Текст
                    NATIONAL DEFENSE RESEARCH COMMITTEE
REPORT NO. A-9O : PROGRESS REPORT
THE EROSION OF GUNS
Part One: Fundamentals of Ordnance
Relating to Gun Erosion
by
John S, Burlew
Approved on September G, 19h2
for submission to the Section Chairman
Approved on September 8, 1?42
for submission to the Division Chairman
Approved on September 9, 19h2
for submission to the Committee
hn S. Burlew, Consultant
Section A, Division A
L. H. Adams, Chaiman
Section A, Division A
z6
Richard 0, Tolman, Chairman
Division A

Preface The work described in this report is pertinent to the pro- ject designated by the War Department Liaison Officer as OD-52, to the project designated by the Navy Department Liaison Officer as NO-23 and to Division A projects PA-230 and РА-2ЦО. Part of the work of preparation of this report was perform- ed under Contract OEMsr-51 with the Carnegie Institution of Washington and elsewhere under the auspices of Section A. Through arrangements made in July, 19h1 with Brigadier General R. H, Somers, then Ordnance Department Liaison Officer with the National Defense Research Committee, the writer was given access to the confidential reports of the Ordnance Depart- ment dealing with gun erosion, for the purpose of preparing this report. The information gained during several months of work in the Ordnance Department Technical Library was supplemented by discussions with both officers and civilian personnel of the War and Navy Departments. This report has been issued in two volumes. The present volume comprises Part One, on the fundamentals of ordnance relat- ing to gun erosion. The second volume will comprise Part Two, on the characteristics of gun erosion, A bibliography and an author index appears in each volume, but the subject index appears only at the end of the second volume. Acknowledgements. — The writer is grateful for the friendly cooperation of the many persons who assisted him in gathering in- formation. He is especially indebted to Mr. S, Faitman, Principal Ordnance Engineer of the Ballistic Section of the Technical Divi- sion, Industrial Service, Ordnance Department, who was a friendly mentor in matters of ordnance, to Mrs, S. H. Pendleton and her assistants in the Ordnance Department Technical Library, and to Mrs. R. E. Gibson, who helped collect data and who prepared the manuscript. Distribution of copies of the report. — The editorial staff of Division A completed preparation of this report for duplication on September 15, 19^2. The initial distribution of copies was as follows: Copies No. 1 to 2li, inclusive, to the Office of the Secretary of the Committee for distribution in the usual manner; Copies No, 25 to 33, inclusive to the Chief of Ordnance, War Department (Attention: Col. W. R. Gcarhardt# Col, S. B. Ritchie, Col. R. R. Studler (2 copies), Lt, Col. D, J. Martin, Capt. С. E. Balleisen, S< Feltman, P. M. Netzer, Technical Library); iii
Copies No. Jli to 36, inclusive, to Aberdeen Proving Ground (Attention; Lt. Col. L. E, Simon, R. H. Kent, J. R. Lane); Copies No, 37 to hO, inclusive, to Watertovm. Arsenal (Atten- tion: Col. H. H. Zornig (2 copies), P. R. Kosting, H. H. Lester); Copy No. 111 to Watervliet Arsenal (Attention: Lt. Col. 3» L. Conner); , Copy No. h-2 to Frankford Arsenal (Attention: Lt. Col. L. S. Fletcher); Copy No. 113 to Springfield Armory (Attention: Capt. A. E* Benson); Copies No. hb to I16, inclusive to the Chief of the Bureau of Ordnance, Navy Department (Attention: Capt. G. L. Schuyler, Lt. B. D. Hills, С. B. Green); Copy No. 117 to the Office of the Coordinator of Research, Navy Department (Attention: Lt. T. C. Wilson); Copy No. L18 to the Naval Research Laboratory (Attention: R. Gunn); Copy No. to the Inspector of ordnance in charge, Naval Proving Ground, Dahlgren, Va. Copy No. 50 to С. C. Lauritsen, Vice Chairman, Division A; Copies No. S1 to 60, inclusive, to L. H, Adams, Chairman Section A, Division A; Copy No. 61 to H. B. Allen, Vice Chairman, Section A, Division A; Copy No. 62 to L. J. Briggs, Member, Section A, Division A; Copies No. 63 and 6Lt to E. L. Rose, Member, Section A, Division A; Copy No. 65 to E. R. tfeidlein, Member, Section A, Division A; Copy No. 66 to J. S. Burlaw, Consultant, Section A, Division A; Copy No. 67 to H. L. Curtis, Consultant, Section A, Division A; Copy No. 6S to R. E. Gibson, Consultant, Section A, Division A; Copy No. 69 to W, F. Jackson, Consultant, Section A, Division A; iv
Copy No. ?0 to J. F. Kincaid, Consultant^ Section A, Division A$ Copy No. 71 to T. C. Poultor, Consultant, Section A, Division Aj Copy No. 72 to J. Wulff, Consultant, Section A, Division A; Copy No, 73 to J. 0. Hirschfelder, Consultant, Section A, Division A; Copy No. to P. Bridgman, Consultant, Section B, Division A; Copy No. 75 to G. B. Kistiakowsky, Chairman, Section B-1, Division B; Copies No. 76 to 80, inclusive, to the Secretary of the Committee for transmittal to L. C. Pauling, Clyde Williams, J. C. W. Frazer, I. M. Stein, J, P,' ilagos. v
CONTENTS Pj'JIT ONE FUNDAMENTALS OF ORDNANCE RELATING TO GUN EROSION Chapter Page Foreword by L. H. Adams .......................... ix I. Introduction ....................................... 1 II. Guns ............................................... 7 III. Projectiles ....................................... 26 TV . Propellants ....................................... 33 . V. Burning of the Powder ........................... Ц2 VI, Movement of the Projectile in the Bore .......♦ 69 VII. Reaction of the Tube to Firing .................... 79 Bibliography and Author Index for Part One .. 89 PART TWO* THE CHARACTERISTICS OF GUN EROSION VIII» Magnitude of Erosion IX. The Eroded Surface X. Experimental Methods of Study of Gun Erosion XI. Relation Between Properties of the Gun Metal and Erosion XII. Conditions of Firing that Influence the Rate of Erosion XIII. Effect on Erosion of Leakage of Gas Past the Projectile XTV. Life of Guns XV. Theories of the Mechanism of Erosion Bibliography and author Index for Part Two Subject Index to Parts One and Two Part Two appears in a second volume designated as NDRC Report A-91. vii
foreword By L. H., Adams Shortly after Section A of Division д was constituted for the purpose of attacking the problem “of gun erosion and other aspects of hyper-velocity, and the facilities of the Geophysical Labora- tory of the Carnegie Institution of Washington were made avail- able for this work, Dr. John S. Burlew of the scientific staff of that laboratory was requested to prepare a report on gun erosion. It was realized that there existed a vast amount of factual mate- rial in this field and a large number of conflicting interpreta- tions and conclusions relating to the observations. The intention was to have an outline of the subject that would be comprehensive but not necessarily exhaustive, and to include merely a cross- section of existing data together with an explanation of the terms used and a summary of the conclusions that have been drawn. Col- lecting and assembling the necessary information was a task of considerable magnitude, which would have been even greater but for the hearty cooperation and encouragement of various individuals in the Army and Navy and in other organizations. The purpose of the report has been to provide the necessary ground work for planning investigations on the mechanism of gun erosion. A considerable portion of it has been available for some time to me and to other persons working on this problem. The pro- gram was first presented and discussed at the Conference on Gun Erosion held at Watertown ^rsenal on October 1?, 19Ш, at which ix
representatives of the National Defense Research Committee as well as of various establishments of the Army and Navy were present» Since then the investigation of the mechanism of gun erosion has been going forward actively at the Geophysical Laboratory and else- where under the auspices of Section A. The present report, which in its completed form provides a convenient background for these investigations, does not include any of the results already obtain- ed in them. Those results arc being presented in a scries of Division A memorandums and reports, the first few of which have already been issued or are in preparation, as mentioned in foot- notes to the appropriate parts of the text of the present report. Washington, D. C. August 11, 1?h2.
THE EROSION OF GUNS Part One: Fundamentals of Ordnance Relating to Gun Erosion’ CHAPTER I. INTRODUCTION 1. The present survey of the erosion of guns aims to collate the principal factual data on the subject and, in addition, to summarize the different theories that have been proposed to explain or to cor- relate these facts. Special care has been exercised to separate the facts from the various interpretations that may be placed upon them, since the latter in many instances depend on unverified assumptions. The sources of all information have been indicated in order that the reader may consult them for further details; for, although this sur- vey attempts to be comprehensive, the treatment of the various items mentioned is not intended to be exhaustive. 2. Definitions A gun consists essentially of a hollow metal tube having' one end — the breech — tightly closed and having within it a projectile that is expelled with a high velocity from the open muzzle by the rapid expansion of the gases which are formed by the burning of the propellent powder placed in the chamber behind the projectile before firing. Erosion of a gun is the gradual removal of metal from the surface of the bore of the tube as a result of firing. It has been suggested that it is desirable to distinguish between the ’’erosion” caused by the hot powder gases and the mechanical "wear” caused by
- 2 - the movement of the projectile; but there is not yet enough evidence 1/ to permit this distinction.-7 Hence in the present discussion the inclusive term erosion will be used to refer to enlargement of the bore without implication as to the cause. This enlargement of the bore radically affects the flight of the projectile. Erosion at the breech end of the tube decreases the range of the gun, and erosion at both ends decreases its accuracy. The two effects combine to determine the useful life of the gun. Some of the methods of estimating the life of guns from the extent of erosion are treated in Chap. XIV. 3. High muzzle velocity Because the rate of erosion increases very rapidly with an in- crease of muzzle velocity, as will be illustrated in detail later (Sec.10$), a muzzle velocity of about 3000 ft/sec is the upper limit that may be employed at present without decreasing the life of a gun too much.' Actually, even most high-powered guns are rated at some- what lower muzzle velocities. V Dr. P.R. Kosting, of Watertown Arsenal, in commenting on this statement pointed out (private communication, Aug. 22, 1?42) that ’’the phenomena accompanying the change in bore diameter at the muzzle end are different from those at the breech end,11 and in parti- cular, that ’’metallographic studies reveal that there is a dif- ference in the effect of service on the bore metal between the muzzle and breech end. In discussions of erosion one hears frequently of references to the wear at the muzzle end and the erosion at the breech end.’1 The distinction that Kosting makes with respect to the phenomena is probably valid; but until we are certain that this dif- ference is anything more than that caused by modification of the same agent by two different sets of local conditions, it seems un- necessary to introduce an extra term in referring to it. Conversa- tions with Navy personnel have shown that they still use the term "muzzle erosion.”
- 3 - The depth of penetration of amor plate by a projectile in- creases as a function of the striking velocity of the projectile. At relatively low velocities — up to about 2£00 ft/sec — the depth of penetration of a projectile is proportional to approximately the three-halves power of its striking velocity as is indicated by such 2/ empirical formulas as that of de Marre [1886]-7 which is still in use today [Naval Ordnance, 1939]- Whether this same relation holds true at velocities exceeding 2J?00 ft/sec is not known with certainty. It has been claimed that the phenomenon of penetration is different at hypervelocities — greater than, say, 3!?00 ft/sec — and that at such velocities projectiles made of lead or other soft metal will penetrate armor as well as steel projectiles. This opinion, which has not been substantiated as yet, was extravagantly insisted upon by Gerlich [1930, 1931, 19331> and it has been repeated even by Hayes [1938, p. 670]. Be that as it may, a gun firing at a muzzle velocity of nearly $000 ft/sec would be an especially effective weapon against the armor carried by tanks, because in this way it would be possible to deliver a projectile having a striking velocity of over 3000 ft/sec at battle ranges up to 1000 yd, and also because the decreased time of flight and flat trajectory would reduce the un- certainty in the estimation of range and position, The probability of hitting an airplane is increased by de- creasing the time of flight of the projectile, which may be accomplished 2/ The name of an author followed by an underscored date refers to the appended bibliography.
by increase of muzzle velocityThe experience of the present war led the Director of Artillery of Great Britain, for instance, to suggest that there is need for an antiaircraft gun capable of sending a projectile to a height of 10,000 ft in only 3 sec, which would require a muzzle velocity of between £000 and 6000 ft/sec [O.B.P., 19М]. "" I The attainment of such high muzzle velocities is prevented, not by the impossibility of designing a gun to produce them, but simply by the realization that such a gun of conventional design would erode so rapidly that its use would not be economically fea- sible. Therefore, if a means of greatly retarding erosion could be discovered, it would not only increase the life of guns in general, but also it would permit the development of these special weapons having high muzzle velocities without bringing in the complication of novelty of design. h. Outline of the present survey A discussion of some of the elementary features of the con- struction and firing of guns has been included in this report, so that the main discussion may be understood by a reader who has had no previous acquaintance with ordnance. This is especially necessary 3/ In a recent discussion of this subject, Weaver [19U2] con- cluded that "in general the probability of hitting varies with at least the inverse second power of the. time of flight; and with an inverse power of the time of flight which may be 3, 4, 5 or even 6 in cases progressively complicated by more severe dodging." Earlier Kent [1933] had calculated that the probability of hitting a maneuvering airplane varies inversely as the fourth power of the time of flight.
- 5 - because this report is primarily intended for the use of civilian scientists. Although these scientists are technically well quali- fied to investigate the problem of erosion of guns, ordinarily they have had no more opportunity to become acquainted with the details of ordnance than most ordnance officers have had to learn as much as they would like to know about the pure sciences underlying the applied science that is a part of ordnance. Those readers who are familiar with ordnance, therefore, may well omit Chapters II, III and IV and a few sections in some of the later chapters* The design of some parts of a gun affects the rate of erosion, and therefore a description is given in the present volume of certain features of gun tubes, projectiles, propellants and the conditions of firing, so that the necessary information will be at hand for later application in a discussion of the relation of these factors to erosion. Other parts of a gun — the breech mechanism, the firing mechanism, the recoil system, the carriage or other mounting, the aiming and laying devices, and the fire control instruments — do not affect the progress of erosion and therefore they are not dis- » cussed. In the second volume — Chaps. VIII to XV — erosion is con- sidered as a result of the interaction of the gun tube with its en- vironment during the few thousandths of a second that elapse during firing. The eroded surface is described in detail and then the effect of varying the different factors of the environment are set forth to the extent that information concerning them is available. The experi- mental methods of study of gun erosion are described; formulas for
- 6 - the life of guns are summarizedj and finally, some of the theories that have been suggested to explain the erosion of guns are dis- cussed.
- 7 - h./ CHAPTER IIA GUNS %. Classification Guns are classified first, with respect to purpose, and second, with respect to size. To a certain extent the two systems are not mutually exclusive, for the use of the weapon often demands a cer- tain size in order that it should be effective. The purpose of the gun is related to erosion only in so far as it may be reflected in the design or in the conditions of firing — for instance, the large projectile and low mutzle velocity of a howitzer or the rapid fire of an antiaircraft gun. The size of a gun is measured in terms of the diameter and the length of the bore. The diameter, which is called the caliber of the gun, is the principal dimension. The length of the gun from the muzzle to the face of the breech block is usually expressed in terms of calibers, that is, it is expressed as the ratio of the actual length to the diameter. Table I shows the dimensions of a representative series of guns. Guns may be broadly classified on the basis of caliber into small arms and cannon (or artillery). A .£o-caliber machine gun is the largest size, standard small arm, and a 20-mm gun is the smallest size cannon in the U.S. Army. The old classification of artillery into guns, howitzers and mortars has lost some of its h/ Most of the information contained in this chapter, for which no other source is indicated, has been taken from Hayes [193.8] or from Naval Ordnance [1939].
Table I. Principal dimensions and other d,ta.lls of representative guns and their proj,ctiles. (Tak,n by parmission from offical drawings in the Ordnance Department, U. S, Army.1) Oun* .30-MG Ml 7A1 .ui-ua М2 20-ram Ю ,M2 37-nn Ю 75-mtajJ М2-М3 3-in. AA М3 90-ram AA M1 105-™ AA М3 h.7-in. AA T2 155-sbb Ml 8-in. Mk VI U-ln. M1920M2 t6-in. M1919M3 Construction [b] Monobloc Built-up Monobloc Monobloc Monobloc Built-up Built-up Built-up Monobloc Monobloc Uonobloc Monobloc sone CW CW liner cw cw cw 2 hoops linsr liner Wire-wound Muzzle velocity(ft/sec) Pressure} desired max. (Ib/in?) 2800 2800 2850 2600 1850-2050 2600 2700 2800 3150 2800 2750 2650 2700 118,000 he,000 112,000 36,000 36,000 3h,000 38,000 36,000 38,000 36,000 38,000 38,000 38,000 Tube, length (calibers) — diameter at breech: 72 72 85.5 53.5 28.5-37.5 50 50 60 60 h5 h5 50 50 inside (in.) O.h7 0.80 0.98 1.98 3.21 3.90 h.71 5.81 7.o6 7.1)0 11.00 16.5 22.0 outside (in.) [c] 1.10 1.68 2.05 h.75 6.60 6.12 10.60 15.50 i5.o ih.6 12.2 19.5 26.2 Bore disaster,lands (in») 0,300 O.pOO 0.787 1-1*57 2.950 3.000. 3.5b3 h.13b h.700 6.100 8.000 11). 000 16.000 — groves (in.) 0.308 0.510 0.81? 1.1)97 2.990 3.080 3.623 h.22h h.79O 6.200 8.11)0 1h.2h0 16.21)0 Depth of rifling (in.) o.ool) 0.005 0.015 0.020 0.020 o.oho 0.01)0 0.0h5 o.oh5 0.050 0.070 0.120 0,120 Travel of projsettle(inJ 21.75 33.25 63.68 69-95 72.5-99.2 127.73 156.1) 221.3 2h0.l) 238.3 298.9 60h.7 679.5 Length of rifling (in.) 21.25 31.75 63.08 68. h5 69.6-96.2 125.83 152.1) 216.7 2hi.6 230.6 288.8 598.3 667.5 Twist of rifling 1/33.3 1/30 1/25.586 1/25 1/25.59 i/ho 1/32 1/30 1/30 1/25 1/25 1/32 1/30 Number of grooves ll 8 9 12 21* 28 32 36 h2 h8 61) 136 llih Width of grooves(in.) [d] 0.167 0.13 0.205 0.231 0.21)8 0.187 0.198 0.211 0.202 O.2h9 0.236 0.173 0,209 Width of lands (in.) 0.06 0.06 0.068 0.15 0.139 0.15 0.1S 0.15 0.15 0,15 0.157 o.i5- o.ih Boeder chamber, length (in.) 2.75 h.25 h.2h 9.20 13.89 2h. 17 2h.25 31.1)7 hO. hO 1)6.80 71.21 102.66 157.3 —* net volume(in*)[•] 0.25 0.98 2.22 19.92 80.59 200 300 600 10h6 1596 5156 19,377 39,960 Charge (lb) 5o gr 250 gr 0.07 0.50 2.00 h.875 6.82 11 2h 30 98.5 h60 832 Density of loading 0.7? 1.01 0.87 0.695 0.686 0.675 0.629 0.508 0.66 0.63 0.52 0.66 0.59 Projectile type [f] Ball 112 Ball Ml H E Mkl АР U51B1 HEHh8[g] И E Mh2A1 H E U71 И E M38A1 Я E «73 И E M1O1 АРМкХХМЗАРИс VI АР Mk IX Fuse None None Mkl/A/ None Point det. 1Д9 Mh3 A2 Mh3 A2 lih3 A2 «61 Point det. Base det, M51 Mk X Mk X Mk X Projectile, — weight (lb) 150 gr 708 gr 0.286 1.92 11). 60 12.81 23.h 32.77 50 9h. 86 261.8 1560 23h0 — length (ln.)[h] 1.12 2.38 3.25 6.36 15.00 10.19 I6.3h 17.96 23-50 27- Sh 38.18 62 [i] 7211) — body diameter (in.) 0.3095 0.5115 0,770 I.h53 2.928 2.980 3.520 h.10 h.67 6.07 7.95 13.93 15.93 Rotating band, diam. (in.) [J] — — 0.828 1.507 3.011 3.098 3.6h3 h.2hh h.82 6.180 8.16 1h.17 16.2h — width, total (in.) — — 0.197 0.7h 0.1)9 1.02 1.20 1.1)2 2.25 2.02 3.30 h.65 5.33 — width,cyl. (in.) [k ] — dist.from bese — — 0,170 0.37 0.35 0.87 1.00 1.17 1,65 0.95 2.13 3.66 3.53 (in.) [1] — —— o.ho 0.58 2,22 1.20 2.60 3.02 3.53 3.1)5 2.00 not specified Bourrelet, diameter(in.) — clearance - lands —— —— O.?8h [0] 2.91)5 2.995 3.537 h. 128 h.693 6.092 7.977 13.98h 15.982 (in.) [n] ** 0.003 0.005 0.005 0.006 0.006 0,007 0.008 0.023 0.016 0.018 — width (in.) — 0.25 1.h80 0.5 0.75 o.85 0.78 1,1 1.62 2.33 2.67 — dist.from base(in,)[m] — — 1.92 — 6.27 5.30 7.23 7.98 8.80 7.h8 10.78 not specified » See pp.10 ff for notes concerning the purposes for which these guns are designed.
Footnotes for Table I [a] The М2 and М3 guns differ in length <нйу> and hence in mussle velocity* [b] CW: ’♦cold-worked.'1 The liner of the 3-in. gun is loose. Those of the 155-mm, 8-in. and 1 Ц-in. guns are shrunk in place. The smaller guns, up to and including the 105-srn, use fixed ammunition; the U.7-in, gun is separate loading with the powder packed in bage in a cartridge case; and the 155чпт and larger guns use separate-loading ammunition without a cartridge ease. The machine guns are automatic. The 75-™» 3-in., 9O-m and 105-mn guns are semiautomatic; that is, the force of the explosion ejects the fired cartridge case and leaves the breech so that it closes automatically when another cartridge is inserted by hand. [c] Outside diameter of the tube or liner. Over-all outside diameters of the built-up guns are: 3-in,, 9.00 in.; 155-чвт, 18.8 in.; 8-in., 33. $ in.; 1h-in., Ц8 in.; 16-in., 65.8 in. [d] The width of the cutter fixes the width of the grooves. The width of the lande ie a nominal residual dimension. [e] For small arms ammunition this volume is called the '♦powder space," but on drawings of cannon using bag ammunition the ’’powder space” is the linear dimension from the obturator pad to the base of the projectile. The "volume of powder chamber" of cannon is the net volume between the obturator pad and the projectile of a bag-loaded gun, and the net volume inside the cartridge case of a c&ee-loaded gun, allowing for the volume of the primer, wade and base of projectile. [f] AP: "armor-piercing"; H E: "high explosive." The projectile listed for each gun is the standard one now being manufactured, although older models on hand are still issued for service. Frequently a different projectile may be fired from the earns gun; or the earns projectile may be fired at reduced charge and hence with lower velocity. [g] An AP projectile (M5l) is used with almost equal frequency. It differs from the H E type in the following dimensions: weight, 1Ц.Ц0 lb; length, 8,91 in.; rotating band, O.?6 in. from base; bourrelet, O.!>1 in. wide and 5.0o in. from base. A semi-armor-piercing projectile (M7U) that is also frequently fired differs from the AP type in not having a cap and windshield. [h] Length includes fuze, if any, and windshields on AP projectiles. [i] This dimension is not specified. The one given is an approximate one taken from a scale drawing. [j] The rotating bands on the larger projectiles have a "hump" that projects beyond the main cylindrical portion. The diameters of these "humps" for the several projectiles are: b.7-in., Ц.98 in.; 155-ran, 6.320 Ln.; 8-in., 8.36 1л.; 1 Ц-in., 1Ц.Ц0 in.; 16-in., 16.58 in. The top of ths hump is only 0.0$ to 0.1 in. wide. [k] All rotating bands are beveled at the front end, and those on ths larger projectiles also at the rear. The cylindrical portion is cut by from one to five circumferential grooves known as cannelures, which provide space for the copper to flow into during the engraving. [1] The distance to the rear of the rotating band or bourrelet. [m] This projectile, which Is of the so-called shot type, has no bourrelet. Ths clearance between the body and the lands is only О.ООЦ in. on the diameter. (nJ This clearance is the minimum. The tolerance on the diameter of the lands is *0.002 in. and that on the diameter of the bourrelet is -0.005 in., so that the tolerance on the clearance is *0.007 in.
- ю - Notes on the purposes of the Guns Listed in Table I The ,30-caliber M1917A1 Browning machine gun is the standard ground machine gun, and is also used as an auxiliary antiaircraft weapon. It is a.recoil-operated and belt-fed automatic gun fir- ing at the rate of hOO to $2$ shots/min from a tripod mount. It is water-cooled. The .50-caliber М2 Browning machine gun is made up in three types — water-cooled, heavy barrel-(air-cooled) and aircraft (air-cooled) — by assembly of the proper barrel and jacket on a basic receiver. The water-cooled gun is the principal antiair- craft machine gun of antiaircraft regiments. It fires at the rate of. 600 rounds/min from a tripod mount. The 20-mm. guns Щ and М2 are automatic aircraft weapons mounted either for firing through the hub of the propeller or as a fixed gun in the wing. The rate of fire is 600-700 rounds/min. The only difference between the M1 and М2 guns, tins former of which is a substitute standard, is in the dimensions and shapes of some' of the receiver parts. The 37"ПДП gun М3 is the smallest field gun used by the in- fantry and cavalry in the U, S. Army, It is intended especially for antitank service. Its carriage Mh, which is designed for great mobility over any kind of ground, can be controlled by one man. The gun is manually operated, Its tube can be easily un- screwed from the breech ring for replacement when it becomes over- heated or worn out. This gun fires both armor-piercing (1,92-lb) and high-explosive (1.63-lb) projectiles. The 37~mm gun M1A2, which is the standard antiaircraft weapon of this caliber, is an automatic gun firing a lighter projectile (1.3h lb) than gun at a rate of 120 rounds/min, with a muzzle velocity of 2600 ft/sec. This gun is slightly longer, and the twist of rifling is 1/30. The 7^-mm guns №2 and М3, which are the same except for length, have been designed for mounting on tanks. The former is a’substitute standard, to be used as long as the stock on hand lasts. They are recent (19h1) adaptations of the original 75-mm gun M1897 (French), Ballistically, the пот guns differ .from the M1897 only in length. They use the same ammunition, The longer of these guns is nearly 20 percent lighter than the Ы1897А2 gun because it is of monobloc construction. The 7^-mm gun M1897A2? which is still the standard light field gun, is Ju,5 calibers long and has a muzzle velocity of 1950 ft/sec. The 7^-mra guns, Ml916 (LT. S.) and И1917 (British), which are limited standards'.for this same service, are only 28.h calibers long and have a muzzle velocity of 1900 ft/sec. Their most distinctive feature is that they.employ rifling of increas- ing twist. These light field guns are to be superseded by the 103-mm howitzer М2Д1,
- 11 The 3-in, ДА gun,М3 is now a limited standard, to be super- seded eventually by the 90-mm gun M1of which production was started in 19U0. The loose liners for the 3-in. gun continue to be made. These two guns are mobile antiaircraft weapons. Al- though the 90-mm gun has about ope-third greater maximum range and fires a much heavier projectile at approximately the same rate of fire (2$ rounds/nin), the total weight of the gun and mount is less than 20 percent greater than that of the 3-in.,gun, because improvements in metallurgy, especially in cold-working, have made possible monobloc construction. The 10^-mm AA gun М3 is now (19^2) the standard fixed heavy antiaircraft weapon? The li,7-in, AA gun T2 is an experimental heavy mobile model under development. Just as the substitution of the 90-mra for the 3-in. gun gives increased effectiveness in light antiaircraft guns, so the h..7-in« is designed for use against targets at higher altitudes (^0,000 ft) than can be reached with the 10^-mn gun. This greater range is obtained by the use of greater muzzle velocity and a heavier projectile of radically im- proved ballistic form. The 1^-mm gun Ml Al is the most powerful of the standard heavy mobile artillery weapons. (The others are the l5£-mm, 8-in. and 2h.0-mm howitzers, all of which have lower velocities and shorter ranges.) It differs from the 1^5-mm gun M1 only in the breech ring, but it differs from the old 155-mm gun Ml916 G. P. F. ("Grand Puissance Filloux"), which it superseded in 19^1, by having greater power. The M1916 gun, which was only 32 calibers long, fired a high-explosive projectile of the same weight at a muzzle velocity of only 2Ц10 ft/sec, using a charge of 26 lb. The 8-in, gun Mk, VI, M3A2 and the 1h-in. gun М192ОЫЗ are the principal standard railway artillery. The former gun has been constructed by rechambering one designed by the U, S. Navy. The 16-in. gun Iu1919M3 is a limited standard for the primary armament of harbor defense installations. It is not expected, how- ever, that any more wire-wound guns will be produced. The present standard for new installations is the 16-in. gun Mk. IIM1, which is a built-up gun obtained from the Navy,
- 12 - precision; but these' names still denote differences of length that are implicit in the different purposes of the three types. A gun, in this special sense of the word, is the longest, heaviest type of weapon of a given caliber and has the highest velocity and the longest range. A howitzer is shorter and lighter and has a lower velocity and a shorter range. A mortar is a very short weapon that * is fired at a high angle of elevation with a relatively short range for indirect fire. Th® term gun is also used in a special sense to refer to the tube (or tubes and jackets in a built-up gun) and breech mechanism as a complete unit as distinguished from the mount, which includes the recoil mechanism. 6. The tube A gun must have walls strong enough to withstand the internal pressure of the powder gases. Because this pressure diminishes toward the muzzle, the wall thickness of the tube can be tapered in that direction. Right at the muzzle, however, the wall thickness of a large gun is usually increased somewhat to form the so-called * bell muzzle, in order to make up for the loss of strength occasioned by the ending of the tube. The necessary strength of tube throughout its length is obtained by using a proper thickness of high-grade alloy steol having a high tensile strength. In addition, large guns are made stronger -- or rather, equally strong with thinner walls — by putting the inner layers of the tube under compression. This is accomplished by wrapping the tube with several layers of wire under tension, by shrinking on the tube a jacket and one or more steel hoops, or by expanding the bore beyond the elastic limit of the steel
- 13 - by hydraulic pressure before it has been machined to final dimen- sions. Wire-wound guns, although still in service, are no longer constructed in this country. Built-up guns are frequently made with two slightly tapered coaxial tubes, the inner one of which constitutes a liner that can be replaced after it has worn out. The replacement requires that the tube be sent to an arsenal, for it has to be expanded by heating in order to disengage the liner, which is made with a taper of about 0.003 times the diameter per inch. The use of a loose liner that can be screwed into a small- caliber tube by the gun crew was tried by the U.S. Army in its 3-in. antiaircraft gun, but it was discontinued because of the difficulty of replacement under field conditions. The British Army has used a loose liner with some success in their 3*7- and h..5-in. antiaircraft guns. d/ The process of radial expansion^ — called also autofrettage and cold-working — has been applied to guns up to 8 in. in diameter. A hydraulic pressure of 110,000 to 1^0,000 Ib/in? is used to produce a permanent enlargement of the diameter of the bore amounting to about 6 percent; then, after the pressure is removed, the outer layers contract and place the inner layers of the tube in a state 111 11 11 "" ' ...... .................. 1 111 ...................... 5/ This process, which was originated by the French "auto- frettage” is French for "self-hooping"), has been developed by the Navy at the Naval Gun Factory, Washington Navy Yard, and by the Army at Watertown Arsenal. A number of reports from Watertown Arsenal dealing with the development of the process are on file in the Ordnance Library, cataloged under "Cold-working.” In 19M, equip- ment for cold-working on a production basis was put in operation at Watervliet Arsenal.
- Ill - of permanent compression. A radially expanded monobloc tube instead of a tube -with a replaceable liner is now preferred by the Army for •small guns, because the former weighs less and hence can be used on a lighter mount. Also it has been found-that the cost of an entire monobloc tube is not much moire than that of a liner alone, and much less than that of a liner and tube (Ord. Dept., 19йОа; II, p. 67]. 7. Tapered bore Although the bore of an ordinary gun is cylindrical, some guns having a tapered bore have been developed in order to obtain a higher muzzle velocity without a corresponding increase of erosion. They have been reviewed separately [Burlew, 1?li2] . 8. The rifling An elongated projectile requires a high rate of rotation for stability in flight and for the operation of fuzes of certain types. This rotation is obtained by the rifling cutting into a layer of soft metal on the circumference of the projectile as the latter is propelled down the tube. The rifling consists of a series of helical grooves in the surface of the bore, the depth of which is of the order of magnitude of 1 percent of the diameter. The raised portions of the bore between the grooves are called lands. The twist of the rifling, which is expressed as the number of calibers of length in one' turn, in combination with the muzzle velocity of the projectile, determines the rotational velocity of the projectile. The design of the rifling, in terms of the number, width, depth and form of the lands and grooves and the degree and uniformity of
- 15 - the twist, is a complex subject. Although many designs of rifling have been tried, there is not complete agreement today as to the most suitable one. The currently produced guns of the U.S. Агщу have rifling of uniform twist that varies from 1 turn in 25 calibers to 1 turn in Ц0 calibers. The number of grooves increases with the caliber, so that the widths, of lands and grooves are approximately the same for guns of all sizes. The lands are about 0.15 in. wide and the grooves slightly wider; and the width of each is uniform throughout the length of a given gun (see Table I). 9. Effect of design of rifling on erosion The effect of the rifling on erosion is not very we11.under- stood, The presence of stress concentrations at the sharp corners where the lands and grooves join has been mentioned as an important contributory factor by various writers [Howe, 1918; Justrow, 1923]t but a quantitative evaluation does not seem to have been attempted. The strength of the lands in shear was calculated by Justrow [1923], who considered that the reduction of bearing surface as erosion proceeds tends to increase subsequent erosion. The twist of rifling may affect erosion resistance. An erosion test was made at Aberdeen Proving Ground [Lane, 1933] of a 3-in. antiaircraft gun liner rifled with straight grooves for the first 23i- in. from the origin of rifling (see Sec. 11) and then with a twist that gradually increased to 1 turn in Ц0 calibers during the next 113г in. For the last 7 in. to. the muzzle the twist was uniform — 1 in Ц0. This liner was fired 2950 rounds, for comparison with another having a uniform twist of 1 in Ц0 over its whole length.
- 16 - Both liners had been made of carbon steel of the same chemical com- position and practically the same heat treatment. Both had been radially expanded, the one with increasing twist slightly more than the other. The gun with increasing twist of rifling eroded less than the other, except at the muzzle, throughout the course of the comparison. After 29$O rounds, for instance, at the origin of rifling, the in- crease of diameter of the lands was only 0.182 in. compared with 0.232 in., and of the grooves 0.112 in. compared with 0.166 in. At the same time, at 10 in. from the- origin, the corresponding in- creases were 0.088 and 0.113 in. on the lands and 0.0Ц6, and 0,059 in. on the grooves. The decrease of muzzle velocity after 2000 rounds was only $0 ft/sec for the liner with increasing twist, but it was 1U£ ft/sec for the other one. The deviations in range were the same. 10. The forcing cone^ The rifling is terminated at the end of the bore toward the breech by a tapered section along which the diameter across the lands increases. This so-called forcing cone (called also band slope and compression slope) continues toward the breech a short distance behind the point at which the diameter of the lands equals that of the grooves. Its rear portion forms the seat for the rotating band on the projectile (see Sec, 18), which is machined to a matching cone; and the forward portion, along which the tops of the lands are 6/ For a drawing showing the various parts of a gun tube, see Naval Ordnance (1939), plate 1, Chap IV.
- 17 - tapered, facilitates the engraving of the rotating band when the projectile begins to move. The forcing cone is usually about in. long. At the rear of it; in a gun firing fixed armunition (see Sec. 21) is the centering cylinder, into which the neck of the cartridge case fits. The diameter of this cylinder is 0.2 to 0.3 in. greater than the diameter across the lands. Behind it an- other enlargement of the diameter, called the chamber slope, joins the centering cylinder with the chamber proper. In guns firing separate-loading ammunition (Sec. 21), the forcing cone is fre- quently elongated, and sometimes — for example, in the 1^5-mm gun M1 — it is combined with the chamber slope, which eliminates the centering cylinder. 11. Origin of rifling The origin of rifling is the point along the axis of the bore that corresponds to the place on the forcing cone where the lands and grooves have the same diameter; the length of rifling is measured from this point to the muzzle. In U. S. Amy reports on the extent of erosion, however, "origin of rifling" refers to the forward end of the forcing cone, where the grooves have attained their full depth. In U, S, Navy and also in British reports, the erosion is commonly measured 1 in. in front of the true origin of rifling. 12 • The chamber The chamber is properly that portion of the gun between the centering cylinder and the face of the breech block; frequently,
- 18 - however, the term is used to embrace the whole space up to the 2/ origin of rifling, in which case the total length of the tube is the sum of the length of the rifling and the length of the chamber. ? The ratio of the' diameter of the chamber to that of the bore, which |. is called the chambrage, varies from 1.0$ in some German guns to 1.50 in the new U.?-in. AA gun M1 of the U. S, Amy. Gun designers of the U. S, Navy consider 1.20 a desirable ratio [Naval Ordnance, 1939, Р» 86]. The walls of the chamber in a gun using a cartridge case are tapered, the diameter at the breech end being a few hundredths of an inch greater than at the forward end, in order to facilitate removal of the case. In large guns using bag ammuni- tion, the rear of the chamber is necked down along a conical see-- tion, called the,chamber rear slope, in order to decrease the size of the breech block. 13. Effect of chamber design on erosion The greater the chambrage, the greater is supposed to be the erosion. According to Hugoniot and Sebert [1882] , Charbonnier [1908, 1922] and Tulloch [1921], a bottle-necked chamber causes eddying of the powder gases at the origin of rifling, with the result that local washing action on the wall is set up. The effect of such turbulence was demonstrated by de Bruin and de Pauw [1931] in the case of an erosion vent plug (see Sec. 86) by means of a plug made in three parts. The first and third parts had holes 1 ип in diameter and 10 nn long, whereas the 7/ The term is used in this way in Table I.
- 19 - intermediate section had a hole 7 mn in diameter and 5 mm long. When such a vent plug was fired, using йО-percent NG powder, at a maximum pressure of 1Я00 atmos, the third part of the plug eroded more than the first, even though the. gas had lost sone of its energy. The recommendation made by Hugoniot, Charbonnier and Tulloch that the chamber should not be much greater in diameter than the bore seems plausible, but no quantitative data to support it are known to exist. The reason for this is that ordinarily when two guns of the sane caliber are made ’with chambers of different dia- meters, all other features of the design are not kept constant. The chamber of larger diameter, for instance, will have a larger volume and hence the correspondingly larger charge used in it nay be the sols cause of the increased erosion (see Sec, 10!?). 14. Examination of the bore The surface of the bore of a gun is examined by moans of a boresearcher, which consists of an illuminated mirror and a tele- scope, For bores of diameter 6 in. or less, the two parts are combined in one. instrument, called a boroscope, which may be quickly focused on any part of the bore surfacef For measure- ment of the diameter of the bore, an instrument called a star gage is used. It consists of either two or three measuring points that nay be extended radially in a plane by the movement of the handle at the opposite end of a long staff until the points make contact with the wall of the bore. Thon the diameter nay be read from a scale — either vernier or micrometer — on the handle.
- 20 - All guns are examined with a boresearcher and measured with a star gage at the time of manufacture, 8/ 1$. Steels used for gun tubes and liners The U. S« Navy has used for many years a plain carbon steel for its gun liners and a 3-perccnt nickel steel for tubes, hoops and jackets. Typical compositions of these two steels are given in Table II, The nickel steel differs from the carbon steel only in its nickel content and in having a slightly smaller amount of carbon. Tho yield strength of the gun stoc-1 must be at least $3,000 Ib/in? and that of the nickel steel at least 6$,000 lb/in? (Naval Ordnance, 1936]. Table II. Chemical composition, in percentages, of steels used in U. S. Navy gunsT 0 Mh Si P s Ni SAfe No. Gun steel 0.42-0.50 0.70 0.27 0.03 0.03 0 io4o Nickel steel 0.35-0,42 0.70 0.27 0.03 0.03 3.00 2340 *After Naval Ordnance [1939]. pp. 118-119. The U. S. /.rmy uses a variety of different compositions of low-alloy steels for gun tubes end liners, procured on the basis of a macro-etch test and tensile properties. Ths following quota- tion from Notes on the selection and use of motels in ordnance design [Ord. Dept., 1940b, p. 86], prepared at Watertown Arsenal, explains tho presont policy. 8/ Tho relation between ths properties of the gun metal and erosion is discussed in Chap. XI.
- 21 "dhemical compositions are normally not prescribed because other factors Such as proper melting and finishing of the metal and its final cleanliness play an important part. The method of pouring, size, relative dimensions and shape of molds, proper method of reduction in hot working, and heat treatment in various stages of the process are also important in securing the quality of product desired in gun steels. 'Every manufacturer, depending on his equipment, staff, con- ditions of work, etc., as well as his previous experience and traditions, will have a preference for certain definite composi- tions. It is advisable to let the manufacturer choose his own composition and assume the responsibility for his work so long as the physical and metallographic properties are satisfactory for gun material. "In normal gun steels, the designer is primarily interested in the physical characteristics obtainable. Specification 57-103 provides for the following physical properties: Yield strength Tensile strength Elongation Reduction of area 65,000 lb/in! 95,000 Ib/in? 18.0 percent 30.0 percent The trend, however, is toward the following requirements which have been found more practical for Ordnance purposes: For tubes and liners (medium strength) — Yield strength (set, 0.05 percent) 65-60,000 lb/in? Reduction of area h5.0 percent minimum For 37-mm, 1.1-in., etc. (high strength) — Yield strength (set, 0.05 percent) 95-110,000 lb/in^ Reduction of area lj.5.0 percent minimum The above values are readily obtainable in centrifugal castings, and the manufacturer of forgings can meet such requirements with reasonably clean steel. "The procurement of gun tubes and liners produced by the centrifugal casting precess is covered by specification 57-66. The high ductility from transverse specimens in the cose of centrifugal castings is due to the method of manufacture, which definitely eliminates the directional properties always present in forgings," The compositions of a number of forgings end centrifugal castings used in guns recently completed by the irray are listed in Table III. The tonsile properties of these steels exceeded
- 22 - Table III. Chemical composition, in percentages, of steels currently used in Lrray' guns# Element A в c D C 0.20-0.211 0.31-Q,3H 0.3I1-0.35 0.1x3-0.1111 Mn 0.60-0.90 0.61-0,77 0.71-0.73 0.59-0.67 Si 0.15-0.23 0.20r0.3Q 0.2 0.18-0.26 P 0.006-0.011 0.027-0,035 0.01Ц-0.0211 0.012-0.02 S 0.016-0.018 0.029-0.038 O.O19-O.O23 o.oi5-o.oi8 Mo 0.Ы1-0.53 0.32^0,38 0.30-0.35 0. li 7-0 .h9 V 0.06-0.16 Q,1?*-0r?3 0.00-0.06 0.19-0.21 Cr 0.91-1.08 0.53-0.97 О.19-О.ЗО Ni —— — 2.26-2.116 0.18-0.36 ‘’'Frcrn Record of measurements -and report of inspection, in the files of the ..rtillery Division, Ordnance Department, U. S. Amy. A. Range of compositions of 111 centrifugal castings made in 19110-Ii1 for 37-mm and 90-mri gun tubes and 3-in. Ал gun liners to be either heat treated to strength or cold-worked. B. Range cf compositions of 8 forgings made in 19110-111 for 155-mm gun tubes, to be cold-worked. C. Range of compositions of 3 forgings made in 19110-Ц1 for 75-nm gun tubes, to be heat treated to strength. D. Range of compositions of 3 forgings made in 19UO-111 for 6-in. gun liners, to be heat treated to strength. the requirements just quoted. The composition of the centrifugal castings in column A represents a development at Waterterm ..rsenal, Donald {19371 recommended essentially this same composition ex- cept for carbon content (0,35-0.110 percent) for making a centri- fugal casting to be heat treated to a yield strength of 125,000 lb/in?, ns a result of a tost of 32 experimental castings of various compositions. Bender and Pappas [19111j, after a study of five slightly different compositions, confirmed the choice.
- 23 - Their work showed that the variations in chromium and manganese had only minor effects on physical properties, but "that sub- stantial changes in physical properties may be brought about by changes in carbon content," The higher the carbon content, the higher the yield strength. Bender and Pappas recommended 0.30-0,35 percent carbon for a yield strength of 121,000 to 125,000 Ib/inf For rifles and machine guns the formerly used manganese steel, W. D, (S.h.E.) 1350, has been superseded by a chromo molyb- denum steal, W. В. (S.h.E.) 4140 modified, which han been found to increase the service life. The compositions of these two steels are given in Table TV. For bans up to 2 in. in diameter the minimum yield strength specified is 110,000 lb/in®, the minimum elongation is 18 percent and the minimum reduction of 2/ area is 50 percent. Those tensile properties represent a consider- able improvement over those of the 23 alloy steels that had been tested (see Sec, 92) in the form of machine-gun barrels at Springfield ..mory in the early nineteen-twenties, only one of which had contained any molybdenum. Table IV,' Chemical composition, in percentages, of stools used in small arms barrels. Element W.D. 1350 W.D, 50 Nod. C 0.45-0.55 0.45-0.55 bin 0.60-0.90 0.60-0.90 Si — —— O.15 max. P o.o45 0.025 max. S o.o55 0.025 max. Or — 0.80-1.10 No •—•• 0.15-0.25 9/ U.S. .xrmy Specification No. 57-107-25, Oct., 11, 1939.
- 2k - 16. Yield strength The yield strength referred to in the foregoing section is determined by the ’’offset method” as described in ’’Federal speci- 10/ fications for metals,” In this method the extensometer reading is plotted as a function of load — by on autographic extenso- meter in many cases — until the curve is no longer linear. Then a straight line is drawn parallel to the initial straight portion of the extensometer curve and offset from that curve by an amount equal to a prescribed set — for example, 0.0^ or 0.1 percent of the gage length. The load corresponding to the point of inter- section of this line and the extensometer curve, divided by the original cross-sectional area of the specimen, is the yield strength. The value of the set now prescribed — usually 0.0f> percent — was chosen so that the resulting yield strength corresponds closely tc that determined by the proportional method, which formerly was used for inspection. For a given sample, the yiold strength is numerically Ц/ larger than the proportional limit, the elastic limit or the 12/ proof stress. As defined in the same specification, these three 10/ Federal Specification [ч^-М-151а, Nov, 27, 1936, par. 22g(2jj; published in the Federal standard stock catalog, Sec. IV (part 5), 11/ The yield-strength determined by the proportional method, which should not be confused with the proportional limit, is de- fined in Federal Specification QQ-M-15>1, par. 22g(l) as follows: 11 Proportional method. — The yield strength shall be calculated at the reading last preceding the first increment of load which pro- duces an increment of strain which clearly exceeds twice the increment of strain taken from the modulus line.” 12/ Footnote 11, pars. 37a, b and c.'
- 25 - limiting stresses correspond, respectively, to 0, 0.003 and 0.01 percent permanent elongation. The yield point, on the other hand, J3/ corresponds to an extension under load of 0.5 percent or greater; and the tensile strength is the greatest load per square inch of original cross-sectional area carried by the material during a tension test. 13/ Footnote 11, pars. 22h and 37o
- 26 - CHAPTER III. PROJECTILES 17. General description of projectiles Modern projectiles are' elongated to secure a largo ratio of mass to cross-sectional area, and.thereby long range. The head of the projectile is provided with a curved surface in order to / decrease the wind resistance 5 and, similarly, the rear end is frequently tapered slightly to give a boat-tailed shape, which reduces air resistance at velocities in the region of the velocity of sound. The longitudinal cross suction of the head is bounded by an ogive, which is an arch-shaped curve, the two branches of which are arcs of two large circles of the sane radius intersecting at the nose of the projectile. The longi- tudinal axis of the projectile' is the perpendicular bisector of the line joining the centers of these two circles. The radius of the ogive, ’which is the radius of the two circular arcs, is expressed in-calibers. Small urns projectiles, which arc called bullets t have a soft outside jacket of copper alloy of diameter large enough to be cut by the grooves of the rifling. the rear of a bullet is a circumferential groove, called a cannelure, into which the mouth of the cartridge case is crimpud, Лп artillery projectile is so long that, in order to keep it centered in the bore, it is provided with a bourrelet at the rear of the head. This is a cylindrical surface that has boon machined to have a slight clearance in the bore (see Table I). Behind the bourrelet, which is about ono-sixth caliber wide, the remainder of the body has a diameter slightly less then that of
- 27 - the bourrelet, ec that it doos not touch the lends. Near the rear of the projectile a rotating band (Sec, 18) is shrunk and forced into an undercut groove machined in the body of the pro- jectile. The dimensions of the cavity of the projectile and the location of the center of gravity are features of design that are important in general but which do not relate to erosion. The general proportions of artillery projectiles are ex- pressed in approximate terras in Table V. Table V. approximate proportions, in calibers, of artillery projectiles# Length Ц to 6 Length of head 2-1/3 to 3 Radius cf ogive 7 to 9 Width of bourrelet 1/6 Width of rotating bond 1/3 Length of cylindrical part of base 1Д ^nglo of boat-tail 9е to 9е Weight (lb) (caliber in inches)3 *.fter Hayes [1938], p. £60. 18. Rotating band The rotating bond — called the driving band by the British — performs taro functions in addition to furnishing a convenient sur- face for the rifling to engage and thus to spin the projectile. It centers the rear of the projectile and it seals the forward end of the powder chamber. Erosion of the gun adversely affects the
- 28 - performance of all "three of these functions of the rotating band. The lack of centering of the rear of the projectile in the bore of a worn gun has been suggested [Miller,. 1920] as a major cause of the increase of dispersion. The increased opportunity for leakage of gas past the rotating band (see Chap. XIII) in a worn case gun (defined in Sec. 21) may account for the increased rate of erosion generally ascribed to such guns as compared to bag guns (Sec. 21), in which tho projectile is rammed hone regardless of how much the forcing cone has been advanced (see Sec. 63). The band must be both soft and strong to serve its several purposes. Copper, which was formerly used exclusively, is still the standard material for the rotating bands of most projectiles used in Navy guns and of those of larger calibers of seacoast artillery. Gilding metal — 90 percent Cu, 10 percent Zn — is widely used for rotating bends by the urmy in order to reduce coppering (see Sec. 60); but although it was triad by the Navy, it is now used by that service only for projectiles for the 1Ц/ 1,1-in. and 6-in. guns. The diameter of the rotating band, as may be seen from Table I, is about 0.02 in. greater than that of the grooves in a new gun, so that the grooves are completely filled when the rotating bund is engraved as the projectile moves down the boro. The starting pressure — or shot start pressure, in British usage — which is required to engrave ths rotating band, is considered as part of the friction of the projectile in tho bore (Sec. ЦЦ). A s1 ill/ Dr. L. T. Е» Thompson, Naval Proving Ground, personal communication, Sept. 13, 19hl.
-29 - recant British investigation. [л, C. 19h-2c] has shown that engrav- ing of the rotating band of a 3-in. proof shot occurred partly by plastic flow ord partly by shear. The amount of shear increasod ’with the age of the gun after it had been two-fifths worn. It also increased with the degree of filling of the grooves and with the hardness of the copper. Many experiments have been fnade with rotating bands for the purpose of reducing "fringing.11 Depending upon the design of the band, excess copper may be dragged back during engraving by the rifling to form n fringe at the rear edge of the band, and then, as the projectile leaves the muzzle, this fringe is flared out by the powder gases and by centrifugal force into a shape like a hoop skirt, which seriously affects the flight of the projec- tile . A resume of the modifications of rotating bands tried dur- ing World VTar I to overcome fringing was given by Veblen and Alger [1919]• Fringing is eliminated by providing some place for the displaced copper to flow into, as, for instance, by cutting a cannelure at the roar of the band. 19. Effects of the rotating band on erosion The rotating band may affect erosion in three different ways. The frictional wear between it and the surface of the bore (Sec. hh) is generally considered important, although there is no means of evaluating its contribution to the total erosion. Some investigators — for instance, Justrow [192-3] — consider that the frictional wear per se is insignificant, whereas others for instance, Kosting [1939b] — conceive of it simply
- 30 - as ths means of ramoval of portions of the bore surface that had already been loosened by previous chemical or physical action. Tulloch [1921] suggested that the frictional wear is increased at tho muzzle by the abrasive action of particles of steel carried forward from the rear of the bore. At any rate, some evidence that wide rotating bands caused more erosion than narrow bands was obtained [Lane, 1933] in the firing of 3-in. antiaircraft liners (see Sec. 9). In the second place, the design of the rotat- ing band helps to determine the extent to which gas can leak past the projectile, which is an effect that is considered in detail in Chap. XIII. In the third place, the material of which the rotat- ing band is made may add another factor to these two mechanical ones by reason of metal fouling, which is considered briefly in Sec. 68. 20, Cartridge case Ammunition is of two general types, depending on whether the powder and primer are, or are not, contained in a cartridge case. The cartridge case, which is usually made of drawn brass, also serves as an obturator to prevent the escape of powder gases through the breech mechanism. The thin body of tho case at the forward end is expanded by the gas pressure until it makes a gas- tight seal with tho wall of the chamber. (Obturation of the breech of a gun that doos not use a cartridge case is obtained by means of an asbestos mushroom pad or other gas check device which is part of the breech mechanism.) The case and the cham- . her into 'which it fits are tapered slightly, in order to facili- tate removal of the empty case after firing. The primer is held
- 31- in a recess in the bottom of the case, so that it is struck by the firing pin that operates through the breech mechanism. A cartridge case is only slightly affected by firing, and there- fore can be used a number of times, if it is resized to the proper dimensions. 21. Fixed and separate-loading ammunition A complete round of fixed ammunition consists of the pro- jectile and fuze (if any) firmly crimped in the end of a cartridge case that contains the powder and primer. Such ammunition is used for small arms and for most guns up to and including the 105-mm gun in the i.rmy and the 5-in. 25-caliber AA gun in the Navy. It is essential for all automatic and semiautomatic guns. Semifixed ammunition, in the nomenclature of the U. S. Army, differs from fixed ammunition only in that the projectile is loose in the end of the cartridge case and the powder is in bags. This type of ammunition is employed for guns, such as the 105-mm howitzer, that use charges of different sizes in order to vary the range. The cartridge case is loaded originally for the maxi- mum range, and then, if a shorter range is desired, ope or more bags of powder arc removed. The projectile and cartridge case are loaded into the gun as one unit. In the U. S. Navy the term semifixed ammunition has a slightly different connotation. It is a-form of separate-loading ammunition in which the powdor charge is packed in a cartridge case for convenience in handl- ing, but in which the projectile and the cartridge case are loaded separately into the gun. This manner of loading gives
- 32 - opportunity to rani hone the projectile against the forcing cone. The 5-in. 38-caliber AA gun and the latest 6-in. guns of the Navy use this form of semifixed ammunition. It is also used.in the nevr Army b.?-in. gun M1 without being designated by any special name. Bag ammunitiont which is the general form of separate- loading ammunition, consists of a projectile, one or more bags of powder and a primer., Tho powder bags are made of silk cloth, and an ignition charge is contained in tho base of each. The primer is attached to the end of the mushroom stem. Bag ammuni- tion is used for all large guns, which sometimes are called bag guns as distinguished from case guns.
- 33 - CHAPTER IV. PROPELLANTS 22. Classification of explosives Propellent powders are always low, or progressive, explosives, characterized by propagation of the explosion — really a very rapid burning — only on exposed surfaces, by the heating of succes sive layers to the ignition temperature (see Chap. V). In high explosives, on the contrary, a detonating wave is transmitted prac- tically instantaneously throughout the mass after initiation of the l£/ explosion by heat or shock. Both nitroglycerin and trinitrotol- uene (TNT) are high explosives, but when either is incorporated in- to a colloid with nitrocellulose, as is done in the manufacture of some propellent powders (see Table VI), it loses its tendency to detonate and merely increases the rate of burning of the mixture. A Iovj- explosive, on the other hand, may acquire the power to det- onate under' special circumstances, such as extreme shock. 2.3. Granulation After General Rodmen of the U. S. Army discovered the principle of the progressive burning of gunpowder in i860 and applied this principle by compressing gunpowder into perforated grains of different sizes, the tern powder as applied to explo- sives lost its etymological connotation of a pulverized material. Powders today are produced in the form of cylinders, cords, tubes, flakes and strips. Most propellent powders in the United States are made in the form of cylindrical grains, the diameters of 13/ A recent discussion of the theory of detonation has been given by Carl [1рЦо].
- 311 - which vary from 0.03 in. for a .30-caliber rifle to 0.9 in. for a 16-in, gun» . The lengths vary correspondingly from 0..06 to 2.1 in. For a weapon of a given caliber several different sizes of powder may be used, depending on the type of projectile. The smaller cylindrical grains have single axial perfora- tions, and the larger ones have seven perforations parallel to the axis. Six of these latter perforations aro at the vertices of a regular hexagon and the seventh is at the center, the dis- tances. between centers cf all adjacent pairs being nearly’' equal. Such niultiperforated powders are said to bo progressively granu- lated, because the burning surface increases as burning proceeds, and hence the rate of burning increases. The web thickness, ivhich is the minimum thickness of the grain between any two boundary surfaces, either internal or ex- ternal, determines the time required to consume a powder of given composition. The web thickness of a singly-perforated grain is g-(D-d) and that of a multipart orated grain is taken as 4(D-3d), where D is the diameter of the grain and d is the diameter of the perforation [Naval Ordnance, 1939, p. 37]. Hence, the grain diameter is about three times the web thickness of a singly-perforated grain, and from five to ten times the web thickness of a multiperfcrated grain, 16/ 2Ц. Chemical composition Gunpowder — frequently called black powder — was used as a propellant until 188Ц, when Vieilia introduced colloidal smokeless 16/ The effect of the chemical composition of propellants on erosion is discussed, in Sec. 106,
- 35 - powder made of nitrocellulose. Today the two general typos of propellant used by all nations are single-base and double-base powders. A single-base powder is one in vrhich the principal in- gredient is nitrocellulose. A double-base powder contains a considerable quantity of nitroglycerin in addition to nitrocellu- lose. In both typos tho nitrocellulose is gelatinized by suit- able solvents. For a single-base powder ether and alcohol are usually used. The nitroglycerin in double-base powder assists considerably in the gelatinization, so that if enough nitro- glycerin is present, no solvent is needed; otherwise acetone is used. Doublo-basc powders, which give butter ballistic results than single-base powders, have been standard propellants in Great Britain and some other countries for many years. . Because their erosive effect is greater, however, double-base powders have not been used extensively in the United Status. 25. American propellents For many years the standard smokeless propollent powder used in this country was pyro powder. This is made from "pyrocellu- lose11 (a nitrocellulose containing 12.6 percent nitrogen) to which is added 0.5 to 1.0 percent diphonylamino as a stabilizer. This is still the standard powder in the Navy. J.Iuch of it is still on hand in the Army, especially for the largor caliber guns; but NH powder ("nonhygroscopic11) and FNH powder ("flashless and non- hygroscopic") are tho onos being produced today. The first lot of FNH powder was made in 192Ц, but the total axiount produced
- зб - before 1930 was very small, . Therefore> much of the data on the erosion of guns in the U,. S. services, apply to guns fired with pyro powder. The composition of the various smokeless powders, other than pyro powder, currently used by the U, S, Army are listed in Table VI, which is based on the confidential specifications of the Ordnance Department. The NH and FNH powders keep bettor than pyro powder because they arc less hygroscopic. This property is obtained by the inclusion of dinitrotoluene (DNT) and dibutyl- phthalate. Double-base powders in general are less hygroscopic than singlo-base ones, and hence much less DNT is needed in I.I2- than in Ш-type powder. These sane two compounds servo also as coolants in the Mi-type FNH powder, causing the powder gases to emerge from the muzzle at a temperature below their flash point when nixed with the oxygen of the air. The flash from FNH, М2 powder is repressed by the inclusion of barium and potassium nitrates. In some of the snail arms powders it is repressed by potassium sulfate, and this compound can be used also as an aid _18/ • in the Mi-type cannon powder. Flashlessness, which depends, on granulation and ignition as well as on composition, has been achieved only to a partial degree in large-caliber weapons. The potassium nitrate used in some .of the powders improves tho igni- tion. The tin and tin oxide were used as docopporing agents 17/ Mr; Bruce E, Andersen, Ammunition Division, Industrial Service, Ordnance Department, personal communication, ^ug. 29, 19ll1< 18/ FHN/P powder also contains potassium sulfate^ sec last footnote (*) under Table VIB.
- 37 - (see Soo. 6?). The graphite glazing on the М2 cannon powder, which is also used sometimes on other powders of small granula- tions, is simply 0 lubricant to make the powder flow more readily and therefore pack bettor. Attempts have been made from time to time to develop a powder that would have a high propellent, power without causing a high temperature in the gun, and that would therefore be flashless and presumably less erosive. The most successful of such cool powders 12/ arc those based on nitreguanidinc, also called picrite, The Du Pont Company experimented with nitroguanidine powders for several years after World War I, but abandoned then after the development of the present FNH powdor. They 'were found excellent as regards flashlessness, but were objectionable because of smoke and ammonia fumes. Also the weight of charge was larger than for pyro powder — 15 percent larger for a 75-mm gun [Storm, 1925>]. Those objections are considered by the British to be outweighed by the advantages of flashlossnoss and reduced erosion (Sec, 108); and so they arc using two picrite powdors (see Table VIB), 26. British powders In 1888 a special committee adopted. as standard for the British services a smokeless powder node by gelatinizing a mixture of highly nitrated nitrocellulose and nitroglycerin with acetone. It was called cordite because it was formed in cords of various *19/ Chemical formula, NH3C(;NH)NHNOg, attempts to make the dinitro derivative, which would presumably have a higher poten- tial, have failed. (Dr. G. B. Kistiakowsky, private communica- tion, ^.pr, 2Ц, 1912.)
- 33 - Table VI. Nominal composition in parts by weight of smokeless powders. A. American Powders Ingredient Cannon Powders Small Arms Powders NH,M1 FNH,M1 FNH,M2 No.5 DIR 1185 Dffi Ц676 IMR hl 66 IMR h8lh ECa Nitrocellulose13 8?.O 85.0 76A5C 99.0 91.0 100.0 97.5 100.0 80. od Nitroglycerin 19.5 Dinitrotoluene 10.0 10.0 1.0 1.0 6.5 7.5 10.0 8.5 Diphenylamine 1.0 1.0 0.60 1.0 0.5 0.70 0.6 0.70 0.75 D ibu tylphthalate 3.0 5.0 Potassium sulfate 1.0 1.0 1.0 Barium nitrate 1.Ц 8.0 Potassium nitrate 0.75 8.0 Tin 1.93 Tin dioxide 1.5 Graphite glazing 0.30 aAlso contains 3.0 parts of starch and 0.25 parts of aurine, a red coloring agent. ^Containing 13.15 percent nitrogen. Containing 13.25 percent nitrogen. ^Containing 13.20 percent nitrogen. В. British Powders (Cordites}# Ingredient Mk. I M.D. к. W.M. s.c. N. N.Q. Nitrocellulose® 37 65 65 65 h9.5r 19 21.5’ Nitroglycerin 58 30 29 29.5 hl .5 18.5 20 Picrite 5U.7 5U.7 Mineral jelly 5 5 6 3.5 Carbamite 2 9 7.5 3.5 Chalk oA 0.2 0.15 Cryolite 0.3 0.3 Containing 13.1 percent nitrogen, p Containing 12.2 percent nitrogen. *After Littler (19h2). To save space we have omitted from this table NH, FNH and FNH/p, which are American-made single-base powders used by the British services. The first two have essentially the compositions given in Table VIA. The FNH/P contains 1 percent of potassium sulfate in addition to the ingredients listed for FNH,M1.
- 39 - Notes for Table VIA (1) Abbreviations♦ — NH, nonhygroscopic; FNH, flashless and nonhygroscopic; IMP,, improved military rifle; EC, Explosive Company, the name of the English manufacturers who first made this powder aft- er its invention in 1882 [Marshall, 1917> p. h8.J; AP, armor-piercing, (2) Use of small arms ammunition. — Caliber ,22, no standard composition. Caliber .1x5, Du Pont Pistol No, 3, Caliber .30 (ball and AP), 1925-hO, ЛЖ1185 principally, but also several lots of pyro and FNH powders; since the summer of 191x1, IMRU676. Caliber, ,30 (blank), EC, Caliber ,50 (ball and AP), before 1928, various smokeless powders to which different metallic compounds had been added; 1928-1x1, EM166; since the spring of 19M, HM8lli. Caliber ,30 (blank), EC. Notes for Table VIВ (1) General notes on British cordites. — Mark I, the original British cordite (1889)5 still made for pistol and ’'blank'1 ammunition, M,D,, manufactured on a large scale for small arms ammunition. W,, Land Service propellant of improved stability; largely replaced li,D, after 1932 for major caliber guns. W ,M,, "Emergency11 Land Service propellant; introduced April 19h0; replaces W, S.C,, the principal Naval propellant; developed by Research Department, Woolwich Arsenal and introduced about 1923; made by solventless process. N,, mainly used for antiaircraft guns (3,7 and h.5-in.), but flashless charges have been determined for most guns and howitzers. N,Q., mainly used for smaller guns (for example, Q.F, 6-pdr) and howitzers (for example, Q,F, 25-pdr, B.L. 3.5-in., etc,), (2) Special notes on nomenclature of flashless propellants. — The shape of the grain (other than cords) or the source of the nitrocellulose is denoted by the following additions after the soliduss tube =/T; slotted tube =/S; multitube =/M; nitrocellu- lose from cotton vraste =/A; nitrocellulose from wood =/F. The web thickness in thousandths of an inch is denoted by a figure following the letter; thus, "cordite N/М. 01l8."
- ЦО - diameters. The original cordite Mk. I caused so much erosion that the amount of nitroglycerin was greatly reduced to give cordite M.D. (= "modified"). The mineral jelly (vaseline) in it serves the double purpose of coolant and preservative [Marhsall, 1917, I: 30h-f>] • The compositions, of these and other more recently devel*« oped British powders are listed in Table VIB. Ballistite was invented by Alfred Nobel in 1888 by gelatin- izing a nitrocellulose of low nitration with nitroglycerin. The ratio of nitrocellulose to nitroglycerin has varied from h.O/60 to 5o/5O, and various stabilizers have been added. Ballistite or powders of nearly equivalent composition ’were used by the German and Italian governments at one time [Marshall, 1917, I: 301-2]. 2?. Stability All smokeless powders deteriorate during storage. Among - the.products of decomposition are oxides of nitrogen, which pro- gressively accelerate further decomposition. Stabilizers act by combining with these oxides and preventing the acceleration, without actually stopping the decomposition. The common 20/ stabilizers are diphenylamine, vaseline and "centralite." Powder that is deteriorating produces objectionable fumes even before its ballistic properties are impaired. Air-dried pyro powder has been found to last 30 to Ц0 years, but water-dried pyro powder lasts ". somewhat shorter time, depending upon the 20/ Also called "carbatnite." The chemical, name is s-diethyldiphcnyl urea [Pike, 1?h2].
- ы - siae of granulation and various conditions of manufacture); On tho basis of accelerated testss NH and FNH powders sere expected to last considerably longer than even air-dried pyro powder. . Because both moisture and high temperatures accelerate to some extent the deterioration of smokeless powder, considerable care is taken to protect it during storage. 28. Primers The initial impulse required for the ignition of a charge of powder is obtained from a small amount of highly sensitive explosive that can be readily set off by percussion, friction or heat. Smokeless powder in all but the finest granulations is so difficult to ignite that in addition to the detonating element most primers contain an igniter charge of black powder which burns very rapidljr with a hot flame and ignites the main powder charge. The best ballistic results might be obtained if the ignition were simultaneous throughout the charge. This condition is approached in practice by tho method of distribu- tion of tho igniter charge, a perforated tube along the axis of the chamber being frequently used .to contain the black powder.- Tho proportion of black powder to smokeless powder varies, be- cause the sane primer is often used with cartridge cases of several different calibers.
- Ц2 - CHAPTER V. ' BURNING OF THE POWDER 2?. Effect of pressure on rate of burning A colloidal propellant burns by parallel layers. This law, first enunciated by Piobert [1839] in the case of gunpowder, has been verified for colloidal propellants by the observation that unburned pieces of powder recovered after having been fired in a gun retain their original chemical and physical properties be- low the surface [Crow and Grimshaw, 1931b, p. 392; Muraour and Aunis, 1938.] Recent observations, however, show that under certain conditions the burning takes place below the surface as £1/ well. The rate of burning increases very rapidly with increase of pressure of the liberated gases. Thus the rate is increased about a thousand-fold by increase of pressure from 1 to 2^00 atmos The highest pressure to which smokeless powder can be sub- jected without having it detonate or burn erratically does not seem to have been determined. During the test of a certain 37-nim gun having an extra strong barrel, a series of five rounds were fired at pressures ranging from 63,hOO to 67,900 lb/in?, as measured by a piezoelectric gage. The velocities, which varied from 3355 to 3375 ft/sec, were acceptably uniform. Hence, as far as the burning of the powder is concerned, a pressure as high as this might perhaps be used in a hypervelocity gun. 21/ R. E. Gibson, personal communication, July 19^2.
- 43 - 30* The mechanism of the burning process The surface of a burning solid propellant remains relatively cold while a gaseous reaction at a very high temperature (Sec. 36) takes place at a distance of approximately 10'*'’ cm from the surface. The slow step in the complex process is the transfer of heat to the surface by the molecules of the burning gas that impinge upon it, for a surface layer of the solid must be kept at the temperature of spontaneous decomposition in order that burning may continue.’ Because an increase of pressure means that more hot gas molecules are available to transfer energy, the rate of burning is increased by pressure. This view of the burning process has been elaborated recently by Boys and Corner [1941] and by Lennard-Jones and Corner [1941]. Crow and Grimshaw [1931b] and Muraour [1931] had suggest- ed essentially the same theory, except that they had postulated a cruder method of heat transfer, 22/ Earlier investigators had derived empirical equations of the form, . 1. c r = a + bp , where r is. the rate of regression of the burning surface, £ is the pressure, and a, b and c are empirical constants. Frequently c was considered unity and a was very small or zero. In such equations, however, the apparent rate-of-burning constant, b, had been found to increase und^r conditions giving rise to increased 22/ See Crew and Grimshaw [1931b, pp. 387-8, 397] for bibliography, Bennott [1921, p. vii], in the'preparation of his Tablas for interior ballistics, assumed that the rate of burning of pyro powder is proportional to the two-third powder of the pressure.
- йй — cooling. This tendency was explained by Crow and Grimshaw by showing that b is in reality a function of the temperature of the powder gases * Instead of assuming that the heating of the surface of the powder to the ignition temperature results from molecular impact, Kent [1935] suggested that radiation from the powder gases is the cause. He argued that the escape of volatiles from smokeless powder while it is being xvarmed up prevents contact of hot powder gas with the surface, whereas radiation could penetrate to the surface. He explained various experimental phenomena by this hypothesis. Ignition of powder at temperatures below the normal igni- tion point by photochemical decomposition caused by the absorp- tion of radiation of selected frequencies was suggested as a possibility by Crawford [1937], in his proposal for experiments involving the application of spectroscopy and photography to interior ballistics. He also suggested the possibility of the photosensitization of the ignition of powder by radiation from cool mercury vapor emitting resonance radiation of wave-length 23/ 253?A. Later Crawford remarked that the efficacy of potassium in. promoting ignition might be explained as the result of emis- sion of radiation of just the proper frequency by excited potassium ions. He suggested that this^hypothesis could be test- ed by exposing some powder to irradiated potassium vapor in an evacuated tube. 23/ Maj. D. J. Crawford, Jr., Ordnance Department, personal communication, Aug. 30j 19k1.
- U5 - 31. Quickness of powder The rats of burning of a powder —- often referred to as its quickness — is one of the most important factors in determining the effectiveness of the powder for a particular purpose. The criterion ’which is applied to the burning of a powder in a gun is that a maximum proportion of the energy of the powder shall bo transferred to the projectile without allowing the powder pressure to exceed the limit imposed by the strength of the walls of the tube. If the powder burns too fast, a dangerously high pressure will be set upj whereas if the powder burns too slowly, some of it will remain unburnod when the projectile loaves the muzzle, and hence its energy will bo wasted. Tho rate of burning is con- trolled by variation of the form and the dimensions of the grains, particularly the web thicknoss. The smaller the web, the faster will a powder bum, because for a given total weight, powder grains with small web have a larger surface area. Direct measurement of the quickness of a powder is now pos- sible by a method developed at tho Burnside Laboratory of the Du Pont Company. While a Sccmplc of the powder is being burned in a closed chamber, an automatic oscillograph record is made of dp/dt, the time-rate of change of the pressure, as a function of the pressure. By comparison of this result with that obtained with a standard powder, the ballistic properties of which had been determined by firings in a particular gun, it is possible to calcu- late the charge of powder required to achieve a desired velocity in a similar gun.
- Ц6 - The uniformity of the rate of burning is also important; for if the rate is irregular, pressure waves may be set up in the bore of the gun. Such waves are potentially dangerous, for by rein- forcement of two of them, a pressure large enough to burst the gun may be attained. Uniformity of burning is aided by proper design of the ignition system, striving toward the ideal situation of enveloping the charge of smokeless powder in a hot flame from the igniter charge so that all the grains are ignited simultaneously. 2Ц/ 32. Heat of explosion and heat of combustion The heat of explosion of a propellent powder is the quantity of heat evolved when the powder burns at constant volume in a closed chamber without added oxygen. It is less than the true heat of combustion, which is the heat liberated when the same powder is burned at constant volume in an excess of oxygen. (Some viritors, as for instance do Bruin and de Pauw [1928] , designate the heat of explosion as the heat of combustion.) The heat of explo- sion is of more practical interest, because it measures the 25/ potential of the powder — that is, its capacity for doing work. 24/ A summary of the heats of formation of the constituents of propellants has been given by Schmidt [193й] and by Pike [19Ц2], 2$/ This definition of powder potential follows Hayes [1938. p. 5ojT J. 0. Hirschfelder [personal' communication, Aug. 29, 19^2] supports an earlier suggestion by R. H. Kent in recommending that the powder potential be taken as the product of T.,, the adiabatic flame temperature (Sec. 36), and Cv, the high-tempera- ture molal heat capacity of the gas at constant volume. This change of definition is equivalent to considering Cv in the pro- duct TeCv as the average value between, say, 2000°K and the adiabatic flame temperature instead of as the average value be- tween 0°C and the adiabatic flame temperature [compare Eq. (V-1) in Sec. 36], The potential calculated in this way is sone 20 percent larger than the heat of explosion.
- и? - MJ In the calorimetric determination of the heat of explosion the calorimeter is usually at or near room temperature, and there- fore the water formed during the reaction is condensed and its heat of vaporization appears as part of the apparent heat of ex- plosion. To obtain the true heat of explosion (water gaseous) the amount of water must be deterrained. The correction for the heat of vaporization amounts to slightly less than 10 percent of the- apparent heat of explosion (water liquid) [Noble 1906, pp. й?6-80; de Bruin and de Pauw 1928, Tables III and IV]. Values for the heats of explosion and of combustion of a number of powders appear in Table VII. The results of Dunkle [1935] showed that the ratio of the heat of explosion Q to the heat of combustion H of a smokeless povzder is a linear function of the ratio of the number of molos cf atomic oxygen per gram of powder to the sum of the corresponding numbers for carbon and hydrogen [0/(C+H2)] in the powder; and also that it is a linear function of the corresponding ratio [C02] »I.H20]/[C0] • [H2] of the products of explosion of the powder. The slight variation of 27/ the heat of explosion with the density of loading [Noble 1906, Plato 14], which is related to the question of change of composi- tion of the products of explosion during coding, is discussed in Sec. 39. 26/ Noble [1905, pp. 216-20]; de Bruin and de Pauw [1928, 1929]; Crow and Grimshaw [ 1931 a]. 27/ The density of loading is the ratio of the weight of charge to the weight of water that would fill the chamber at 4 °C; hence in the metric system it is numerically equal to the weight of charge in grains divided by "the volume of the chamber in milliliters.
- 1x8 - Table VII. Heat of explosion, heat of combustion and temperature of explosion of propellent powders.* Compositiona (percent by wt.) Nitrocellulose Powders Double-base Powders FNH Powder N3 X-927** C2 1511** 1358** Nitrocellulose (12.6^ N) 900 95.23 (13.21 N) 70.16 78.20 83.97 Nitroglycerin ——— ——— 28.6? 19.80 — Dinitrotolueno — — —— ——— 9.88 Dibutylphthalate ——— —— — —— L.9U D iphenylamine o.U5 0.1x9 —- 0.99 1.00 Volatile matter lx.75 Lx.28 1.17 1.01 0.22 Total 100.00 100.00 100.00 100.00 100.01 H(kilocal/gm)b 831 861 1220 1153 751x Q(kilоcal/gm)c —— 2б11х — 2377 2980 Temp. (°C)d 21x85 — 3U33 — T—** Temp. calc. (°C)4' 2Ц00 21x25 3300 3075 — aSee Table IX for composition of products of combustion of these same powders. ^Heat of explosion, water liquid. Heat of combustion. ^Calculated by Crew and Grimshaw from heat of explosion and heat capacity of pewdor gases. “Calculated for this report from composition by method of de Bruin end de Pauw [1929]. (J.S.B.) 'rCrow and Grimshaw [1931b]. ^Dunkle [1935].
- 49 - Table VIII, Products of combustion of gunpowder J*-5, Loading^3 Pressure0 0.10 1.6 . o.5o 10.7 0.90 35.6 Gases Composition (percentage) Wt. Vol. Wt. Vol. Wt. Vol. C02 CO H2 CH^ H2S 25.97 3.03 12.01 o.o5 0.06 O.lll 48.99 8.98 35.60 2.07 0.29 4.06 25.22 5.63 10.22 0.0? 0.16 0.66 47.21 17.04 30.29 3.01 0.84 1.61 27.50 3.56 10.85 0.03 0.15 0.77 52.65 Ю.73 32.64 1.27 0.81 1.90 Total 41.53d 99.99 41 .96 100.00 42,86 100.00 Solids Composition (percentage) K2CO3 k2s2o3 K2sot K2S KCNS KNO3 (nhu)2co3 c Free S 30.07 11.66 11.71 2.30 0.00 0.32 o.o3 ’ 0.72 ' 0.41 35.20 14.70 2.69 2.06 0.17 0.29, 0,06 • 0.00 . ' 2.8'7 37.55 4.91 4.8? 4.13 0.21 0.11 0.09 0.00 5.27 Total A 57.22, 58.04 57.14 *After Noble qnd Abel [18?51, Tables 2, 3, 4. ^Large-grain rifle powder^ Waltham Abbey. Composition: KNO3, 74.95%; KsS0i , 0.15%; S, 10.2?%; charcoal, 13.52%; water, 1.11%. ^Proportion of space occupied by the charge; not "density of loading." « cLong tens per square inch. ^Should be 42.78% - error in original for one of the individual percentages.
- 50 - 33. Products of combustion of gunpcwder The composition of the products of conbustipn of propellent ponders after explosion in a closed chamber have been determined by various investigators. Noble and Abel [1875], for instance, as a part of their classic research on Fired gunpowder analysed the products of combustion of several kinds of gunpowder at different densities of loading. Their results for a representa- tive gunpowder at three densities of loading are given in Table VIII. Over half the weight of the material is solid, which is responsible for the smoko from burning gunpowder. Among gaseous products the volume percentages of nitrogen and of carbon dioxide are much greater than from colloidal powder, whereas the percent- ages of carbon monoxide and of hydrogen are much less (compare Tables VIII and IX), These differences reflect the higher nitro- gen and oxygen contents cf gunpowder as compared with smokeless powders. Table IX. Products of combustion of smokeless povrders, in moles per kilogram of powder, Table VII, in which these powders are listed in the same order, gives their composition. Product Nitrocellulose Powders .Double-Base Powders FNH Powder N3* X-927* 02* 1511** 1358** C02 3.200 5.2Ц2 6.607 8.322 3.292 CO 20.21Ш 16Л8Ц 12.771 11.704 2O.5U3 ' Ha0 8.82? 5.922 11.272 7.923 5.001 N2 u.269 5.242 5.195 5.604 5.894 H2 6.979 8.219 1.676 5.120 9.139 *Crow and Grimshaw [ 1931b] . *H!Dunklc [1?35] • Density of loading, 0.12.
- 51 - 3t. Products of combustion of smokeless powder Because smokeless powder burns without leaving any appreciable solid residue, the determination of its products of combustion is just a specialized application of the standard technic of gas analy- sis. The water formed during combustion, after it has condensed in the explosion chamber, is cither wiped out and v/eighed as liquid water [Noble, 1905, p. 203], or else it is distilled out of the bomb and absorbed [de Bruin and de Pauw, 1928, p. 9], which is pre- ferable. Analyses of the products of combustion obtained from different kinds of smokeless powder have been published by Sarrau and Vieillc [188Ц], by Noble [189h], by Macnab and Ristori [189Ц, by Noble [19O6] and by de Bruin and de Pauw [1928]. A number of unpublished reports from Picatinny Arsenal [Valente, 192 9; Dunkle and Volsk, 1932; Volsk and Dunklo, 1932; and Dunkle, 1935] contain analyses for the new FNH powders, and also for some experimental powders of special compositions. A selection of results from Dunkle [1935] at a density of loading of 0.12 is given in Table IX in tho columns headed X-9273 1511 and 1358. The variation in composition of powder gases with change of density of loading is illustrated by the results of Noble, some of which are reproduced in Table X. h'horeas the amounts of nitrogen and of wator show only slight variation with density of loading for all three powders, the amount of carbon monoxide decreases while that of carbon dioxide increases as the density of loading, and hence the pressure, is increased. Similarly the ratio of methane to hydrogen increases very greatly with increase of density of
Table X* Variation in composition pf powder gases with change of density c-f iouding.’^ ' “ A. Composition of powders, in percentages by weight. Component''^ NC C-I C-M.D. Nitrocellulose (sol.) 85.5 37.0 63.o Nitrocellulose (insol.) 1h.5 Nitroglycerin 38.0 30.0 Mineral jelly (vaseline) 5.0 3.0 B, Pressure and volume of gas. Density of Loading Pressure . (atnos) Total Gas . (ml/gm) Permanent Gas (ml/gm) _ж_ C-I C-M.D. .110 C-I c-e.o. ... NC C-I C-M.D. o,o3 h37 h93 h37 993.1 870.0 933. h 81Ц.7 670.0 781.8 .10 973 1066 990 980.0 878.3 9h8.O 810.5 692.5 790.0 .13 1З16 1783 1600 938.3 880.0 931.0 795.0 699.0 786.5 .20 2103 2391 233h 93h.O 873.3 913.5 776.0 697.0 773.0 .23 2789 3h67 3193 906.3 863.0 893.5 751.5 688.0 75h.O .30 3h98 hhoh ho6? 883.0 8h8.O 873.0 730.0 671.5 733.5 .hO 3228 6302 3830 8h1.0 820.0 832.0 695.0 6hh.9 693.5 .30 7133 812Ц 73hh 802.0 798.8 789.3 659.5 623.6 653.5 *After Noble [1901, pp, Ш-7, h80]. NO, nitrocellulose; C-I, cordite Mark I; C-M.D,, cordite M.D. [Table X continued cn next page,]
- 53 - TABLE X. (Continued.) G. Total composition in percentage by volume. Density of Leading (gm/ml) o.o5 0.10 o'.i5 i o.2o| 0.2$ 0.30 o.ho o.5o I. Nitrocellulose co2 111.95 16.19 17.50 19.110 21.50 23.50 27.60 31.60 co 35.36 3h.6O 33.35 31.90 29.95 28.25 211.65 20.96 H 20.01 20.67 19.90 18.00 15.80 ili.o5 11.60 10.00 CHU O.h9 0.71 1.58' 2.89 li.8O 6.20 8.I1I1 9.80 N 11.19 10.75 10.62 10.83 11.011 11.16 11.00 11.00 H20 18.00 17.08 17.05 16.98 16.91 16.8a 16.71 16.611 Total 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 IT. Cordite Lark I co2 21.07 21 .98 23.13 211.37 25.8? 27,611 30.56 32.118 co 26.116 26.13 25.20 23.91 22,31 20,1111 17.116 15ЛЗ H 13.53 15.12 15.60 15.53 111,96 13.93. 11.73 9.117 снц 0.23 0.32 o.51i 1.00 1.65 2,115 3.98 5.51i N 15.97 15.08 111.93 111.86 111.90 111.95 15.11 15.60 H20 22.7I1 21.37 20.60 20.33 20.31, 20.59 21.16 21.118 Total . 100.00 100.00 100.00 100,00 100.00 100,00 100.00 100.00 III. Cordite M,D. C02 111.85 16.115 18 35 20.30 22,25 211,15 28.15 32.35 CO 311.87 33.95 32 hO 30.50 28.115 26.115 22.60 19.00 H 18.95 19.20 19 00 18.00 16.65 15.00 11.80 8.115 СНц 0.29 0.83 1 60 2.70 3.95 5 .Цо 7.во 9.95 N 12.89 12.57 12 115 12.50 12.65 12.85' 13.05 13.1.15 нго 18.15 17.00 16 20 16.00 16.05 16.15 16.60 16.80 Total 100.00 100.00 100 00 100.00 100.00 100.00 100.00 100.00
- 5Ь - loading. The cause of these variations is considered in Sec. 39 in connection with a discussion of the gaseous equilibria exist- ing at the temperature of explosion. 33, Composition of powder gases from a gun The experiments that have boon referred to so far were per- formed with closed chambers. A few analyses have been made, how- ever, of the gases formed in an actual gun. Thus Noble [1906, pp. h.55-6] obtained the results shown in the second column of Table XI by analysis of a sample of gas withdrawn from the breech end of a 9.2-in, gun just after it had fired a charge of 103 lb of cordite, giving a pressure of 16.1 tcns/in? and a muzzle velo- city of 2б00 ft/soc. For comparison, tho last two columns of this table give Noble’s analyses of the gases obtained from firing charges of cordito at a density of loading of 0.05 in a closed chamber, which in one case had been exhausted and in the other had been initially filled with air at atmospheric pressure. The com- position of the gases in the gun agrees closely with those in the closed chamber. Poppenbcrg and Stephan [1909b] found from an analysis of gaseous mixtures taken at various parts of the barrel that there is more carbon monoxide t the moment of explosion than there is after the gases have cooled down. The presence of'considerable quantities of iron in the powder gases was claimed by Siwy [1908], but he published no supporting evidence. He did not even describe the method of detection.
-Sa- lable XI, Products of combustion of cordite in a gun. compared with those obtained in a closed chamber, in percentages. Product Gun (Mean of Two Rounds) Closed Chamber With Air Evacuated co2 25.6 г?.15 26.35 CO 35.8 3U.35 35.05 H 19.0 17.50 19.50 CHU 0.5 о.зо 0.60 N 19.1 20.70 18.50 Total 100.0 100.00 100.00 .• 36, Explosion temperature 28/ The theoretical explosion temperature refers to the tempera- ture to which tho products of an explosion would be raised if the reaction took place adiabatically. This temperature on the centi- grade scale is TG * Q/CVJ (V-1) where Q is the racial heat of explosion at constant volume (referred to products cooled to 0cG), summed over all the components of the original explosive, and Gv is the mean racial heat capacity from 0 to TcC, summed ever all tho products of the explosion. Calcula- tions of the explosion temperature based on this aquation arc dis- cussed in the next section. Various attempts have boon made to measure the temperature of explosion directly. Noble and Abel (1875; p. 1?1 in 1906 reprint] 28/ Also called adiabatic flame temperature.
- $6 - showed that the temperature of explosion of gunpowder is higher than the melting point of platinum, by firing a very thin sheet of platinum and a coil of fine platinum wire with large charges of gunpowder in a closed chamber. Portions of the wire were weld- ed to the sheet, and the-, shoot itself showed the beginning of fusion'when examined under a microscope. Measurement of the explosion temperature of smokeless powders by means of thermocouples was attempted by Macnab and Ristori [1900J . Because the . heat losff is so groat, the naxfiw-l tempera- ture recorded by a platinum-p^atinrhodium thermocouple in a closed chamber depends on the diameter of the wire. Each couple of a series of ten made of wires ranging in diameter from 0.010 to O.OhO in. was used to measure the temperature of explosion of the same charge of powder under the sane conditions, The enf of the galva- nometer was recorded as a function of time on a fixed photographic plate fitted with a moving slit. The maximum deflection was given by the couple made of the thinnest wires; the deflections of the others wore progressively less. It was hoped to extrapolate the deflections to a thermocouple of zero thickness, and to calibrate them in terms of temperature; but the attempt was not entirely successful* It was demonstrated that the explosion temperatures of the three explosives, guncotton, cordite, ballistite, increased in the order named, which is the same order as that obtained from calculated temperatures. It has been suggested that explosion temperatures might be
- 57 - 29/ measured by means of the spectra of the burning gas, a method that has been discussed recently by Brinkman [i9ho], The line- reversal method has been used successfully in measuring the flame temperatures in the internal combustion engine. It was suggested by Lewis and von Elbe [1938д p. 337] that this method might be adapted to the measurement of the explosion temperature in a closed chamber ”by suspending highly dispersed sodium salts in explosion mixtures and recording the reversal point photographi- cally by progressive adjustment of the comparison radiator.’1’ 37. Explosion pressure The explosion pressure is the maximum pressure that would be developed by the powder gases if the reaction took place adiabati- cally in a closed chamber. It is related to the temperature of explosion TQ by the equation of state of the pewder gases. At the high temperatures of explosion, the internolecular attraction term in the equation of state can be neglected, and the following simple form of equation suffices: Po (v-oc) = RTg, (V-2) where v is the specific volume of the powder gases at the pressure P , a. is thoir average cuvplume and R is tho gas constant per gram. 29/ Thu temperature of the powder gases burning inside a closeTchamber has been determined recently at tho Geophysical Laboratory by measuring with photoelectric cells the intensity of radiation -in different parts of the spectrum that is transmit- ted by a quartz window in the side of the chamber. This technic will bo described by F. 0. Kracek, ff. Benedict and L. G. Bonner in a forthcoming NDRC report.
- 58 - 3£/ Since the density of loading A is the reciprocal of v, this equa- tion nay be written in the form Pe = ДНТэ/(1 (V-3) Since ex. is approximately 0.001 for the six gases commonly found as products of combustion of powder [Sarrau, 1893; Burlot, 192Ц], the explosion pressure increases almost linearly with an increase in density of loading. Because the number of moles of gas per gram of powder depends on the gas composition, the constant R for a given charge also depends on tho composition. Because of the rapidity with which the powder gases lose their 31/ heat to the walls of the chamber, the pressure measured in a closed chamber never quite attains the theoretical maximum. For the same reason it is doubtful whether tho adiabatic explosion temperature is actually reached. Some of Noble's experiments [1905. P. 227] demonstrated the extreme rapidity of cooling. Thus the pressure of the powder gases from cordite fired in a closed chamber decreased from 39,000 Ib/in? to 60 percent of that value in 1 sec. The loss of pressure is greater at a low density of leading. Noble, nevertheless, did not correct the observed pressures, and hence his measurements — some of which are reproduced in Table X,B — show considorable departure from linearity at low densities of loading 30/ This relation assumes that the weight V/ of tho powder gases is the same as the weight of the original charge, Then v = V/U, and in the metric system, A is numerically equal to W/V, where V is the volume of tho chamber (see footnote 27). The expansion of the chamber by the pressure is neglected. 31/ See Chap. X for methods of measurement.
- 59 - and approach linearity at high densities. He had previously cone to the conclusion [Noble and Abel, 18?5] that the pressure loss due to the cooling would not exceed 1 percent. However, the spe- cial experinents of Muraour [1923, 1925a, and 1925b] and of Crow and Grimshaw [1931b] have shown that this loss nay easily exceed 10 percent, especially at low densities of loading or in a snail chamber. In a gun the conditions are much more complex than in a closed chamber, because of the expansion of the gases as the pro- jectile moves forward. They will be considered in Sec. 50 in con- junction with a discussion of the movement of the projectile. 38. Approximate calculation cf temperature of powder gases The temperature of the powder gases at the tine of explosion may be calculated by either of two independent methods, which aro based on Eqs. (V-1) and (V-3), respectively. The former is related to the thermochemical properties cf the powder and of the evolved gases. Equation (V-1) seems to have been used first for this pur- pose by Mallard and Le Chatelier [1883]. Equation (V-3) was devel- oped by Abel as an empirical relation in the form P = XA/(1 - a A) , (V-h) where A and oc were empirical constants, and it is often referred to as the Abel, or the Noble and Abel, equation. The calculation of the temperature of explosion from the maxi- mum pressure developed in a closed chamber, in accordance with Eq. (V-3), involves two difficulties. The equation itself is only an approximation, and there is not at present sufficient knowledge of
- 60 - tho kinetic properties of gases under the conditions of tempera- ture and pressure concerned for it to be improved. Tho covoluno, in particular, is imperfectly known. An equally important diffi- culty is that of measuring the true pressure of the explosion. As already mentioned, tho gas is cooled sc quickly by the walls of tho chamber that the maximum- pressure measured is somewhat less than the true pressure. The correction to bo added to the observed pressure was evaluated by Muraour [1923, p. 323; 1925a, p, h60] from experiments in a closed chamber in which the area of the sur- face exposed to the gases could be varied. It was calculated by Crow and Grimshaw [1931а» P« 50] as an example of heat conduction, from pressures measured in two chambers of widely different sur- face areas. In the early applications of Eq. (V-1) to the calculation of the temperature of explosion, Cv was evaluated on the basis of the composition of the cooled powder gases. It was expressed as a linear function of the temoerature; vrhoreupon Eq. (V-1) could be 32/ solved for T. It was soon realized, howeverthat the products of combustion found in a closed chamber after it has cooled to room temperature are not necessarily the same as those present under the conditions of high temperature and high pressure exist- ing at tho time of the explosion, and also that thq heat capaci- ties of gases at high temperatures are not linear functions of the temperature. 32/ Seo Hayes [1938» p. 53] for an exposition of this s olution.
- 61 - 39. Secondary reactions during cooling Experiments intended to elucidate the changes that take place during cooling of the products of the explosion were con- ducted by Poppenberg and Stephan. They confirmed Noble’s ob- servations (see Table X) that the ratio к = £Hs0] • [CO]/[CO2].’[H2] decreased with increase of density of loading. They explained this variation by the assumption that the mixture cooled more slowly at high densities of loading, and hence that the "water gas" reaction [Eq. (V-6)] had more time to proceed. This view v/as substantiated by firings [Poppenberg and Stephan, 1909a] made at the same density of loading with the explosion chamber cooled to -120°C in one case and heated to 90°C in the other. More carbon dioxide was produced at the high than at the low temperature. The formation of methane was shown to be a secondary re- action by later experiments made by Poppenberg and Stephan [19*1 On] and by Poppenberg [1923]. bi some of them the explosion took place inside a small porcelain bomb contained in an evacuated chamber. The sudden expansion of the gases when tho porcelain bomb burst so cooled them that the equilibrium 'was frozen. Criticism by Kast [1910] of the interpretation of those results was met by a further discussion by Poppenberg and Stephan [1910b]. Their conclusion with respect to tho secondary formation of meth- ane was confinaed by Muraour [1919a], who quenched the equilibri- um mixture by having it expand through a 1-m canal into an evac- uated vessel. It also is in agreement with the latest calcula- tions [Kent, 1938a].
- 62. - bO, Calculation of corrections for composition of powder gases After it was realized that the composition of the cooled gases reflected secondary reactions during cooling, methods of calcula- tion were developed to correct for then. These corrections, com- bined with improved,data for heat capacities of gases at high tem- peratures and with increasingly precise measurements of explosion pressures, gradually decreased the discrepancy between the values of explosion temperatures calculated from the maximum pressure and those calculated from heat capacities. At first [Noble, 1906, pp, Ц63-5; Popponberg, 190.9; Muraour, 1919b] the temperatures calcu- lated from the maximum pressure were much too low, except at-high densities of loading. In a later comparison [Muraour} 1g2ba, 1g2bb, 192bc, 1926] the disagreement was much loss.. Finally the very thorough investigation of Grow and Grimshaw [1931b] resulted in agreement within 2 percent for the value of X — tho ballis- tic force of the propellant — derived from the heat capacity and heat of explosion and that derived from direct measurements of the explosion pressures at different densities of loading. This func- tion is.the leading coefficient in the equation of state adopted by Crow end Grimshaw, ' . Po = A (A + «.А3 + a«.3A3 + (V-5) v/hich is merely a more elaborate form of the Noble and Abel equa- tion, Eq, (V-b). The mpthod of calculation used by do Bruin and de Pauw [1928] and by Schmidt [192g] tc correct for the change in gas composition by tho secondary reactions in cooling consisted in calculating the
- 63 - corrections to tho ’composition of the cooled, leases, by application 33/ of the equilibrium constant for tho secondary water-gas reaction, CO + H20 C02 + H2. (V-6) 3h/ 111, Calculation of temperature and composition of powder gases Tho method of Henderson and Hasse [1922], which was later used by Poppenberg [1923] and by Crow and Grimshaw [1931b], is superior in that it dees not require a knowledge of tho composition of the cooled gases. By a series of successive approximations the equi- librium composition of the hot gases is derived from the following experimental data: (i) the molal heat of explosion at constant volume; (ii) tho chemical composition of the powder in terms of the chemical elements present; (iii) the equilibrium constant as a func- tion of the temperature for each of the reactions in which the pro- ducts of combustion may be assumed to participate; and (iv) the mean molal heat capacity from 0cC to the explosion temperature for each of the substances that it is assumed might be present in the explosion. The molal proportions of tho chemical species assumed to participate in the equilibria at the temperature of the explosion 33/ For a discussion of this reaction, particularly in flames, see Bono and -Tovmsend [ 192 7 j Р» 322]. 3h/ Hirachfelder, McClure and Curtiss [19112 j have systema- tized the calculation of the adiabatic flame temperature and the composition of the powder gas. Their method is based on the same data, as is the one described in Sec. ill; but instead of carrying out the approximation by assuming successive values for the flame temperature, they calculate the energy loss at two different assumed flame temperatures, and then, by linear interpolation be- tween these values for that flame temperature which would corre- spond to zero energy loss, they obtain the true flame temperature.
- 6Ц - form the unknowns of a series of simultaneous algebraic equations. Ona group of these equations of condition is based on the known chemical composition of the powder. The other is derived from the equilibrium constants for the assumed equilibria, evaluated at a provisionally assumed temperature. The solution of these simul- taneous equations gives the first approximation to the molal com- position at the time of explosion. (Henderson and Hasse con- sidered only the water-gas reaction in the main calculation, and thon used the equilibrium constants for various other possible re- actions to obtain second-order corrections.) From this result and the assumed temperature, tho mean molal heat capacity Cv of the whole mixture is calculated, and then tho experimental value for the heat of explosion is divided by Gv to compute tho temperature [Eq. (V-1)]. This value of the temperature, ’which will differ somewhat from that originally assumed, is used as a basis for a repetition of the calculation. Finally, after one or more repeti- tions, the computed temperature will agree with tho assumed tempera- ture. This value and tho corresponding molal composition are con- sidered as those of the burning gases. This method of calculation, which involves the simultaneous consideration of a number of equilibria, was used by Kent [1938a] in a recalculation of the composition of the gases from two of the powders previously considered by Crow and Grimshaw [1931b]. The equilibria involving the formation of methane and nitric oxide and those for the dissociation of -water and nitrogen were included in addition to the water-gas reaction, which had been tho only one
- 65 - 35/ considered by Grow and Grimshaw, This recalculation made use of the latest values for heat.capacities of gases at high tempera- tures, obtained largely from measurements of band spectra [David and Leah, 193U; Lewis and von Elbe, 1938, Chap. XVI]. The results, which are reproduced in. Table XII, show that the early assumption of the dominance of the water-gas reaction was correct. Later un- published computations [Kent and Lane, 19Ц1] involving a total of twelve different equilibria have not changed the results materi- ally. The water-gas reaction does not involve a volume change if the gases are ideal, and hence the equilibrium is not affected by change of pressure. On this basis it has boon argued that the apparent change of explosion temperature and of heat of explosion with variation in density of loading arc spurious effects. The existence of reactions involving volume changes in the powder gases, however, brings in a slight variation of the. explosion temperature with change of pressure, as shown in Table XII. h.2. Equilibrium conditions during burning Although the calculations just discussed furnish us with our best knowledge of the composition and temperature of powder gases at the time of explosion, there remain two possible sources of uncertainty, in addition to the uncertainties of the experimental data involved in the calculation. The variation of the heat ca- pacity of the gases with pressure [Muracur, 192ца, p. 330] has 35/ Hirschfaldor, hicClure and Curtiss (reference Зй) have repeated the calculation, using a variation of this method with the additional refinement of correcting for the fact that the gas is nonideal.
- 66 - Table XII. Partial pressures, in atmospheres, of the constituents of powder gases, and tempera' ture of explosion. Powder Nitrocellulose Ballistite Pressure (atmes) 3U0 3U00 3U0 3U00 CO 138.Ц8 1386.5 81.0U 756.8 C02 UO.92 U11.0 73.86 827.5 H20 82.00 827.9 88.89 96U.8 n2 36.17 326.6 50.8U 52.2.5 H2 ЦО.69 U07.U 12.60 111.2 02 —— 5.08 26.0 N 0.01 — 0.17 1.3 H .8U 2.9 3.11 -16.1 0 .01 — 1.76 7.2 OH 1.02 3.5 18.99 133.9 NO 0.06 0.2 3.86 3U.7 Temp.(°C) 2663 2689 3398 368U *Aftcr Kent [1938] . not been taken into account. Moesveld [1928, p. 3U] concluded, however, on the basis of the van de? Vfaals equation, that the variation is negligible at high temperatures. The other source of uncertainty, namely, tho assumption of a state of equilibrium among the atoms of a gas, is more impor- tant, because this assumption underlies all the calculations, Muraour [l92Ua, p. 335] suggested that during combustion equilib- rium may nut be maintained^ oxygen, for instance, may react more rapidly with hydrogen than with carbon. He considered, never- theless, that equilibrium would be established as soon as burn- ing was completed.
- 67 - Lewis and von Elbe [1938, p. 306] pointed out that in the absorption of the energy of explosion by the gas molecules there maybe a lag in the excitation of the translational degrees of freedom of carbon dioxide, carbon monoxide and nitrogen. The resulting momentary excess of translational energy causes "a higher pressure which lasts for a period comparable with explo- sion times. Ultimately, energy equilibrium among all tho degrees of freedom is established. Thus, tho heat capacity of those gases is a time function." A somewhat similar possibility of nonoquilibrium conditions is that suggested by Crawford [1937], who considered that because of the very rapid rate of burning in a gun, tho gas kinetic energy of tho products of combustion lags behind the radiant energy field of the explosion system. This would cause a considerable deviation from tho van der Waals equation. An investigation of equilibrium has been made recently at Aberdeen Proving Ground in collaboration with Professor H, C, Urey [Kent and Lane, 19^1]. A small amount of barium carbonate containing a heavy isotope of carbon was fired with a charge of powder, and then the distribution of the heavy isotope between carbon monoxide and carbon dioxide was determined in the pro- ducts of tho combustion after they had been quenched by rapid expansion. According to a preliminary report of these experi- 36/ ments, the heavy isotope was found uniformly distributed 36/ R. H. Kent, personal communication to R, E. Gibson, Jan. 19U2.
- 68 - between the two gases. This is considered as evidence that chemical equilibrium had been reached during the combustion of the powder.
- 69 - CHAPTER VI. MOVEMENT OF THE PROJECTILE IN THE- BbRE ’ Ths detailed mathematical description cf ths movement of the projectile in the boro of a ;gun, which is one of tho princi- 37/ pal problems of interior ballistics, is too involved to be pre- sented in this report. Instead, a few outstanding details of that description will be mentioned in their relation to erosion. hU. Rate of increase of pressure The rate at which the pressure increases behind a projec- tile in a gun depends on the interaction of several factors. It is controlled principally, of course, by tho rate of burning of the powder, but this reciprocally depends on the pressure. In a closed chamber the pressure rises at an ever-increasing rate until the powder is nearly all burned. In a gun, however, tho rate of increase of pressure is partly checked by the in- crease cf volume caused by the movement of the projectile. At first the expansion is slow, because tho pressure is low and also because the resistance to movement of the projectile is high until the rotating band has been engraved by the rifling (soe Sec. Ц6). Later tho expansion is fast enough to offset partially tho in- creased rate of burning cf the powder. Next the amount of burn- ing powder becomes so small that the evolution of gas begins to 37/ The Le Luc equation for the velocity of the projectile in the bore, which is used in designing new guns, is described in Hayes [1938, pp. 7U-81] and in Naval ordnance [1939, pp. 50-85]• A more refined form of senienpirieal calculation is afforded by the use of tho Tables for interior ballistics pre- pared by Bennett [1921]. '
- 70 - decrease. This effect combines with the over-increasing expan- sion to cause the rate of increase of pressure to become zero. After the attainment cf a maxinun pressure, the remainder of the powder is burned within a short interval of time during which the pressure decreases slowly. Finally the pressure continues to decrease approximately as in an adiabatic expansion. The curves of pressure versus time for several guns (Fig. 1) clearly show those variations cf the pressure. They arc photographic records made by means of a cathode-ray oscillograph attached tc a piezo- electric gage during the firing of actual guns. Position of projectile at maximum pressure Although the time required for the pressure to build up to a maximum is a considerable fraction of the total time the projec- tile is in the bore, up to the moment of maximum pressure the pro- jectile has moved only a few caliber lengths from its initial posi- tion. At the time of maximum pressure that part.of the bore of the gun between the breech and tho base of the projectile is subject to tho maximum pressure. After that the pressure decreases slow- ly as the projectile continues tc move toward tho muzzle. The muzzle pressure, the exact value of which depends on the speed of burning of the powder and on the length of the tube, is of the order of one-fourth the maximum pressure. 38/ h-6. Friction of the projectile in the bore W Both static and dynamic measurements have been made of 4-t------------------;........................................ . 38/ See Soc. 53 for heat of friction. 39/ Some of the earliest static measurements were made at Watertown Arsenal with the "United States Testing Machine" and wore reported in Tests of metals published annually by that Arsenal*
Г> S'S'&-/с/ ’/ '^/_j/
- 72 - the initial resistance of the projectile to movement, caused by the work required to engrave the rotating band — or the soft metal jacket, in the case of a bullet — and the subsequent fric- tion between the boro and the two parts of the projectile that touch it, namely, the rotating band (Sec. 18) and the bourrelet (Sec. 17). bi one case, for instance, the resistance in a worn 75-nm gun, M1897 (fired 7755 rounds) was found to be I4.0,000 lb at the origin of rifling and about 15,000 lb at the muzzle [Ord. Dept., 1922b]. Tho actual resistance to a moving projectile, however, is probably much less than the result of such a static measurement. This was indicated by the fact that when projec- -tiles having rotating bands of reduced diameter wore fired from a 21i0-mn howitzer, practically no effect on pressure or velocity was found [Ord. Dept., 1922a]. this conclusion is in agreement with the suggestion of Justrow [1923] that the friction between the band and the bore corresponds approximately to that of a piston adjusted for suction. A summary of several dynamic methods of measuring bore fric- tion has been given by Kutterer [19^2]. In Cranz' method a shot is fired from a shortened tube from which it initially protrudes about one-third its length. The engraving resistance is taken equal to the mean pressure corresponding to a series of equal charges which are just sufficient to eject the projectile about half the time. In several other methods the frictional force is evaluated as the difference between the force tending to push the projectile forward and the force of recoil. In Sebert’s method the frictional force is evaluated as the difference be- tween the force acting on the projectile and that acting on a
- 73 - frictionless piston inserted in on axial hole through the shot. These two forces are measured by differentiation of two space- time curves obtained by means of recording tuning forks on frictionless slides attached to the shot and the piston. Liebessart improved Sebert’s method by measuring the displace- ment optically. The method of Cranz and Schardin [1932] utilizes a breech block consisting of a heavy frictionless piston, having the same area as the projectile, so that the forces on them are the same. Hence, when the gun is fired from a free recoil carriage, the tube is pulled forward by the frictional drag. The.frictional force is the product of the mass of the tube and its accelera- tion. The latter is obtained by tho double differentiation of the space-tine curve for the tubs’. Kutterer [19h2] modified this method by measuring the acceleration of the tube directly by attaching to it. a frictionless piston which registered against a piezoelectric quartz gage. The methods mentioned have required a modification of the operation of the gun or projectile. Tho friction of a standard projectile in the bore of a 1h-in. Navy gun was determined in- directly during an investigation of the ballistics of this gun [Bur. Stand., 192 Ца] . The pressure on the base of the projec- tile computed from the acceleration of recoil was less than the gas pressure in-the chamber measured with a. spring pressure gage. Цо/ This difference was considered a measure of tho friction. ЦО/ Preparations are being made for application of this methoa to measurement of the friction and also other ballistic properties of several guns of moderate caliber, under the super- vision of.H, L. Curtis of the National Bureau of Standards, for Section A, Division A, NDPC.
- 7h - kl/ A direct-reading gage has been suggested recently by Smith [19h2]. It consists of a quartz piezoelectric gage mounted in the base of the projectile and connected to an oscillograph by means of a wire that comes out the muzzle of the gun. UY* Relation of friction to erosion The divergent views concerning the influence of the rotat- ing band on erosion have already ^en mentioned (Sec..19). The magnitude of neither that influence nor that of friction between the bourrelet and the bore has been evaluated. Nonsymmetrical U2/ muzzle erosion has been ascribed to the wear between the bourrelet and a particular land with which it remains in contact during most of its path down the bore. It is generally agreed, however, that the amount of erosion caused by friction between the projectile and the bore is greater in small arms than in guns firing projectiles vd_th rotating bands. The difference between the erosion in the two kinds of weapons, however, seems to be merely one of degree rather than of kind. This point of view was emphasized by Greaves, Abram and Rees [1929, p. 158], who suggested that although the total amount of metal eroded per round increases with caliber (sec Table XVIII), a smaller proportion is removed by friction in the larger guns. This Ц/ This gage is now being used in a series of measure- ments with a 3-in. gun at Aberdeen Proving Ground under the auspices of Section A, Division A, NDRC. See Crocker and Smith U2/ Arthur E. Jewell, Gage Section, Aberdeen Proving Ground, personal communication.
- 7$ - conclusion is based on the fact that the force applied to unit area of the surface of contact between the rotating band and the bore is about the same for all guns fired at the sane pressure, whereas the heat input to unit area of the bore surface increases very rapidly — because the ratio of the weight of charge to bore surface increases linearly — with an increase of caliber. Il8. Velocity of the projectile in the tube The velocity of the projectile continues to increase as the gases expand after the tine of maxinum pressure. Noble invented the chronoscope for the determination of the velocity at different points down the tube by recording the tines when the projectile passed these points. In the original arrangement described by Noble and Abel [187!?], the tube of the gun was drilled at a series of points along its length. Each hole was fitted with a plug which carried a wire arranged in such a way that it was cut when the projectile passed that point. Cutting the wire inter- rupted the primary current of an induction coil, whereupon a spark recorded itself on a piece of paper attached to a disk, 36 in. in circumference, that was rotating with a constant speed of about 1250 rev/min. The instrument was capable of recording the millionth part of a second, and when it was in good working, order, the probable error of a single observation did not exceed h or 5 x 10"6 sec. nfter the distance of travel of the projec- tile had been expressed as a function of tho tine, the velocity and the pressure were computed by two differentiations. A some- what similar contacting arrangement was used at Aberdeen Proving
- 76 - Ground [Kent and Hitchcock, 1923] in a study of tho internal ballistics of a.2U0-mm howitzer» A scries of 13 contacting plugs were connected to an oscillograph to obtain a direct tine- travel record, 1x9. Muzzle velocity and energy The nuzzle velocity of the projectile, which is an impor- tant characteristic of the gun, is determined by measuring with 43/ a chronograph the time required for the projectile to pass be- tween two velocity frames set up at measured distances in front of the gun, and then applying a correction to take care of the small change of velocity between the nuzzle and the point of measurement. For the first 10 yd or so tho velocity is increased by a few feet per second by the rush of the escaping gases, and then it decreases. The muzzle velocities of the guns in use to- day generally range from 15OO to 3000 ft/sec at full charge. If the full range is not needed, sone of then, howitzers especially, are fired at reduced charge, which yields a lower muzzle velocity. The muzzle energy, which is a measure of tho poxver of the projectile, is determined by the nuzzle velocity, the velocity of rotation and the mass of the projectile, Tho proportion of the powder energy represented by the muzzle velocity is the thermo- dynamic efficiency of the gun. This efficiency is about 30 per- cent, as is shown in Table XV (Sec. 92). The mechanical effi- ciency of a gun, considered as the ratio of the muzzle energy to 1x3/ See Ordnance Dept. [1936, No. 1x0-17] for a description of the procedure of this measurement.
- 77 - the work represented by the area of the ideal indicator diagram, was calculated by Henderson and Hasse [1922p, Ц73] as 8h to 90 percent for a number of British грдпз fired with cordite M.D. at nuzzle velocities of about 2700 ft/sec. So, Effect of expansion of powder gases Although the pressure in a gun firing a service round is about the sane as in a closed chamber at a density of loading of 0,20 to 0,2$ (see Tables I and X), the gases in the two en- closures arc not subject to the sane conditions up to the moment of maximum pressure. In the closed chamber this pressure is attained without the performance of any work by the gas, where- as in the gun the gas expands at the same tine that the pressure increases. Because the secondary water-gas reaction [Eq, (V-6)] involves no volume change if the gases are ideal, this expansion has less effect oh the gross composition of the powder gases than it otherwise night. On the other hand, a possibility which nay be of special significance in connection with erosion is that the expansion may promote some reaction that is affected by the change of pressure, and thereby cause the. formation in the gas mixture of a chemical combination that might readily react with tho steel bore. Another possible cause of a local disturbance of the gase- ous equilibrium is the cooling of the gas. In a 3-in. antiaij>- craft gun, for instance, the cooling has been found by calcula- tion to be about 1000°C by the tine the gases reach the muzzle, which corresponds to a .gradient of about 3° рог centimeter
- 78 - Ы±/ travel of the projectile. Most of this cooling is caused by the work performed in pushing the projectile down the bore. Because it affects the mass of gas as a whole, any change of composition resulting from this part of the cooling is distributed uniformly. Part of the cooling, however, is caused by contact of some of the gas molecules with the cold surface of the bore; and this change is not uniform. Not only is the cooling confined to the layer of molecules adjacent to the wall of the bore, but the return of these molecules to the main mass of gas is hindered because they are drawn forward by the movement of the projectile. This uni- directional gas stream may have a different composition from that of the rest of the gas, since an exothermic reaction would be pro- moted by cooling. This possibility may be important with respect to erosion, because this stream of gas adjacent to the surface of the bore might react with it. ИД/ The adiabatic explosion temperature (see Sec, 36) is independent of the amount of the propellant. The temperature of the gases at the muzzle, however, is lower the smaller the amount of charge. For a 3-in, AA gun Ю, the muzzle gas temperature was calculated to be 1бДо°С for a charge of U.8h lb of pyro powder, for which the explosion temperature was 2?00°C. With a charge of 5.00 lb of NH powder, which had an explosion temperature of 2Д00°С the muzzle gas temperature was calculated to be 1Д30°С [Tolch, 1936b, p. Д].
- 79 - CHAPTER VII. REACTION OF THE TUBE TO FIRING $1. Temperature of the tube Direct measurements of the temperature of a gun tube haflre been made by means of thermocouples welded to its outer surface. The average results obtained for a number of different guns at Aberdeen Proving Ground are summarized in Table XIII, which has been copied from a survey by Lane [1938] . In arriving at this average rise of temperature per round, allowance was made for cooling effects when the fire was slow enough to warrant such a correction. In the case of the 3-in. AA gun Ю, for instance, the cooling was found to be unimportant at rates of fire of 10 rounds/min and more [Tolch, 1936b]. The larger temperature rise per round shown in Table XIII for the measurements near the muzzle compared with those near the breech end of the tube re- sulted from the heating of a smaller amount of metal rather than from the presence of a larger quantity of heat. This situation is demonstrated by the figures in the last column, which show that the most heat was absorbed by the part of the tube farthest from the muzzle. Translation of these results into terms of the heating of a gun under service conditions is difficult except in the cases Of very rapid fire or of sustained fire at a steady rate. The ex- pected temperature rise for a single burst fired very rapidly is practically the product of the number of rounds fired in one burst and the average rise in temperature per round. Thus Lane [1935] calculated that the bore surface of an aircraft machine gun reached
Table Kill. Heating of gun tabes firing single-base powder* Gun Projectile and Velocity Firings Tube Diam. (in.) Dist. from Muzzle (in.) Max. Temp. Measured (°C) Average Temp. Rise (°C/md) Average Heat Absorbed (cal/cm2 rnd) Cal. .30 tank machine gun 172-gm ball 2600 ft/sec Mean of thou- sands of rounds 1.21 1.21 3.7 11.2 200-400 Maintained <> • О p 3> i d «3 OJ CO C\ 1 1 ' " ' | co t L_! Cal. .50 KoG., heavy barrel air-cooled 750-gm ball 2600 ft/sec 100 md and 2(199) rnd with 15 min cooling between groups 1.5 1.87 2.31 1.9b 5.25 17.1 23.5 27.25 38b 370 314 322 1.36 1.26 .89 1.13 3.3b b,95 5.b5 li. 76 Cal. .50 M.G., water-cooled 750-gm ball 2б00 ft/sec 2(100) rnd and 1(198) rnd —— 100Ь — - 75-mm M1897E3 15.96-lb slug 1755 ft/sec Group of 300 md b.80 5.28 5.67 12 36 55.5 320 328 328 1.51 l.bo 1.33 b.55 1 5.65 co 6.51i 3-in. AA T8 15-lb slug 2600 ft/sec Four groups 100 md each 5.03 5.86 12 b8 352 295 3.30 2.39 11.15 12.5b 3-in. AA М3 15-lb slug 2600 ft/sec Group of 2b7 rnd 5.95 7.28 13 63 b30 360 2.18 1.59 12.0 lb.2 Ю5-ПШ1 AA M1 33-lb slug mod. 2800 ft/sec Group of 169 md 6.95 8.33 9.69 12 88 16b 285 303 282 2.01 2.06 1.89 9.5 16.2 21.8 *After Lane [1938, Table I]. Corresponding values for double-base powder were 1,1b and l55°C/md and 3.011 and b.25 oal/cm2rnd. ^Temperature of outside barrel practically constant at temperature of boiling water.
-81 - 137O°C after having fired 600 rounds in min. In this calcula- tion the cooling effect was allowed for to the extent of h. per- cent of the heat input. In calculating the temperature of the tube after sustained fire at a steady rate it is necessary to make allowance for the cooling, which depends materially on the wind velocity. Cooling rates for the same guns listed in Table XIII have been- compiled separately by Lane [1939a]. One important way of expressing the heating of the tube is in terms of the rate of fire necessary to maintain a particular temperature. For the caliber .30 machine gun these rates, meas- ured directly for temperatures of 200°, 300° and ЦОО°С,above ambient temperature, are given in Table XIV in comparison with similar rates that had been calculated from the results of temper- ature measurements with other guns. Such data are of practical importance in deciding on allowable maximum rates of fire; for if the temperature at the breech end of the tube increases too much, the steel will be weakened to such an extent that the gun may burst. Tolch [1936a] calculated for the 10^-mm AA gun M1 that at 16Ц in, from the muzzle the maximum allowable temperature was йО£°С; that is, at this temperature the tensile strength of the steel had decreased to such an extent that the factor of safety was reduced to unity. Tho effect of artificial cooling of the barrel, as by the use of a self-circulating water cooler, is illustrated by the data in Tables XIII and XV for the air-cooled and water-cooled caliber .5>0 machine guns. After similar firing of the same ammunition, the
- 82 - Table XIV» Temperature of the breech end of the gun tube as a function of the rate of fire; single-base povrder. Temperature Above Eate of Fire Necessary to Maintain Temperature (rounds/min) Ambient C'c ' Cal. .,30lCal. Л0 bi.O. Q. a i r.iv_r>on?r>r'c ...... 7/ Ml 897E3C 3-in. AA Gun M3d 105-mm AA Gun Ше 200 1U 8.7 3,3 1.1 0.9 300 17.8 14.3 3.5 1.8 1.4 4oo 22J 20.8 7,7 —_ 2.6 2.0 Corresponding rates of fire obtained with double-bo.se powder were; 200^C, 9.1;'300cC, 1^.4; 400°C, 18.8 md/min [Tolch, Ш9 . bTolch [1936c]. cThis gun had.been modified by the removal of. the bronze jacket from tho regular gun of this model. It was concluded as a result of the temperature measurements during firing tests that ’’the increase in heating effect due to the modification is partially, if not ’wholly, compensated for by the increase in cooling rate1’ [Dickinson, 1936] . ^Tolch [1936b 1. eTolch [1936a]. barrel of the former had reached an average temperature of nearly ЗбО°С, whereas that of the latter was not much above 100°C. The quantity of heat absorbed by tho water-cooled barrel was nearly 40 percent greater than that absorbed by the air-cooled one; and yet its projectile derived 1 percent more energy from the same amount of powder. Thus the lowered temperature of the barrel, which tends to increase the life of the barrel, was secured with- out any loss of efficiency. i>2. Distribution of energy in gun The proportion of tho total energy of the powder that is absorbed by the gun tube is shown in Table XV in comparison with
- 83- - tho distribution of the remainder of the energy of the powdor. The amount shown as absorbed by the tube includes that absorbed by the brass cartridge cases. For the caliber .50 machine guns this part was found by a calorimetric measurement to amount to about 300 cal, which raised the temperature of the case to about 81°C. The percentages of energy absorbed by the tube, shown in Table XV, agree reasonably well with the figure of 22 percent calculated by Cranz and Rothe [1908] and with that of 15 percent deduced by Muraour [1925, p. 1x77] from closed-chamber measure- ments . Table XV, Percentage distribution of energy of powder, Distribution Cal. ,5o М2 Machine Gun 3-in, AA Gun Юс Air-cooled, Pyro Powdera Water-cooled, Pyro Powder^ Pyro Powder NH Powder Velocity and spin of projectile Heat of gun Velocity of gases Heat of gases 28 16 27) > 56 29 J WO 29 22 271 >59 22 J Too 29 11 60 100 32 9 $9 Energy (cal/rnd) 11,31x0 11,31x0 1,035,000 1,51x0,000 aMuzzle velocity, 2600 ft/sec. Fired 5>00 rounds in three groups at different rates [Tolch, 1936c], ^Muzzle velocity, 2600 ft/sec. Fired 1x00 rounds in three groups at different rates [Tolch, 19371 . &Muzzle velocity, 2800 ft/sec. Pyro powder: 15 rounds in U5 min; NH powder; 25 rounds in 80 min [Tolch, 1936b], 53. Heat of friction Transfer of some of the energy of the powder to the tube takes place indirectly by means of the heat of friction of the
- ВЦ - projectile as it passes down the bore (see Sec. Ц6) as well as directly by contact of the hot gases. A measurement of the ex- tent of this indirect transferwas attempted [Ord. Dept. 192Ц] by comparison of the heat absorbed by two 37-mm guns that had been fired 100 rounds each under the same conditions, except that the projectiles for the one gun had standard bands, where- as those for the other gun had very narrow bands, the cylindri- cal portion having been 0.03 in. wide instead of 0.5 in. Each gun was immersed in the same quantity of water at the end of firing. The rise in temperature of the water was 5°C in both cases. Hence, as far as this somewhat rough measurement was concerned, the heat of friction was probably less than 10 per- cent of the heat absorbed from the powder gases. 54. Temperature of the surface of the bore No direct measurement of the temperature of the surface of the boro of a gun has been attempted during firing. Some measure ments have been made directly after firing by pushing a thermo- couple junction into the muzzle of the gun and holding it against the bore surface with an asbestos pad. Such a procedure yields a result of very questionable validity, because of cooling before the measurement can be made. The closest approach to a direct measurement of the tempera- ture of the bore surface was made by Kosting [15351 on a Garand caliber .30 automatic rifle. A thermocouple junction was placed at the bottom of a hole drilled into the barrel 7 in. from the breech end (0.06 in. in front of origin of rifling) so that the
-•85 w U5/ junction was only 0.077 in- iron the. bore surface. The maximum temperature after firing £1 rounds in 1.3 min, with the tube already hot from previous firing, amounted to U85°C. At the same time the temperature at the surface on the sane radius was b28°C, Hence the temperature of the bore surface was estimated to be somewhat in excess of 5>00°G. Measurements at several other points along the barrel were made simultaneously by means of recording potentiometers. They indicated that longitudinal heat flow was much greater than radial heat flew. The temperature of the bore surface was computed from that of the outside surface of the tube of a 10^-mm AA gun Ml by Tolch [1936a, p. 10], using the Fourier equation for heat flow. The greatest difference between inside and outside temperatures was at the measuring point nearest to the breech. There the maximum outside temperature was 279°C, and the corresponding bore tempera- ture was calculated to be 3h9°C; but this result depends on several somewhat doubtful assumptions. The calculation of the temperature to which the bore surface is raised by a single round is even more uncertain. Lane [1939b] developed a method suggested by R. H. Kent that is based on a known rate of heat input. For a 75-mn gun an increase of 179°C U5/ In a private communication, dated Aug. -22, 19^2, Kosting pointed out “that because of the use of the recording potentiometer the time lag would be such as to measure average temperatures and not peak temperatures, and also that the thermo- couples placed within the holo drilled into the barrel were held in place by friction and were not spot-welded to the steel. I now look with disfavor on any experimental work for evaluating temperatures in which the thermocouple is not spot-welded at the point where the temperature is to be measured.1*
- 86 - was indicated at the interface, whereas it was only 2° at a depth of 0.0^5 in. below the surface. Kent [19h1] himself criticized this method, on the grounds that it assumes that the heat capacity and conductivity of the surface layer of the bore are those of the steel throughout the tube, whereas he pointed out that possibly these properties are considerably different in the altered surface layer. He recommended that an attempt be made to measure them. 55. Dilation of the bore Part of the pressure of the powder gases during firing neutralizes the residual compressional stress that is present in a modern gun as a result of the method of manufacture (see Sec. 6), and then the remainder of the pressure places the tube in a state of tension. There is no quantitative information as to how much compressive stress is retained by a built-up gun or by a radially expanded (cold-worked) one. The final machining operations, especially the rifling, must relieve some of the compressive stress, and later, when the gun is in service, probably more of it is gradually dissipated. The calculation of the increase of diameter of the bore from the pressure is further complicated by the nonhonogeneity of the stress distribution that is occasioned w by the rifling. Justrow [1923] rather naively disregarded these Ц6/ Dr. P. R. Kosting (private communication, Aug. 22, 19h.2) remarks that recent British firings have shown that the engraved portions of the rotating bands of some recovered projectiles were larger than the diameter of the bore of the gun. Hence the tube must have expanded ahead of the rotating band; and therefore the stress distribution in the tube is more complicated than if the gun is considered merely as a cylinder subject to a uniform inter- nal pressure.
- 8? - factors in his elaborate calculations of the dilation of the bore. The bore diameter is also changed by the rise of temperature. If the whole tube heats up gradually, the inside diameter increases. Thus for a 105-mm gun the increase would amount to about O.OOli in. for each 100°C rise of temperature above ambient [Tolch, 1936a, p. 6] . 56. Contraction of the bore Contraction of the bore may occur whenever the bore layers are stressed beyond their yield point, which varies inversely as some function of the temperature. A sudden increase of 100°C in the temperature of the bore would correspond to an increase of 2 4?/ 30,000 lb/in, in the compressional stress, Fleming [1918] as- cribed to this cause the complete elimination of the rifling in a Lewis antiaircraft machine gun that had been fired 2500 rounds automatically (cyclic rate not known), during which time the barrel had been observed to have become red hot. He suggested that the projectile acted as a mandrel in the contracted bore. 57• Elongation of the tube Compressive stress in the inner layers of the tube may be re- lieved by elongation of the tube. This action is assisted by the longitudinal stress caused by friction between the projectile and the bore. Liners have been known to be caused to protrude in this manner, and even 15-in. gun tubes have been reported [Miller, 1920, p. 53] to have been permanently elongated as much as J-in. Ц7/ R* W. Goranson, private communication.
- 88 - 58. Tangential stress The rotary motion of the projectile during firing imposes a considerable tangential stress on the tube. For а 2Ц0-тт howitzer this stress was calculated under various assumptions to be as high as 9000 lb/in^ It was therefore considered to have been a factor that contributed to the failure of a number of such guns by burst- ing or swelling [Ord. Dept., 1931]. Justrow [1923] attributed to the tangential stress on the sides of the lands considerable in- fluence as a cause of erosion.
- 89 - BIBLIOGRAPHY AND AUTHOR INDExkl/ for PART ONE A. C. [19U2c] (Sec. 18) ’’Deformation of driving bands on 3-in. proof shot/’ Adv, Council Sci. Res. Tech. Dev. to British Ministry of Supply, A. C. 2000 (Apr. 28, 19h2). Bender, V. E. and J. N. Pappas [19k1 ] (Sec. 15) ’’Investigation of steel compositions for centrifugal castings that can be heat-treated to a yield strength of 121,000 to 135,000 lb/in.8," Watertown Arsenal Re pt. No. 3lh/U2 (July 22, 191+1), 6 p. Bennett, Albert A. [1921] (Sec. 30, h3) ’’Tables for interior ballistics,’’ Office of the Chief of Ord- nance, Washington, D.C. (Apr. 1921). Bone, W. A. and D. T. A. Townsend [1927] (Sec. Ц0) Flame and combustion in gases, Londons Longmans, Green (1927)t ------------------------------ Boys, S. F. and J. Corner [ 19li1 ] (Sep. 30) ’’Structure of reaction zone in burning of a colloidal propel- lant,’’ Adv. Council Sci. Res. Tech. Dev. to British Ministry of Supply, A. C. 1139 (Aug. 22, 19Ш). Brinkman, H, [19Ц0] (Sec. 36) ["Measurement of high temperatures of gases by the emission spectra"], Nederland. Tijdschr. Natuurkunde 7, 65-85 (191|O); Ch. Abs. 3h, 7670. Bruin, G. de [1930] "La poudre sans fumee et I1erosion des bouches a feu. I. Court aper<£u historique et critique du sujet," Communique de la Ste. Anme. "Fabrique Neerlandaises d’Explosifs" No. 10 (1930), 39 P* English tr. by Lt. Col. Aiken Simons, Mil. Sales Div., E. I. du Pont de Nemours and Co. (Mar. 1931), h9 p.; Ord. Dept. Lib, Bruin, G. de and P. F. de Pauw [1928] (Sec. 32, 33, Ц0) "La temperature de combustion de la poudre sans fumee en fonction de sa composition chimique. II. Partie experimentale’’ [See Moesveld, 1928, for Part I»], Communique de la Ste. Anme. 4’Fabrique Neerlandaises d’Explosifs" No. 8 (1928); Ord. Dept. Lib, Ц7/ The section number given on the same line as the author’s name refers to the present report.
- 90 - Bruin. G. de and P. F. de Pauw [1 929] (Sec. 32, 33) "La temperature de combustion de la poudre sans fumec en fonction de sa composition chimique. III.,11 Communique de la Ste. Anme. "Fabrique Neerlandaises d’Explosifs" No. 9 (1929); Ord. Dept. Lib. Bruin, G. de and P. F. de Pauw [1931] (Sec. 13) "La poudre sans fumee et 1*erosion des bouches a feu. II* Partie experimentale," Communique de la Ste* Anme. "Fabrique Neerlan- daises d’Explosifs" No. 11 (1931), 27 p. English tr. by M. S. P., Mil. Intell. Div., War Dept.; Ordnance Dept. Library. Bur. Stand. [192Ца] (Sec. Ц6) "A digest of ballistic data on large naval guns obtained for the Bureau of Ordnance, Navy Department by the Bureau of Stand- ards," Washington, D.C. (Oct. 192Ц). Bur lew, J. S. [19h-2] (Sec. 7) A brief history of tapered bore guns, NDRC Report A-h3 (OSRD No. 5l5) (Apr. 15, 19U2). Burlot, E. [192Ц] (Sec. 37) "Note relative a la determination experimentale de la force et du covolume des explosifs," Mem. poudres 21, Ц11-90 (192Ц). Carl, Leroy R. [19h-0] (Sec. 22) "The breaking theory of detonation," J, Franklin Inst. 230, 75-9h, 207-227, 355-37U (19Ц0). Charbonnier, P. [1908] (Sec. 13) Balistique Intdrieure, Paris; Doin (1908). Charbonnier, P. [1922] (Sec. 13) "La veine gazeuse," Men. artillerie franqaise 1, 1001-11 (1922). Cranz, C. and R. Rothe [1908] (Sec. 52) "Temperatur und Warme Messungen am Infanterie gewehr M98S," Z. ges. Schiess u. Sprengstoffw. 3, 301-5, 327“32 (1908). Cranz, C. and H. Schardin [1932j (Sec. Ц6) "Eine neue Methode zur Messung des Geschosswiderstandes in Rohr," Z. tech. Physik 13, 12Ц (1932). Reprint translation in Ordnance Dept. Library. Crawford, D. J., Jr. [1937] (Sec. 30, Ц2) "The applicability of photography and spectroscopy to some problems in interior ballistics," Watertown Arsenal Rept. No. 122/9 (1937), 13 p. and bibliography of 5 p.
- 91 - Crocker, J. A. and N. M. Smith, Jr. [19U2] (Sec. Ц6) "Bore-frictipn gages,11 Memo, to H. H. Porter OSRD/Sec. t/TR 1107 (Mayli-, 19й2), 3 p. , Crow, A. D. and W. E. Grimshaw [1931a] (Sec. 32, 38) ’•On the equations of state of propellant gas os,” Trans. Roy. Soc. (London) A23O/;39-r?3 (1931)^ Crow, A. D. and W. E. Grimshaw [1931b] (Sec. 29, 30, 33» 37, hl) ’•The combustion of colloidal propellants,” Trans. Roy. Soc. (London) A230, 387-h.il (1931). David, W. T. and-A. S. Leah [193h] (Sec. hi) ’•Specific heat of gases at high temperatures,” Phil. Mag. 18, 307-21 (193h). de Bruin, G. See; Bruin, G. de de Marre, Jacob.- See: Marre, Jacob de de Pauw,. P. F. See;. Bruin, G. de and P. F. de Pauw Dickinson, W. D. [1 936] (Sec. 51) "Study of the heating and cooling df the bore of the 75-ram gun M189733,“Aberdeen P. G., Ballistic Res. Lab. Rept. No. Ц9 (Apr. 22, 1936), 6 p. Donald, W. G. [1937] (Sec. 15) ’•Investigation of high-strength steel compositions,” Water-? tow Arsenal Rept. No. 31Ц/2О (May 12, 1937), 7 p. Dunkle. Sec also: Volsk and Dunkle Dunkle, 0. G. [1935] (Sec. 32, 33, 3h) "Calorific values of smokeless powders as affected by varia- tions in composition, granulation, etc.,” Picatinny Arsenal Tech. Rept. No. 620 (Apr. 23, 1935), 13 p. Dunkle, G. and N. T. Volsk [1932] (Sec. 311) "Calorific values of smokeless powders as affected by variations in composition, granulation, etc.,” Picatinny Arsenal Tech. Rept. No. 231 (June’ 7, -1932), 7 p. Fleming,. J. .Р,- [-1918] .(Sec. 56) "Erosion tests of high silicon-chrome-vanadium steel in Lewis, aircraft automatic rifle barrels,” manuscript, Ordnance Dept. Library (June 1918).
- 92 - Gerlich, H. [1930] (Sec., 3) "Increased bullet speeds. Effects and possibilities of a German rifle and its ammunition,'1 Army Ordnance 1Q, -398-RDh (1930). (Reprinted in American Rifleman, Sept. 193G.) Gerlich, H. [1931] (Sec. 3) • 4An advanced era of bullet velocities. The attainment of high speeds and their effects on armor,11 Army Ordnance 12., wo-юз (1931). ; .. . Gerlich, H. [1933] (Sec. 3) "The ballistic efficiency of guns.. A great increase of velocity has been attained," Army Ordnance 13,215-9 (1933)* Greaves, R.. H., H. H. Abram and S. H, Rees [1 929] (Sec. 4?) "Erosion of guns," J. Iron and Steel Inst. 109,.’113-16? (1929). Includes bibliography of Ц p. Hayes, T. J., editor [1938] (Sec. 3, 5, 17, 32, Ц3) Elements of ordnance; A textbook for use of cadets of' the ULS?' Military Academy, New York: Wiley (1938), 715 p. Henderson, J. B. and H. P. Hasse [1922] (Sec. Ц1, Ц9) "A contribution to the thermodynamical, theory of explosions," Proc. Roy. Soc. (London) A100, Ц61-82 (1922); Mem. artillerie frangaise 2, 1078-1101 (1922J. Hirschfelder, J. 0., F. T.. McClure ahd 0. F. Curtiss [19Ц2] (Sec. Ul] Thermochemistry and the equation of state of propellant gases, M)RC Rept. A-h8 G5SrD~~No] 5U7) (Apr- 8, 19h2). Howe, Henry M. [1918] (Sec. 9) . • "The erosion of guns," Trans. Am. Inst. Min. Met. Engrs. 58, 513-96 (1918). Hugoniot and Sebert [1882] (Sec. 13) "Etudes des offets de la poudre dans un canon de 10 cm," Mem. artillerie marine 10 (1882). Justrow, Capt. [1923] (Sec. 9, 19, I4.6, 58) "Theoretical consideration concerning the life of the tubes of our guns, trench mortars, rifles and pistols and their relation to the behavior of projectiles," "Heerstechnik," (Jan. 1923), translation, Ordnance Dept. Library.
- 93 - Kast, H. [19Ю] (Sec. 39) ["The determination of explosion temperature’1], Z. ges, Schiess. u. Sprengstoffw. 5, 205-8, 248-54 (1910); Ch. Abs. 4, 2877. Kent, R. H. [1933] (Sec. 3) ’'Times of flight of a Gerlich 3-in. antiaircraft gun/’ first addendum and Appendix C of "First and final report of test of Halger-Ultra rifle designed by H, Gerlich," 29th partial Rapt, of Ordnance Program 4975 and 2nd partial Rept. of Ordnance Program £019, Aberdeen P. G. (Jan. 1933). Kent, R. H. [1935] (Sec. 30) "The influence of radiation on the ignition and burning of powders," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 9 (1935), 7 P. Kent, R. H. [1938a] (Sec. 39, 41) "The flight of the projectile," J. Franklin Inst. 226, 19-33 (1938). “ Kent, R. H. [19U1] (Sec. 54) Remarks at conference on gun erosion, Watertown Arsenal, Watertown, Mass. (Oct. 15, 1941). Report of conference, Watertown Arsenal, p. 15. Kent, R. H. and H. P. Hitchcock [1929] (Sec. 48) "Interior ballistic firings in the 240-min howitzer No. 292 on free recoil carriage," 4th partial Rept. of Ordnance Program 4259 (Sept. 19, 1929). Kent, R. H. and J. R. Lane [ 194-1 ] (Sec, 41, 42) Remarks at conference on gun eirosion, Watertown arsenal, Watertown, Mass. (Oct. 15, 1941). Report of conference, Y/atertown Arsenal, p. 33. Kosting, P. R. [1935] (Sec. 54) "Investigations of the cause of bore enlargement in the Garand caliber .30 semiautomatic rifle No. 49," Watertown Arsenal Rept. No. 739/3D (June 6, 1935). Kosting, P. R. [1939] (Sec. 19) "A metallographic and diffraction study of erosion of guns," manuscript, Ordnance Dept. Library, 3 p. Kutterer, R. E. [1942] (Sec. 46) "The measurement of shot resistance in the bore of a gun," Adv. Council Sci. Res. Tech. Dev. to British Ministry of Supply, л. C. 2244 (July 11, 1942). Extended abstract of
- 9h - articles in. Wehrtechnische Monatshefte 39, h96, $36 (1935) and in Меи. artillorio francaise 15 (193^). Lane, J. R. [1933] (Sec. 9, 19) Memorandum report on test of six 3-in. antiaircraft guns, Ml 925 — radially expanded, Ordnance Program R6l5 (Apr. 21|, 1933), 6p. Lane, J. R. [1935] (Sec. 51) ’’Temperature of an aircraft machine-gun barrel when fired at a rate of 1200 rnd/min,” Aberdeen P. G., Ballistic Res. Lab. Rept. No. 23 (Nov. 1, 1935), 6 p. Lane, J. R. [1938] (Sec. 51) ’’Heating of guns,” Aberdeen P. G., Ballistic Res. Lab. Rept. No. 10U (May 20, 1938), 12 p. Lane, J. R. [1939a] (Sec. 51) ’’The cooling of guns,” Aberdeen P. G., Ballistic Res; Lab. Rept. No. 1Ц6 (Apr. 26, 1939), 15 p. Lane, J. R. [1939b] (Sec. 5й) "Temperature of 75~mm cartridge case," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 159 (Sept. 27, 1939), 8 p. Lennard-Jones, J. E. and J. Corner [19Ц1] (Sec. 30) "Burning of propellant and effect of gas flow over its sur- face," Adv. Council Sci. Res. Tech. Bev. to British Ministry of Supply, A. C. 1173 (Sept. 5, 191+1). Lewis, Bernard and G. von Elbe [1938] (Sec. 36, hl, h2) Combustion, flames and, expl isions of gases, Cambridge Univ. Press (193Ю, Ul5 p. Littler, W. B. [19U2] (Sec. 26) '•Summary of propellants in use in British service,” Adv. Council Sci. Res. Tech. Bov, to British Ministry of Sunply, A. C. 1802 (Mar. 7, 19U2). Macnab/U, and E. Ristori [189h] (Sec. 3h) "Researches on modern explosives,” Proc. Roy. Soc. (London) 56, 8-19 (189U). Macnab, W. and E. Ristori [1900] (Sec. 36) "Researches on modern explosives; second communication," Proc. Roy. Soc. (London) 66, 221-32 (1900).
- уз - Mallard., Е. and Н. Le Chatelier [1883] (Sec. 38) Constants for equations for temperature," Ann. Mines Ц, 379 (1883). Marre, Jacob de [1886] (Sec. 2) '‘Perforation des tiles et plaques de fer et d’acier en tir normal/1 Mem. artillerie francaise 1h (1886). Marshall, Arthur [191 7] (Sec. 25, 26) Explosives, London: Churchill; Philadelphia: Blakiston (ed. 2, 1917) , vol. I and II, 795 p., vol. Ill, 286 p. Miller, H. W. . [1920] (Sec. 18, 57) "Preliminary report on character, cause, rate and effect of the various phenomena of the wear and erosion of guns,1’ manuscript, Ordnance Dept. Library (1920), 180 p. Moesveld, A. L. Th. [1928] (Sec. Ц2) '’La temperature de combustion de la poudre sans fumee en fonc- tion de sa composition chimique. I. Partie theorique" [See de Bruin and de Pauw, 1928, fir Part II.], Communique de la Ste. Anme, "Fabrique Neerlandaises d’Explosifs1’ No. 7 (1928). Muraour, H, [1919a] (Sec. 39) ['’Determination of the temperature reached in explosive reactions'1], Gomptes Rendus 168, 995-7 (1919); Ch. Abs. 13, 1766. Muraour, H. [1919b] (Sec. hO) , ["Comparison of temperatures of explosion calculated from specific heats with those calculated from explosion pressures1’], Comptes Rendus 169, 723-6 (1919); Ch. Abs. 1U, 631. Muranir, H. [1923] (Sec. 37) "Note sur 1*influence du refroidissement dans la mesure des pressions explosives," Mem. poudres 20, 323~3h (1923), tr. by Major Aiken Simons, .Ordnance Dept. Library. Muraour, H. [192Ца] (Sec, ЦО, Ц2 ) "Notes sur la theorie des explisifs," Mem. artillerie francaise 3, 329-338 (192Ц). & Muraour, H. [192ЦЬ] (Sec. Ц0) "Comparison entre les temperatures d’explosions calculees a partir des pressions explosives et les temperatures d’explo- sions calculees a partir des chaleur specifiques," Mem. artillerie francaise 3, 3h9-63 (192Ц),
- 96 - Muraour, JI. [192Цс] (Sec. ЦО) "Note sur la chaleur specifique des’gaz h haute temperature. (Comparison des experiences de Berthelot et’Vieille avec les experiences effectuoes au Laboratoire de Nerhst)," Mem. ar- tillerie fraricaise 3, 373-80 (192Ц). Muraour, H. [1925] (Sec. 37j 52) "Notes de balistique interieure. II. Sur 1*influence du refroidissement dans la mesure des pressions explosives. III. Sur ,1a perte par le refroidissement dans le tir au canon," Mem. artillerie francaise'4, Цб0-7Ь, Ц75-8-(-1925). Muraour, H. [1926] (Sec. Ц0) ["?i comparison between explosion temperatures calculated from the specific heats and those calculated from-the explosion temperatures"], Bull. Soc. chim. 39, 389-96 (1926); Gh. Abs. 20, 2Ц13. . Muraour, H. [1931 ] (Sec. 30) ' "Sur une theorie de la combustion, en vas clqs> des poudres colloidales," Comptes Rendus 192, 227-9. (1931). Muraour, H. and G. Aunjs [1938] (Sec. 29) "Verification de la loi de combustion par couches paralleles des poudre colloidales," Comptes Rendus 206, 1723-6 (1938). Naval Ordnance [1936] (Sec. 15) "Specifications for gun forgings for the United States Army and Navy," Naval:Ordnance Pamphlet No. 9, WashingtonJ Govern- ment •Printing Office (5th rev., May 1936). ...' Naval Ordnance [1939] (Sec. 3, Ю, 12; 15, 23, U3) Naval ordnance* a textbook prepared for the use of midshipmen at the' U.'S. Naval Academy, by,Officers of the U.S. Navy, Annapolis: U.-S. .Naval Institute (1939), Ц81 p. Noble, Л. [189Ц] (Sec. 3U) "On methods that have, been adopted for measuring pressures in the bore of guns," Oxford: British Assoc, (ij&9U); reprinted in Noble (1906 reprint], pp. Ц82-98. Noble, A. [1905] (Sec. 32, 3h, 3?) 4Researches on explosives, III," Trans. Roy. Soc. (London). ‘ 205л> 201-36 (1905). Noble, A, [1906] (Sec. 32, 3U, 35,. hO) "Researches on explosives',. IV," Trans. Roy. Socv (Londun)' 206A, L53 (1906). ,'
- 97 Noble, л. [1906 reprint] (Sec. 36) Artillery and explosives; essays and lectures written and delivered at various times, New York: Dutton (1906), 5118 p« Noble A. and F. Abel [187$] (Sec. 33, 36, 37, 118) ’’Researches on explosives, I,” Trans. Roy. Soc. (London) (1875); reprinted in Noble [1906 reprint], pp. 99-230. °- в* p- [19M] (Sec. 3) Ordnance Board (Great Britain), Proceedings No. 11379 (Mar. 13, 19111), ”High-velocity guns on the Gerlich and sm jothbore principles.11 Crd. Dept. [1922a] (Sec. Ц.6) ’’Report of 21i0-mm howitzer tests. Firing record No. 191$, referred to in memorandum dated Nov. 28, 1922 by A. Adelman,” Technical Staff Test Program 1922-101. Ord. Dept. [1922b] (Sec. Ц6) ’’Report of experimental determination of the frictional resis- tance offered by the rotating bands of projectiles vriien passing through 75-mm guns,” Technical Staff Test Program 1922-108. Ord. Dept. [1921;] (Sec. 53) ’’Heating of rifle barrel due to friction of rotating bands,” first progress report of Ordnance Program 111179 (Apr. 9, 1921г). Ord. Dept. [1931] (Sec. 58) ’’Report of investigation of 2Ц.0-ПШ1 howitzers 111918 and M1918A1. Investigation of causes of bursting of howitzer No. 5 (Dec. 3, 1931); TSTP 1932-657. Ord. Dept. [19110a] (Sec. 6) ’’Ordnance Materiel — General,” Aberdeen P. G., the Ordnance School (Vol. I, Apr. 19Ц0; Vol. II, May 19Ц0). Ord. Dept. [ 19110b] (Sec. 15) ’’Notes on the selection and use of metals in ordnance design,” p. 86; Ordnance Document No. 2050 (May 19110). Pike, H. H. H. [19112] (Sec. 27, 32) ’’Thermochemical data for the products of propellant explosions,” Adv. Council Sci. Res... Tech. Dev. to British Ministry of Supply, a. C. 1862 (Mar. 20, 19112). Piobert, G. [ 1839] (Sec. 29) Traite d’Artillcris, Proprieties et Sffets de la Poudre (1839).
. - 98 - Poppenberg, 0. [1909] (Sec. ho) [’’Calculating the temperature of explosion ..in explosives”], Z. ges. Schiess u. Sprengstoffw. 4, 36 7 (T9O9); Ch. Abs. 4, 16?U Poppenberg, 0. [1923] (Sec. 39, hl) "Uber die Zersetzung von Explosivetoffen,” Z. angew. Chen. 36, 80-8$ (1923); Ch. Abs. 17, 2361i. Poppenberg, 0. [1924] ”Sur la decomposition des substances explosives,” Men. artillerie francaise 3, 383-402 (1924). (Tr. of Poppenberg,0. [1923].) Poppenberg, 0. and E. Stephan [1909a] (Sec. 39) ["Decomposition, heat of combustion and temperature of explosion of explosives”], Z. ges. Schiess u. Sprengstoffw. 4, 281-303 (1909); Ch. Abs. 4, 1675- Poppenberg, 0. and E. Stephan [1909b] (Sec. 35) ["Decomposition of povjders in guns during firing”], Z. ges. Schiess u. Sprengstoffw. 4, 388 (1909); Ch.- Abs. 4, 1674. Poppenberg, 0. and E. Stephan [1910a] (Sec. 39) ["The decomposition of explosives”], Z. ges. Schiess u. Spreng- stoffw. 5, 266-91 (1910); Ch. Abs. 4, 3139. Poppenberg, 0. and E. Stephan [1910b] (Sec, 39) ("The decomposition of explosives”], Z. ges. Schiess u. Spreng- stoffw. 5, 452, 474 (1910); Ch. Abs. 5, 1$15. Sarrau, E. [1893] (Sec. 37) Introduction a la theorie des explosifs, Paris (1893). Sarrau, E. and P. Vieille [1884] (Sec. 34) Mem. poudres et salpetres 2, 126, 337 (1884-5). Schmidt, Albert [1 929] (Sec. 40) "Beitrage zur thermodynamischen Behandlung oxplosibler Vorgange,” Z. ges. Schiess u. Sprengstoffw., parts I and.II;.24, 41-46, 90-93 (1929); part III, 25, 97-101, 144-48 (1930),“" Schmidt, Albert [1934] (Sec. 32) "Therraochenische Tabellen fur die Explosivchemie,” Z. ges.' Schiess u. Sprengstoffw. 29, 2$9-66, 296-301 (1934),
- 99 - Siwy, P. [1906] (Sec. 35) ’ ' '‘Die Abnutzung der C-eschtltze und deren Ursachen," 2. ges. Schiess u. Sprengstoffw. 3, h2., 66 (1908); Ch. Abs. 2, 2862. Smith, N. M., Jr. [19h2] (Sec. U6) The piezoelectric projectile accelerometer and a bore- friction gage, NDRC Rept. A-59 (OSRD No. 808) (Aug. 7, 19Ц2). Storm, C. G. [1925] (Sec. 25) "’The development of flashless, nonhygroscopic powders,1’ manuscript, Ordnance Dept. Library (Feb. 18, 1925), 16 p. Tolch, N. A. [1935] (Sec. 51) ’’The rate of erosion of the Browning caliber .30 tank type machine gun Ml 919 as dependent on the temperature of the gun and other factors,11 Aberdeen P. G., Ballistic Res. Lab. Rept. No. 13; 21 p. and graphs. Tolch, N. A. [1936a] (Sec. 51. 5h, 55) "On heating effects and endurance properties of the 105-ram antiaircraft gun Ml as determined by rapid-fire tests,11 Aberdeen P. G., Ballistic Re's. Lab. Rept. No. 35 (Feb. 12, 1936), 16 p. Tolch, N. A. [1936b] (Sec. 50, 51, 52) "On heating effects and endurance properties of the 3_in. antiaircraft gun IQ as determined by rapid-fire tests," Aberdeen P. G., Ballistic Res. Lab. Rept. No, 39 (Mar. 2, 1936), 19 p. Tolch, N. a. [1936c] (Sec. '51, 52) "Heating effects and distribution of the energy of the charge of the caliber .50 М2 machine gun, heavy barrel," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 61 (Oct. 1, 1936), 15 p. Tolch, N.,A. [1937] (Sec. 52) "Heating effects and distribution of the energy of the charge of the caliber .50 М2 machine gun, water-cooled type," Aber- deen P. G., Ballistic Res.’ Lab. Rept. No, 7U (Apr. 5, 1937), 13 ръ Tulloch, T. G. [1921] (Sec. 13, 19) Vear in big guns, Woolwichs Royal Artillery Institution Printing House (1921, written in 191?), 10 p. Valente, F. A. [1929] (Sec. 3h) "Calorific values of smokeless powder as affected by variations in composition, granulation, ignition, etc.," Picatinny Arsenal Rept. No. R72 (May 15, 1929).
- 100 - Veblen, 0., and P. L. Alger [1919] (Sec. 18) "Rotating bands,’1 J. U. S. Artillery 355-390 (1919). Volsk, N. T. and C. G. Dunkle [1932] (Sec. 3U) "Study of relation between volatile solvent content and calorific value of.FNH powders," Picatinny Arsenal Rept. No. 20$ (Apr. 20, 1932). Weaver, Warren [19Ц2] (Sec. 3) "The way nuzzle velocity affects the probability of hitting aircraft by AA fire," NDRC Memo, Section D-2 (Fire Control) (Apr. 1, I9h2), 5 p.
NATIONAL DEFENSE RESEARCH COMMITTEE REPORT NO. A-91 : PROGRESS REPORT THE EROSION OF GUNS Part Two: The Characteristics of Gun Erosion by John S. Burlew
NATIONAL DEFENSE RESEARCH COMMITTEE REPORT NO. A-91 : PROGRESS REPORT THE EROSION OF GUNS Part Two: The Characteristics of Gun Erosion by John S. Burlew Approved on October 31, 19ц 2 for submission to the Section Chairman John S. Burlew, Consultant Section A, Division A Approved on October 31, 19b2 for submission to the Division Chairman Approved on November 2,1 9Ц2 for submission to the Committee L. H. Adams, Chairman Section A, Division A Richard C. Tolman, Chairman Division A
Preface The work described in this report is pertinent to the project designated by the War Department Liaison Officer as OD-f>2, to the project designated by the Navy Department Liaison Officer as NO-23 and to Division A projects PA-230 and РА-2ЦО. Part of the work of preparation of this report was performed under Contract OEMsr-f>1 with the Carnegie Institution of Washington. This report has been issued in two volumes. The present volume comprises Part Two., on the characteristics of gun erosion. The pre- ceding volume was issued as NDRC Report A-9Q.(OSRD No. 882) and com- prised Part One, on the fundamentals of ordnance relating to gun erosion. The '’Bibliography and Author Index" on page 2I4.3 of the present volume pertains only to Part Two, but the "Subject Index" on page 2^9 pertains to both parts of the report. Distribution of copies of the report. — The editorial staff of Division A completed preparation of this report for duplication on November Ц; 19u2.The initial distribution of copies was as follows: Copies No. 1 to 2Ц, inclusive, to the Office of the Secretary of the Committee for distribution in the usual manner; Copies No. 25 to 33> inclusive to the Chief of Ordnance, War Department (Attentions Col. W. R. Gearhardt, Col. S. B. Ritchie, Col. R. R. Studler (2 copies), Lt. Col. D. J. Martin, Capt. С. E. Balleisen, S. Feltman, P. M. Netzer, Technical Library); Copies No. 3U to 36, inclusive, to Aberdeen Proving Ground (Attention: Lt. Col. L. E. Simon, R. H. Kent, J. R. Lane); Copies No. 37 to Ц0, inclusive, to Watertown Arsenal (Attention: Col. H. H. Zornig (2 copies), P. R. Kosting, H. H. Lester); Copy No. I4/I to Watervliet Arsenal (Attention: Lt. Col. S. L. Conner); Copy No. Ц2 to Frankford Arsenal (Attention: Lt. Col. L. S. Fletcher); Copy No. ЦЗ to Springfield Armory (Attention: Capt. A. E, Benson); Copies No. ЦЦ to I46, inclusive to the Chief of the Bureau of Ordnance, Navy Department (Attention: Capt. G. L. Schuyler, Lt. B. D. Mills, С. B. Green); Copy Цр. h7 to the Office of the Coordinator of Research, Navy Department (Attention: Lt. T. C. Wilson); Preceding Page15 Blank iii
Copy No. Ц8 to the Naval Research Laboratory (Attention: R. Gunn); Copy No. Ц? to the Inspector of Ordnance in charge, Naval Proving Ground, Dahlgren, Va.; Copy No.-£0 to С. C. Lauritsen, Vice Chairman, Division A; Copies No, 51 to 60, inclusive, to L. H. Adams, Chairman, Section A, Division A; Copy No. 61 to H. B. Allen, Vice Chairman, Section A, Division A; Copy No. 62 to L. J. Briggs, Member, Section A, Division A; Copies No. 63 and 6J4 to E. L. Rose, Member, Section A, Division A; Copy No. 65 to E. R. Weidlein, Member, Section A, Division A; * Copy No. 66 to J.- S. Burlew, Consultant, Section A, Division A; Copy No. 6? to H. L. Curtis, Consultant, Section A, Division A 5 Copy No. 68 to R; E. Gibson, Consultant, Section A, Division A; Vice Chairman, Section H, Division A; Copy No. 6? to W. F. Jackson, Consultant, Section A, Divi- sion A; Copy No. 70 to J. F. Kincaid, Consilltant, Section A, Divi- sion A; Copy No. 71 to T. C. Poulter, Consultant, Section A, Divi- sion A; Copy No. 72 to J. Wulff, Consultant, Section A, Division A; Copy No. 73 to J. 0. Hirschfelder, Consultant, Section A, Division A; Copy No. 7U tO'P. W. Bridgman, Consultant, Section B, Division, A; Copy No., 75 to G. B. Kistiakowsky, Chairman, Section B-1, Division B; Copies No. 76 to 80, inclusive, to the Secretary of the Com- mittee for transmittal to L. C. Pauling, Clyde Williams, J. C. W. Frazer,'I.'If. Stein, J. P. Magbs; Copy No. 8l to M. A. Tuve, Chairman, Section T, Office of Scientific Research and Development; Copy fob. 82 to Lt. Col. K. Maertens, Secretary to Joint Com- mittee on New Weapons and Equipment; Copy No. 83 to the Bureau of Ordnance (Attentions Lt. H. W. Drake); Copy No. 8Ц to Frankford Arsenal (Attention: The Librarian); Copy No. 85 to the Chief of Ordnance, Firing Records, Tech- nical Division (Attention: Mrs. E. T. Ryan). iv
TABLE OF CONTENTS PART ONE* FUNDAMENTALS OF ORDNANCE RELATING TO GUN EROSION Page Foreword by L. H. Adams .................................... ix Chapter I. Introduction ......................................... 1 II. Guns .......................................... 7 III. Projectiles ......................................... 26 IV. Propellants ........................................ 33 . V. Burning of the Powder ............................. Ц2 VI. Movement of the Projectile in'the Bore ............. 6? VII. Reaction of the Tube to Firing ..................... 79 Bibliography and Author Index for Part One .................. 89 PART TNG THE CHARACTERISTICS OF GUN EROSION VIII. Magnitude of Erosion .............................. 101 IX. The Eroded Surface ................................. 119 X. Experimental Methods of Study of Gun Erosion ...... 1J?2 XI. Relation Between Properties of the Gun Metal and Erosion................................................... 168 XII. Conditions of Firing that Influence the Rate of Erosion........................................... 192 XIII. Effect on Erosion of Leakage of Gas Past the Projectile................................................ 209 XIV. Life of Guns ................... ................. 2Г9 XV. Theories of the Mechanism of Erosion........... 235 Bibliography and Author Index for Part Two................... 2^3 Subject Index to Parts One and Two ......................... 2^9 Supplementary Index of Guns, by Caliber..................... 267 z<Part One appears in a previous volume which is designated as NDRC Report A-90 (OSRD No. 882). v
List of Tables Table /- Page I. Principal dimensions and other details of repre- sentative guns and their projectiles .................... 8 II. Compositions of steels used in Navy guns .......... 20 III. Compositions of steels used in Army guns .......... 22 IV. Compositions of steels used in small arms barrels .23 V. Approximate proportions of artillery projectiles .. 27 VI. Nominal compositions of smokeless powders .......... 38 VII. Heat of explosion,, heat of combustion and tempera- ture of explosion df propellent powders .................. 18 VIII. Products of combustion of gunpowder..........,..... 19 IX. Products of combustion of smokeless powders ........ 50 X,- Variation in composition of powder gases with change of density of loading ..................... 52 XI. Products of combustion of cordite in a gun com- pared with those obtained in a closed chamber .... 55 .. XII. Partial pressures of the constituents of powder gases, and temperature of explosion......................... 66 XIII. . Heating of gun tubes firing single-base powder .... 80 XIV. Temperature of the breech end of the gun tube as a function of the rate of fire; single-base powder., 82 XV. Distribution of energy of powder .................. 83 XVI. Erosion of 155-mm guns ....................... 108 XVII. Comparison of average erosion per round for guns of different caliber; velocity, 2500 to 2750 ft/sec .............................................. 109 XVIII. Average thickness of the four parts of the altered layer 'at the origin of rifling in 37~mm gun No. СЦ6С....................................... 131 XIX. Hardness of altered layer at surface of bore of 12-in. gun specimens ............................... 138 vi
Table Page XX. The microstructural constituents of steel .......... 1Ц2 XXI. Cost of ammunition needed for firing guns to destruction ............................................. 155 XXII. Properties of special steels tested for drowning machine-gun barrels ...................................... 170 XXIII. Firing tests of Browning machine-gun barrels made of special steels ....................................... 173 XXIV. Erosion of pure metals as a function of melting point .................................................... 179 XXV. Loss of weight of erosion vent plugs ............... 181,182 XXVI. Accuracy of life of machine-gun barrels fired at different temperatures ................................... 19h XXVII. Comparison of life of guns and muzzle velocity .... 196 XXVIII. Calculated accuracy lives of guns .............. 23h List of Figures Figure Page 1. Typical pressure-time curves .....................« 71 2. Erosion of 3-in. gun liner ..................... 10Ц 3. Erosion of 3-in. gun after 108Ц rounds ........ 10£ Ц. Erosion as a function of the number of rounds .... 106 9. Section of the bore surface at the origin of rifling of a 37-mm Browning automatic gun M1 after 8032 rounds ........................ 121 6. Interior voids, uneven bore edge and cracks in 37-nim gun СЦЦбС ................................. 12£ 7. Hard layer in 3-in. AA gun.No. 138 ................. 127 vii
THE EROSION OF GUNS Part Two: The Characteristics of Gun Erosion CHAPTER VIII. MAGNITUDE OF EROSION 59- The two chief characteristics of erosion are its extent and the nature of the eroded surface. The exact dimensions of the bore of a particular gun at any given time depend on so many factors, some of them uncontrollable, that each gun is a separate problem. On the whole, however, the erosion of all guns — except those fired extremely rapidly — follows the same pattern, so that it is possible to des- cribe these two characteristics in general terms. 60. Variation along the bore The increase in diameter of the bore of an eroded gun is not uniform along its length, but is highly localized at the two ends. In the middle of the bore, between the two eroded regions, the diameter frequently is less than it had been originally, because of fouling or coppering, which is the formation of a metallic deposit on the surface of the bore (Sec. 67) • The greatest amount of erosion occurs at the breech end, where a region of considerable enlargement of the bore extends forward from the origin of rifling (Sec. 11 ) for a distance of several calibers, the maximum enlargement being at a small fraction of a caliber in front of the origin, namely, at the point where the lands had first attained their full height when the gun was new. The erosion of the lands is usually two or three times as great as that of the
- 102 - grooves,—' as is illustrated for a 3-in. М3 gun liner No, 881 in Figs. 2 and 3- The former is a combination of five star-gage records made at different periods in the life of the gun. the latter is.a cross-sectional diagram with an exaggerated scale. Only erosion at the breech end is apparent because that at the muzzle was so slight in comparison. In a case gun (Sec. 21) erosion of the centering cylinder extends toward the chamber from the origin of rifling only as far as the mouth of the cartridge case. The increase in diameter in the middle of this region is usually a little greater than that of the grooves at the origin. In a bag gun (Sec. 21), on the other hand, the increase in diameter of the centering cylinder becomes progressively less to the rear of the forcing cone, and is negligible after a distance of one to two calibers. The chamber does not erode. Muzzle erosion is always much less than breech erosion, and in some small-caliber guns it is scarcely noticeable until the end of life. The maximum increase of diameter in this region occurs at the muzzle itself. Here the Lands usually wear more rapidly than the grooves. Erosion usually is not circularly symmetrical. However, the dissymmetry is frequent.ly not reported, for one of the two common types of star gage (Sec. 1Ц) gives only an average diameter, based Ц8/ Dr. P. R.- Kosting (personal communication, Aug. 22, 19U2) points out, however, that in a recent endurance test of the 37-mm gun M1A2.,(see Sec. 107), it had been observed that when double-base powder was used as the prope.llant the increase of diameter at the origin of rifling was greater across the grooves (0.035 in.) than across the lands (0.030 in.) after slightly more than 1200 rounds.
- юз - on lengths of three radii 120° apart. Some writers have suggested that erosion of the muzzle is regularly greater at certain positions around the circumference of the tube (referred to as different '^•clocks'* in analogy to the positions on a clock face), but such a generality does not seem to apply to all guns. In three well-worn 155-nnn guns, for instance', in which both vertical and horizontal diameters were measured, the difference between the twb diameters across the lands at different positions along the last three caliber lengths of the gun varied from 0 to 0.03 in., the horizontal diameter being the greater in two of the guns and the vertical the greater in the other [Weyher, 1939]. 6l. Variation with number of rounds The average extent of erosion is frequently measured by the increase in diameter of the lands at the origin of rifling or just in front of it. When this increase — denoted usually by E and ex- pressed in inches — is plotted as a function of the number of rounds, the typical curve which is obtained for cannon fired slowly shows that the rate of increase in diameter of the lands at the origin be- comes less during the later stages of erosion. This relation is illustrated by Curves A and В in Fig. Curve A which is the standard curve for an unplated 1li-in./50-caliber Navy gun, has been expressed by the exponential function, E = O.3U9 ( 1 - е-0-ОЮО), (VIII-l) 1l9 / where N is the number of - equivalent service rounds. — . Curve В 119/ Naval Proving Ground photograph No. 8302.
2ООг /го г u । e I <ь l Ал is 'l к /оо /во - О /60 <30 /40 60 /20 - 40 О юо - 20 &О О 60 - •С Erosion of lands 8} 1 <b i <u i /034 4o - 03!^ FIG. 2. EZOBIOK OF~ B~IN. GUN L/NE&. бгооие d/'ometer J.oSo in. (when necu) ZO 40 60 SO /ОО /20 /40 Distance from rear face of breech (mJ Bore d/ameter z^g,2a 5ooo/n.(<ohenneaj) • /оо /го /40 3-in. A A Gun /tfB, Liner No. 361 Chamber, о to 235 in. Forcing Cone, 235 to 24.5 7n. Bore, 24.5 to /50.57m. i О * Advance of Ercing Cone (in J Rounds Vertical t-toriz. 229 ( not measured ) 292 0.4/ 0.36 5Z7 Z.34 2./3 <33/ &.O4- 7.ZZ /034 11.4е/ //02
EROSION OF LANDS I.SOO'R 3^" EROSION OF GROOVES З.ооо1' !.540“R -S-O! £7g.<3. Erosion of-3-/П.gun offer /084 rounds. 7У1& verfic/e sco/e is five f/'тез fhe hor/jon/a/ ^sca/e.
— / OG ~
- Ю7 - represents data for sixty-tiro 133-nim guns, G.P.F., in different stages of wear [Weyher, 1939]. For a number of 3-in. antiaircraft guns, М3, which had been fired more rapidly, however, the rate of erosion did not decrease with ^0 / time. Curve C-—' in Fig. Ц , which shows the increase in diameter of these guns at the origin of rifling, is slightly concave upward. Curves A, В and C in Fig. Ц extend to the estimated accuracy life of the guns considered (see Chap. XIV). Some European guns that have been fired far beyond this point have shown a subsequent increase in the rate of erosion, the complete curve in such cases being S-shaped.— 32/ A similar condition has recently been observed— in the case of a 37-mrn AA gun M1A2, which was fired mostly rapid fire (3 or 6 rnd/min). From round 3000 to round 6800 there was scarcely any change in the pre- vious increase (0.022 in.) in the diameter of the lands; and then suddenly the diameter increased 0.0Ц8 in. during the next 1000 rounds. These data on the variation of erosion with the number of rounds of the l33-mm guns depend on measurements on different guns, instead of on a series of measurements at different stages in the life of the same gun. Hence they may reflect individual peculiarities of the guns. 30/ This curve has been drawn on the basis of star-gage records in the files of the Ballistics Section, Ordnance Department, U.S. Army. 3l/ P. R. Kosting, personal communication, Aug. 13, 19Ц1 32/ Watertown Arsenal photograph No. W. А. 639-ЦО76, based on Firing Record No. 8097? P. R. Kosting, personal communication, Aug. 22, 19Ц2.
- 108 - The three guns,that had been fired the largest number of rounds were of French manufacture and of unknown characteristics; the other five had been made in the. United States at the close of World War I. One of the latter group showed an excessively high rate of erosion, but it was suggested that this might have been an erroneous indica- tion caused by an error in the number (112$) of rounds recorded* It has been noticed by the present writer that the points (solid dots) in Fig. 1* representing the increase in diameter of the bore of these guns (see fourth column of Table XVI) do not fit at all on Curve B. Instead, it has been found that they fit reason- ably 1 well on a parabolic curve with a slight upward curvature. There remains the more likely possibility that the data for the two dif- ferent makes of gun should be considered to lie on two separate table XVI. Erosion of 155-mm guns. Gun No. Number of Velocity Eb Fc Loss of Weight (gm) Rounds (ft/sec) (in.) (in.) Total Per Round 81*2 ' 812 2398 0.037 0.33 1*52 0.56 56? 1609 2383 .062 • 70 590 .1*1* 853 212? 2379 .088 1.06 1023 .U9 .. 1*31* • . . 11.2^ • 2365 . .110 2.62 1761 1.57 li 9* 2853 2283 .11*3 5.1*7 2889 1.01 UU* 3289 221*7 . 156 7.33 2725 . 0.83 32* 3597 , 2215 ' .175 8.01* 1*006 1.11 aThe -gun numbers marked with an asterisk indicate those Of French manufacture (Puteaux). The others were American made. bE, increase in diameter at origin of rifling. CF, advance of forcing cone (see Sec. 125d). dIn the original report [Weyher, 1939] it is suggested that the number of rounds reported for this gun may be too small. On the other hand, the steel in this gun may have been of poor quality; for, on the graph showing the erosion of 62 155-mm guns as a function of the number of rounds, several other guns showed an equally high rate of erosion.
- 109 - curves, each of the general shape of Curve B. This possibility does not seem to have been considered in the original investigation, for the report [Weyher, 1939] correlates the various measurements as if the guns were all the same except for the number of rounds fired. 62. Variation with caliber The depth of erosion per round and also the maximum depth of erosion at the end of life increase very greatly with increase of . caliber. Representative values of the former quantity for several guns of about the same muzzle velocity but different calibers are given in Table XVII. It should be remembered that the variation in the rate Table XVII. Comparison of average erosion per round for guns of different caliber; velocity, 2500 to 2?5O ft/sec7 Gun Rounds Erosion at Origin of Rifling Lands(10-3in.) Grooves(10-3in.) 37-mm М3 Tube No. 22708 902 to Ш72 0.0Ц7 0.026 3~in. AA Gun М3 No.6Ц0 326 to 528 .17 .12 8-in. Ml888 Mil No. 56 62 to 127 .35 .27 1 Ц-in. M1920 Mil No. 11 123 to 15Ц 1.1 . 96 16-in. M1919 Mil No. 2 123 to 152 2.8 1.Ц of erosion of guns of the same caliber and muzzle velocity is quite large, and that therefore the values given in Table XVII should not be considered as establishing a functional relation between erosion and caliber. They are merely illustrative of possible values.
- 110 - 63» Advance of the forcing cone— The forcing cone (Sec^TO) does not retain its original posi- tion, but gradually moves forward as erosion proceeds. At the same time the angle of the cone becomes less steep, because of rounding off of the front, of the conical surface. The advance of the forcing cone, which is one way of expressing the extent of erosion, is measured with a special tapered plug gage— that is inserted through the breech until it is stopped by contact with the forcing cone. The i . fifth column of Table XVI lists the advance F of the forcing cone for some 155-чпт. guns. 6Ц. Weight of metal lost The weight of metal lost during the erosion of a gun is not readily determined. It is not directly proportional to the increase in diameter at the origin of rifling, because the length of the eroded region increases toward the muzzle at the same time that the diameter at the origin is increasing. On the other hand, this forward extension is slower than the increase of diameter, so that the loss of weight is proportional, not to the square of the in- crease of diameter, but presumably to some power less than two, 53/ The use of the advance of the forcing cone as a criterion of the end of life is discussed in Sec. 125(d). 5Ц/ A plug gage that made contact at only two diametrically opposite points in. the bore was recommended by Weyher [19391j but subsequent experience at Aberdeen Proving Ground showed that if this gage were inserted in.the bore at a slight angle, it could be pushed beyond the forcing cone. Hence its use for routine measurements has been discontinued. (Arthur E. Jewell, Instruments Section, Aberdeen Proving Ground, personal communication, Sept. 2, 19Ц2.)
- 111 An estimate of the weight of metal lost from the breech end of a 155-mm gun tube — that at the muzzle is insignificant -- has been made by the present writer on the basis of the star-gage records of the guns -listed in Table XVI. The area— under the erosion curve of the star-gage measurement of the diameter plotted as a function of distance along the bore was computed separately for the lands and for the grooves. The space formerly occupied by. the material eroded away from the lands was considered as being repre- sented by half the volume of the solid of revolution obtained by rotation of the area under the erosion curve about the axis of the bore; and similarly for the grooves. Hence the total volume Vj of material lost is, according to the theorem of Pappus, VT = VL + VG = (DA)L + in (DA)g, (VIII-2) where D is the mean diameter of the section resolved, A is its 56/ area,— and the subscripts L and G refer to lands and grooves, 55/ De Sveshnikoff [1933, p. 659] had measured the area under the erosion curve of caliber .30 machine-gun barrels with a plani- meter. He remarked that "aside from giving a general idea of the dimensional change of the bore at the breech and the muzzle end, the measurements do not appear to yield information of any value." 56/ The grooves are regularly wider than the lands, sometimes as much as 25 percent (cf. Table I), Hence the following equation is more formally correct than Eq. (VIII-2): VT = Г77 n (DA)l + I ГТ?) " ^G' (VIII-3) where L and G represent the width of the lands and of the grooves, respectively. For L/(L+G) = 0.Ц, Eq. (VIII-2) gives a result 10 per- cent greater than that obtained from Eq. (VIII-3)j but this dif- ference is scarcely as great as the uncertainties inherent in the calculation, since no provision can be made for local irregularities of the bore surface between the points touched by the star gage.
- Ir- respectively. The corresponding weight of material lost wascom- puted by the use of a nominal factor of 7«3 gm/cm3 for the density of steel. The total weight of metal lost and also the average loss per round are recorded in Table XVI. It will be seen that the guns (neglecting No. h3h) may be separated into two groups on the basis of the rate of the loss of weight. Interpretation of this result is complicated by the fact that the two groups differed in two respects, -namely, in the number of rounds fired and in manufacture. It might be concluded that the rate of erosion, as measured by the loss of weight, suddenly increased above about 2000 rounds; on the other hand, it seems somewhat more likely that the two kinds of gun differed in erosibility (Sec. 61). A definite choice between these two possibilities cannot be made on the basis of the data available. What is needed are measurements of the same gun at different stages of its life. It was stated by Siwy [1908, p. 67] that a 28-cm gun lost dbdut 1/3 kg of steel at each round, but he made no mention of the basis “of this estimate. It seems to be fairly reliable, as indi- J cated by a calculation made by the writer of the loss of weight of 57/ a 12-in. (30.£-cm)/L5-caliber Mk. V gun.—' On the assumption that the lands occupied percent of the circumference of the bore, the weight lost from the lands alone was Ц10 gm/round. The fact that $7/ The calculation was based on the erosion curve of the lands of this gun, published as Sketch No, 1Ц in Naval Ord. [1912],
- 113 - this gun -was somewhat larger than that referred to by Siwy gives ample allowance for the weight lost by erosion of the grooves, which would have been less than that from the lands. It is important to notice that the rate of removal of metal is several hundred times as great for this size of gun as for the l55-mm gun already referred to, the diameter of which is half that of the 12-in. gun. 65. Scoring VJhen the surface of the bore of a gun tube has been eroded into relatively deep troughs, it is said to have been scored. This condi- tion may result from local imperfections in the steel or, more usually, it is simply the final stage in erosion that at first was fairly uniform. The latest report by Kosting and Peterson [19b2] concerning eroded guns points out that the scoring in the 37-mm gun M1A2 which they had examined occurred in a region where hot gases could begin to expand freely after escaping past the projectile before it started to move forward. This observation seems to confirm the opinion held in different quarters, which was summarized by Greaves, Abram and Rees [1929, p. 118] in these words: "It seems probable, therefore, that scoring begins (if it occurs at all) after some general enlargement of the bore has already taken place, and is due, at least in part, to gas-washing produced by the escape of propellent gases past the driving band during the initial stages of motion of the projectile."
- 11Ц - 66. Pastilles Pastilles are small shallow depressions in the bore of a gun, ranging in depth from 0.02 to 0.0£ in., and having varying diameters. They are of two types — bulbous and granular. They were found only in 7J>-mm guns and in a few l££-mm howitzers, used by both the French and the A.E.F. in World War I. Scarcely any were reported by the. British or Germans. None were found in antiaircraft guns [Miller, 1920, pp. 61 -6I4]. A graph [Miller, 1920, Fig. 20] of the location of £26 pastilles found in 223 7£-™i guns along the bore of the gun showed that the maximum frequency of occurrence was at a distance of 67. £ in. from the muzzle,—/ which was near the point of maximum pressure in the gun. Nearly all the pastilles occurred in the bottom quadrant of the bore. ( Two possible causes of pastilles have been proposed. Some French ordnance inspectors in World War I suggested that unburned fragments of US No. 3 powder (multiperforated grain), left in the bore of the gun, detonated by impact of the next shell. Soldiers operating 7£-mm guns in the field, however, suggested that the tin- lead anticoppering compound formed a globule of bronze by fusion with copper and then was pressed into the bore by the next projec- tile [Miller, 1920, pp. 67, 68, 71]. Neither opinion has been veri- fied. £8/ The length of rifling was 87.8 in.
- 115 - Pastilles have been alleged as a cause of prematures. Hence they have a great effect on the morale of a gun crew, in some in- stances causing the crew to destroy a gun which had them. They are also supposed to reduce muzzle velocity by allowing escape of gas, and to serve as a starting point for further erosion [Miller, 1920, pp. 72-7Ы- However, this latter effect seems very doubtful in view of the findings of Kosting [1936]. As a result'of a compilation of all essential data on pastilles preparatory to initiating a metallurgical investigation of them, Kosting found that "the dimen- sions of the pastilles do not change within the accuracy of measure- ments (1/6Ц in. approximately) during subsequent firing." For example, the pastilles in four 153-mm howitzers examined after 23, 177j 307 and Ц11 rounds subsequent to the’original measurement of the pastille, showed no enlargement. Hence, pastilles probably " can be safely neglected-as far as erosion is concerned. 67. Coppering, or metal fouling The metallic deposit that forms on the surface of the bore of a gun, usually referred to as copper, is generally considered to consist principally of metal rubbed off the rotating band. Thus Tolch [1936a] reported that the copper in a ЮЗ-mm gun consisted of a mixture of copper and iron oxides in a matrix of free copper. The analysis showed: total copper, 71.Ц2 percent; total iron, 9.6£ per- cent; lead, 0.33 percent; and oxygen combined as CuO and FeO, 18.61 per- cent. The presence of copper is more extensive than a superficial examination might.indicate, for its surface is frequently blackened
- 116 - by copper oxide. Heating a section of the bore in a current of hydrogen reduces the oxide, and then the true extent of the copper- ing is apparent. Coppering is greatest in the central section of the bore, where it may decrease the diameter as much as 0.01 in. In a given gun it is usually thicker in the grooves than on the lands. It sometimes occurs on the surface of the eroded section near the breech. Inasmuch as a deposit in this region is likely to be re- moved by later firing, it may be the cause of irregularities in the apparent erosion at that point. The principal ill effect of coppering is that the movement of the projectile is slightly less regular. The corresponding irregu- larity in the rate of burning affects the muzzle velocity of the projectile and hence its range. 68. Effect of fouling on erosion The possible influence of fouling of the bore on erosion was suggested by Hardcastle [190Ц], who had observed that cupronickel from the jacket, of the bullet worked into cracks in a caliber . 303 rifle barrel. Its paefcence was shown by immersing a polished section -of a barrel in ammonia for a week. Hardcastle claimed that erosion was partially, caused by alloying of the steel with the cupro- nickel. The. mechanical enlargement of cracks by copper from the rotating band was considered by Howe [1918] to be an important cauqe of erosion (see Sec. ?0)«
- 117 - It has also been suggested— that the zinc in;gilding-metal bands might have a deleterious effect on the steel of the bore sur- face, which might explain the observation [Ord. Dept., 1926] that * in the first trials of such bands a slight increase occurred in the erosion of a 3-in. liner, No. 35» The original report on the firing had suggested that the greater erosion might have been caused by the increased width of the band. None of these suggestions has been investigated. 69. Removal of copper Copper may be removed from a gun by several methods. Tin, lead-tin alloy or lead may be mixed with the powder, either in every round or in special decoppering rounds. Projectiles may be fitted with rings or disks made of lead-tin alloy. Copper may also be re- moved by swabbing the bore with an alkaline solution, or by scraping. The ,British have used tinfoil regularly in ammunition., but the • American services have not, except for small arms. Within the past year, however, the U.S. Navy, having found that its guns were copper- ing badly during sustained rapid fire, has begun the practice of ' 60 adding about 0.2 percent of pulverized lead tb' the powder every round.—' Coppering of U. S. 37-mm guns in service in "the Middle East has been reported by the British to be very serious, and they have requested that the American-made ammunition for these guns be supplied with tin- foil [O.B.P., 19h2b], $9/ By Col. G. F. Jenks; personal communication, Aug. 20, 19hl. 60/ Lt. B. D. Mills, personal communication, August 19Ц2.
- 118 - According to Paquelier [1929], decoppering by the use of solder is harmful, as shown by some firings of two new U7~n®i guns at GSvre in 1925. The gun that had been fired with projectiles having bands of solder had eroded more after 650 rounds than the other, which had been fired under exactly the same conditions ex- cept that no solder bands were used. In the one case the muzzle velocity decreased 13& ft/sec, whereas in the other it decreased only 79 ft/sec. No other reference to this effect has been found.
- 119 - 'CHAPTER IX. THE ERODED SURFACE ?O. .. The system of cracks During the firing of a gun the surface of the bore gradually • becomes covered with.fine cracks which are frequently referred to as "heat checks." A typical example of them is shown in Fig. 5,— which is a photograph of a section of the bore surface at the origin of, rifling of a 37-nm Browning automatic gun Ml after 8O32 rounds^ which таз about the end of its life [Kosting, 19h0]. Just how early in the life of a gun such cracks begin to develop does not seem to have been determined. The pattern of the cracks consists of small areas, nearly rectangular in outline, that are bounded by two principal series of discontinuous cracks at right angles to each other. Within each such rectangle a series of finer cracks forms a separate network. Another characteristic of the cracks is that they are puckered, so that they resemble the cracks formed in river-bottom mud when it dries out in the sun. Some of the cracks are wider and deeper than others-. On the lands — shown by the darker parts of Fig. 5 — these prominent cracks are transverse, whereas on the grooves they are longitudinal. De Sveshnikoff [1921a, p. 20; 1922, p. 306], -when he noticed this condition in eroded machine-gun barrels, remarked that these 61/ This photograph and also Figs ..-6 and 7 were obtained for this report through the courtesy of Lt. Col. S. B. Ritchie and Dr. P. R. Kosting, of Watertown Arsenal.
- 120 - directions were the same as those of the tool marks left on the lands and on the grooves, respectively. Carter [Guion, Carter and Reed, 1937] also observed that many cracks in a 1,1-in. Navy gun 62 / started at tool marks. Kosting,--' however, although he does not deny that many cracks start from tool marks, emphasizes that many more cracks occur than there are tool marks to start them. The radial depth of the cracks varies along the length of the bore, being in general greater toward the breech end. The deepest cracks of all usually occur where the driving edge of a land meets a groove. (This side of the lands is toward the top of Fig. 5.) The depth of the cracks varies considerably in different guns, but without any apparent regularity. Thus de Sveshnikoff [1933, Table II] reported that the maximum depth of cracks in 26 machine guns varied from 0.6 to 3.2 mm, whereas Fay [1916] reported that the maximum depth of cracks in a 12-in. gun he examined was 1 anm. . Fay also observed that in the region of greatest erosion in this gun — near the origin of rifling — the cracks were the deepest and also the fewest. The number of cracks per inch at the bore edge of a section of an eroded gun near the origin of rifling was reported by Kosting [ 19Ц0, Table 7] as 32J? for a worn-out 37-mm Browning automatic gun Ml and by Kosting and Peterson [ 19h2 ] as Ц01 for a worn-out 37~n®i gun M1A2. Some of these cracks, which varied in depth from a few , _ , , . _ , 63/ microns to 0.6 mm, are shown in Fig. 6. — 1 62/ P. RT Kosting, personal communication, Aug. 22, 19h2. 63/ The present Fig. 6 was Fig. 8 in Kosting [19h0]. J
Fig. 5. Section of the bore surface at the origin of rifling of a 37-nrn Browning automatic gun Ml after 8032 rounds. I I
- 123 - The radial cracks ordinarily are not deep enough to affect the strength of the tube. Howorth [1910] expressed the opinion that a split, when it does occur, is an extension of such a crack; this suggestion was confirmed by later experience at Woolwich Arsenal [Greaves, Abram and Rees, 1929] with respect to cracked carbon-steel liners. The cracks in a ih-in. gun were found by Howe [1918 j Sec. 29] to be filled with copper, which was covered with a layer of black copper oxide and dirt. This he removed by heating a section of the tube in hydrogen. He reported that a 11 careful examination of the longitudinal and transverse sections indicates that this copper reaches nearly, if not quite, to the roots of the cracks, but in many cases it does not reach to their tops, that is, it does not fill them flush, though from many cracks of the main network, it projects in the form of a thin network of dikes.” He also found that the copper color of these dikes could be revealed by rubbing them lightly with fine emery paper. The cracks are generally considered to be caused by expansion and contraction of the surface of the bore, as described, for example, by Tschernoff [1913] and by Howe.[1 918]. The heating of the surface of the bore by the powder gases is so rapid that only a very thin layer of metal becomes hot enough to be in the plastic state. The expansion of this layer is restrained to two dimensions by the rigid colder layers beneath, and hence it tends to buckle; then, after the hot gases have passed, it is quenched so rapidly by the cold metal beneath that it becomes rigid before it has time to
- 12Ц - contract to its original state. The volume changes involved in the thermal transformation of the steel exaggerate the results of normal thermal expansion. Lack of ductility imparted by the alteration of structure that takes place in the surface layers (Sec. 71) is usually assumed to be a strong contributing factor; but Greaves, Abram and Rees [1929, p. 117] pointed out that "experiments show that surface cracking of an intercrystalline character may occur under the action of hot gases in metals which nill not harden on quenching." At any rate the cracks are not limited to the altered layer. Their continuation to several times the depth of this layer was considered by Howe [1918, Sec. 31] as evidence that they had been deepened by the "wedging action" of the copper that filled them. Greaves, Abram and Rees [1929, Fig. 28] also published a photograph showing this action. 71. The altered layer Microscopic examination of the polished cross-sectional sur- face of a segment of an eroded gun tube shows that the structure of the metal has been altered for a short distance beneath the sur- face of the bore. Although the alteration can be seen in an unetched specimen, it is made more apparent by etching the surface, for instance with 2-percent nital, which is a 2-percent solution of nitric acid in ethyl alcohol. Then the altered layer, which does not etch readily, appears white in contrast :to the rest of the steel, as shown in Figs. 6(b), 7(a) and 7(b),and hence has frequently 6Ц/ The present Fig. 7 shows photomicrographs obtained by Kosting for a 3-in. antiaircraft gun similar to those that he published for a 7f?-mm gun [1939b] and a 37-™ gun [19h0].
- 125 - Fig. 6. Interior voids, uneven bore edge and cracks in 37-mm gun СЦЦбС. The specimens were copperplated prior to polishing. The bore edge is on top. (a) Un- etched, showing interior voids in the region of maximum erosion, x 2$, (b) Etched with nital, showing cracks and uneven bore edge, x 100; T, troostite bandj W, white layer; M, martensitic layer, (c) Etched with nital, x 56*0; the horizontal arrow at the Bottom indicates an incipient crack with grey outline. [Watertown Arsenal 639-2578.]
- 127 - Fig. 7. Hard layer in 3-in. AA gun No. 138. The specimen was copperplated prior to polishing. The bore edge is at the top, as indicated by the arrows, (a) Unetched, hard layer revealed by polishing, x 100. (b) Etched with nital for 8 sec, x 100. (c) Etched with nital for 18 sec, x 100. (d) Etched with nital for 60 sec, showing the white band, dark-etching layer and dark-etching band, x 100. (e) Same as (d), showing two white layers, x 1 JtOO.” (f) EtchecTwith nital for 30 min 5 the white band is~very complex in places — x 1500. [Watertown Arsenal, 639-2039.]
- 129 - been called ’’the white layer." This term seems to have been used first in a report, Tests Of metals for 1916, probably by F. 0. Langenberg of Watertown Arsenal, to designate what had been referred to earlier by Fay [1916] as the "hard layer" and by Howe [1918] as the "hardened layer." The British writers Greaves, Abram and Rees [1929] called it the "hardened skin." The illustrations of the white layer published by de Sveshnikoff [192$, Fig. 1 ], by Lester [l929a, Fig. Ц], by Greaves, Abram and Rees [1929? Fig. 1] and by Shumate [1938, Fig. 18], closely resemble Fig. 7(b). Prolonged etching of the altered layer has been shown by Kosting [1 939b, 19Ц0] to darken most of it and reveal its composite nature. The altered layer at a magnification of 100 diameters, as shown in Fig. 7 (8), consists of a narrow dark-etching layer and an outer white band, which under higher magnification (x 1$00) is seen in Fig. 7(e) to be made up of two white layers. The narrow dark-etching transition band had been observed earlier, for example by Bihlman [1921], by de Sveshnikoff [192$, Fig. 1] and by Lester [1929a, Fig. 6] and had been identified by them as troostite.—7^ The resolution of the main part of the altered layer, however, had not been made before. Hence in reading and talking about "the white layer" in a gun tube, it is important to remember that most writers before 1939 were using this term to refer to the relatively thick white layer shown in Fig. 7(b) 6$/ The term troostite in the past has not had an exact definition, but has been used to refer to several different types of microconstituent of steel. The authors cited seem to have intended - it in the sense of "secondary troostite" as defined by Epstein [1936, p. 266]. See Table XX for a comparison of this structure with re- lated ones.
- 130 - and that Kosting since that date has been using it to refer to the much thinner layers shewn in Fig. 7(e). The outer white layer is easily removed by metallographic polishing, if the surface of the bore is not first protected by plating it with CoppeiS nickel or chromium. Kosting [1939b, p. 12] remarked that therefore it may have been absent from some of the specimens examined by other investigators, because not all of them adequately protected the surface. The method of obtaining a high degree of polish of the specimen for examination at very high magni- fications has been improved by bliss Mary R. Norton [193$, 1937] at Watertown Arsenal. 72. Thickness of the altered layer The altered.layer regularly increases in thickness toward the breech end of a gun; but, except that it frequently has been reported as being thicker on the driving edge of the lands, it does not vary much in thickness around a circumference at a given dis- tance down the tube. The thickness of the altered layer in a worn-out gun increases with the caliber of the gun. Greaves, Abram and Rees [1929, Table 3] compiled a table of the approximate average thickness of the "hardened skin" for guns of different calibers studied at Woolwich Arsenal, England, that had fired cor- dite M.D. This table shows that the average depth at the origin of rifling in thousandths of an inch is almost equal numerically to the caliber in inches. A recent measurement by Jones [19Ц2] of the thickness of the altered layer of the inner "A" tube of a
- 131 16-in. gun—' showed that it varied from 0.0088 to 0.0113 in. (0.22 to 0.29 mm) in different sections of the bore, but in the report the variation in depth was not correlated with the location of the different sections. This altered layer was made up of a hard layer 0.19 to 0.2Ц mm thick plus a narrow dark-etching band O.O3I4. to O.OI4.7 mm thick. The thickness of the four parts of the altered layer — "hard layer" — were measured separately by Kosting [19I4.O] at 21 places along the bore edge of a 1-in. section cut from the origin of rifling of the eroded 37“®ni gun shown in Fig. 5- The results are summarized in Table XVIII. Another section of the bore located Table XVIII. Average thickness of the four parts of the altered layer at the origin of rifling in 37-mm gun No. .СЦЦбС.^ Layer Average Thickness (in.) (mm) First (outer) white layer 0.0003 0.008 Second w hite layer .0002 .00S Fartensite layer .0012 .030 Troostite band .000Ц .010 * After Kosting [19h0, p. 10]. 5 in. from the muzzle of the same gun showed no altered layer. This result, which has been observed also in other guns, is interpreted 66/ No. S. 15 (Vickers), fired 225 E.F.C. (see footnote 9Ц), with erosion of 0.633 in. at 1 in. from origin of rifling. The percentage composition of the steel was: C, 0.26; Si, 0.1^; Mn, O.36; S, 0.023; P, 0.03h; Ni, 3.60; Or, 0.6£.
- 132 - by Kosting^X/ to indicate "that wear at the muzzle should be dif- ferentiated from erosion at the breech end." In a subsequent examination of an eroded 37-nm gui; И1А2, No» 109, Kosting and Peterson [19Ц2] found that whereas the entire altered layer.was about 60 percent thicker than it had been in 37_mm gun No. СЦЦ6С (Table XVIII), the white layer on the average was only about Ц0 percent as thick as that in the other gun. Also in gun No. 109 the white layer was seldom complex. These two guns differed both in composition — No. СЩ6С contained only Ni as an alloying element; No. 109 contained 2.30 percent Ni, 1.07 percent Or and 0.31 percent Mo — and in conditions of firing.—/ For a number of machine-gun barrels — the number was not stated — Snair and Wood [1939, p. 609] reported that This layer extends for a distance of approximately two and one-half inches along the bore from the origin of rifling. It varies in depth along this distance, and if the depth were plotted versus the distance along the bore, the resulting curve would closely resemble a powder pressure- distance curve. This fact establishes a definite relation- ship between powder pressure developed and layer formation. This conclusion is open to question, inasmuch as the occurrence of the maximum depth of the altered layer a short distance in front of the origin of rifling might more logically be ascribed to a maximum heating effect there, just as de Sveshnikoff [1921a, p. 18] considered this to be the cause of a maximum number of cracks at that point in the machine-gun barrels which he examined. Snair and 67/ Personal communication, Aug. 22, 19Ц2. See also footnote 1. 68/ A detailed history of the firing is not available for either gun.
-133 - / Wood seem to have disregarded the fact that the maximum on the pressure-distance curve means that -- neglecting localized variations of pressure within the gas space — the maximum pressure is applied all along the barrel from the breech up to the base of the projec- tile at the point indicated and not just at that point alone. Hence I "a definite relationship between the powder pressure developed and the layer formation" would be expected to be accompanied by a depth- distance curve which was parallel to the distance axis from the breech up to the point of maximum pressure and which then decreased toward the muzzle in conformity with the pressure-distance curve. One eroded gun in which no "nonetching layer" — that is, no white layer — was found was a Browning machine-gun barrel that had been fired a total of f?000 rounds in 1$ bursts of 100 to Ц00 rounds each at the rate of 1000 rnd/min. The interpretation of this result was complicated by the fact that it was found that the steel inthe gun was WD 1060 — a carbon steel practically the same as SAE 1060 — which is seldom used for machine-gun barrels, and that the steel was very dirty, with coarse dendrites present [Kosting, Carter and Reed, 1937]. 73. Similar altered layer on other steel surfaces The erosion surface of a gun tube resembles in many respects surfaces that may be produced by treatment of steel in a number of different ways. In the gun itself the surface of the chamber __ i,f a cartridge case is not used — is covered with cracks that differ from those in the bore only in that the network' is more uniform and
- 13b lacks the prominent unidirectional cracks which characterize the surface of the lands and grooves. The structure of the surface is altered in the chamber just as in the bore. - The pressure plugs inserted in a 12-in. gun for measuring the maximum pressure also were found by Fay [ 1916, p. 22Ц1] to show an altered layer around the edge of the central hole and around the edge of the plug itself. The surface of the plug was covered with cracks. Greaves, Abram and Rees [1929, p. 116] used this phenomenon to study the formation of the altered layer. They placed cylinders of 0.8-percent carbon steel (pearlitic) in the chamber of a 15-in. gun .during firing. After one round the altered layer was 0.0037 in. thick, after five rounds, 0.011Ц in., and after 10 rounds 0.0118 in- • The last thickness is the same as that which they listed for the bore of a l£-in.. gun. A similar altered layer was reported [Greaves, Abram and Rees, 192 9, p. 11] as occurring "on the base of proof shot or shell of large caliber, and in fact on any steel surface on which the propellant gas has acted for a suf- ficient time." The surface of an 'erosion vent plug frequently shows an altered layer — "white layer," in the older sense, — such as that illustrated by Greaves, Abram and Rees [1929, Figs. 56 to 71 ]• Among hl alloy steels.tested in erosion-plug experiments and after- wards examined at Watertown Arsenal [see Sec. 9h(b)], only a few specimens showed a white layer. These specimens were low in manganese; with 2 percent or more of manganese, no white layer was formed. The three steels that showed the most pronounced white layer contained vanadium [Ritchie, 1928].
- 135 - A phenomenon that may be related to the erosion of steel by powder gases is the Munroe effect. During his experiments on smoke- less powder, Charles E. Munroe discovered that when a disk of gun- cotton is detonated on a horizontal steel plate, any indentations or raised characters originally on the under side of the disk are repro- duced in reverse on the upper surface of the plate [Marshall, 1920; Browne, 1939, p» 1306]. No metallographic examination seems to have been made of a steel surface acted upon in this way. Steel surfaces that have been suddenly heated by friction also show an altered layer. Greaves, Abram and Rees [1929, p. 116] have enumerated the following cases with original references:—' (i) steel rails; (ii) heavily braked railway tires; (iii) wire rope on a pulley; and (iv) a ball bearing run dry. The heat'checking of the surface of brake drums, described by Walther [l 938], was explained by him- in much the same terms as were used in Sec. ?0 for the cracks on the bore surface of a gun. Steel can also be eroded by an electric arc. De Sveshnikoff [1921a, p. 10; 1921b, p. 163] reported that when an electric arc was struck with either a copper or a carbon electrode against a piece of gun steel, the area around the central hole produced by the arc developed a number of cracks and acquired a martensitic structure— 69/ (i) J. E. Stead, Proc. Inst. Meeh. Eng. (1899), p. 77; W. H. Merrett, Proc. Inst. Meeh. Eng. (190Ц), p. II4.O; J. W. Bampfylde, J. Photomicrographic Soc. 1Ц, 23(l 927). (ii) H. Brearley, Metal- lurgist 1_, 132(1925). (iiij E. A. Atkins, J. Iron and Steel Inst. (1927), p. hh3- A more recent contribution to the third subject is E. M. Trent, "The formation and properties of martensite on the surface of a wire rope," J. Iron and Steel Inst. 1ЦЗ, Ц01-Ц19(19Ц1). . 70/ See Table XX for definitions of these structural designa- tions.
- 136 - that merged rather abruptly into the original ferrite-sorbite structure—^ of the specimen. Greaves, Abram and Rees [1929> Fig, 15>] have also published an illustration of the structure of the altered layer produced in this way. Chemical action also can produce an altered structure on the-surface of steel. Four steel bottles used by the General Chemical.Company to contain nitrogen and hydrogen at a pressure of 15>00 Ib/in? and a temperature of $00 to 600°G failed after vary- ing periods of service. Wheeler [1922, p. 2$71 found that "when these containers were cut open and the cross-section surface polished, they showed an inside zone with a different luster from the rest of the metal. Upon etching, this zone was almost un- affected while the rest of the metal etched' normally." A similar alteration of steel was obtained by exposing it to the action of ammonia at a temperature of 600 to 700°F. This effect has been confirmed by later investigators. Thus microscopic examination of a longitudinal section of a machine- gun barrel made of "nitralloy G" that was nitrided by heating in ammonia after-machining showed that'a white layer was present both inside and outside the tube [Lester, 1929b]♦ A white layer was formed on the surface of specimens of a series of steel alloys of about the composition of centrifugally cast steel, when they were casehardened both by nitriding and also by cyaniding [Shumate, 1938]. A white layer similar in appearance to that present in an eroded gun bore, but much thinner, was formed on a piece of steel heated in gaseous ammonia at 900°F for 1^ hr. It was suggested
-137- that the layer in the gun tube is thicker than that produced arti- ficially by reason of the high pressure in the gun [Snair and Wood, 1939, p. 619]. The appearance of a fused surface that is often seen in an eroded gun bore was produced artifically by treating a rough groove in a piece of steel with ammonia at 6J>O°C— and then rough polishing the piece. A photograph of such a surface at a magnification of f>0 diameters was published [Wheeler, 1922, p. 296]. Hydrogen at high temperatures and pressures has a marked action on steel [Inglis and Andrews, 1933j Naumann, 1937]• As pointed out by Kosting,—^ various types of layer are formed which may vary in their resistance to attack by common etchants. In some respects some of these layers are not unlike those observed in eroded machine-gun barrels. V. Stefanides, in a discussion printed with the paper by Snair and Wood [1939] on the white layer, suggested that hydrogen and water vapor "may play a more important role than they are being given credit for. The austenite, when formed in the presence of H30, which evidently breaks down, and the oxygen goes into solution with the carbides, forms a more stable austenite which is hard to break down, requiring higher temperatures than austenites without the oxygen in solution." 71/ The duration of the treatment was not stated; other ex- periments lasted 20 hr at this temperature. 72/ Personal communication, Aug. 22, 19h2. .
- 138 - 7h. Hardness of the altered layer Although many writers in describing the bore surface of an eroded gun have referred to the altered layer as a ’’hardened layer," few measurements exist to show just how hard it is. In fact, some of the early metallurgists seem to have considered it a hardened layer simply by analogy with previously known hardened layers on tempered steel. Thus Osmond [l901J made no mention of having determined the hardness of the layer he described. Fay [1913, p« 133] did measure the hardness, both with a scelerbscope and with a Brinell ball. Although the results by the two methods, are decidedly inconsistent, as is shown in Table XIX, they do indicate a greater hardness at the surface of the bore than deeper within the metal of the tube. This conclusion was confirmed by the measurements of Jones [19h2] of the bore of an eroded :l6**in. gun liner.—/ He found that the hardness at the surface Table XIX. Hardness of altered layer at surface of bore of 12-in. gun specimens.: ~ ’ Specimen Hardness Number No. . Distance from Muzzle (in.) Sceleroscope Brinell Land Groove Back® Groove Back® A 0 33 - 27 31 192 170 В 120 3u 28 31 166 156 C 2h0 38 33 31 187 170 D 325 h7 h3 32 187 1h9 E 337 Ц8 Ц8 30 187 1h9 * After Fay [1913, P. 133]. a The reverse side of the trepanned specimen, to indi- cate the hardness of the unaltered steel. 73/ See footnote 66.
- 139 - varied from 63O to 633 D.P.H,, whereas at a depth of 0.001 in. below the surface it was a little greater, from 660 to 710 D.P.H. Measure- ments at successively greater depths showed thatthe hardness decreased uniformly with depth until below 0.009 in. the hardness was 260 D.P.H., that of the unaltered steel. Shumate [1938> p. 22] reported that he was not able to measure with certainty the hardness of the altered layer on the'surface of the bore of an eroded liner for a 103-mm gun, using a Vickers hardness tester. Greaves, Abram and Rees [1929, P« 116], instead of measuring the hardness of the altered layer on the bore surface itself, measured the hardness of the layer produced on steel cylinders placed in a l3_in. gun (Secs. 73 and 7З). After the cylinders had been fired ten rounds, the Brinell hardness, measured with a 1-mm ball and 30-kg load, had increased from 209 to 3h7- The scratch hardness of the altered layer was determined roughly by Bihlman [1921, p. 3]- He found that a scelerometer scratch З.Ц microns wide on the unaltered steel surface was reduced to 2.8 microns in width where it crossed the white layer. A photomicrograph of a similar scratch but without a measurement of its width# was published by de Sveshnikoff [1923, Fig. 3] , who remarked (p. 796) that it '‘showed that this surface layer was considerably harder than the original steel." The only indication of the hardness of the thin outer white layers are photomicrographs of a scratch across the white layers in a 73-mm gun [Kosting, 1939b, p. 13] and a Bierbaum microcharacter scratch across the white layers in a 37-mm gun [Kosting, 19h0, Fig. 7(b)]. The
- що - former photograph was said to show that “the outside of the white layer is slightly harder than the inside." 75. The altered layer considered as martensite—/ Examinations of the altered layer were made first by metal- lurgists, who tried to explain the hardness of that layer in terms of some process already known to cause hardening of a surface layer on a piece of steel. Fay 916] summarized the three possibilities as (i) cementation, (ii) heat treatment, and (iii) cold-working. He ruled out the presence of free cementite, because the hardened layer did not etch when treated with boiling sodium picrate. He assumed instead that the surface layer was heated considerably above Acx andthen chilled rapidly to martensite by the cold layers below the surface. Although no needles of martensite were found, it was concluded that the foregoing assumption was verified by the fact that the altered layer could easily be changed to troostite by tempering. As a result of experiments with pressure plugs that were locally cold-worked and then fired in a gun, it was concluded that cold-working, as by the rubbing of the projectile on the driving edge of the lands, was an essential part of the hardening process. Fay and'various later writers quoted Rosenhain and Beliew [1913] in support of the view that the altered layer consisted of martensite; but these citations are of little worth because they refer simply to separate statements, made during a discussion at a Scientific congress, that a distinctly martensitic structure had Th/ The several terms used in describing the metallographic structure of steel are defined in Table XX and the notes that accompany it.
- 1h1 - been observed. There -was no opportunity at that time to support the statements by photographic evidence, and none seems to have been published since then by these investigators. Howe [1918] followed Osmond [1901]in considering that the altered layer was either troostite or martensite, depending on the extent to which the alteration had proceeded, the troostite being an intermediate stage. Bihlman [1921, p. 3] was in general agree- ment with this conclusion, for he reported that photomicrographs at 2^0 diameters showed a structureless white layer on the surface, a dark-etching layer immediately behind it and then the original structure of the steel. "The characteristic structure of martensite was not observed in the white layer of any of the specimens examined"; but a determination of the relative scratch hardness of the altered layer (Sec. 73) was said to "bear out the suggestion of Fay that the edge structure is a martensite-troostite-pearlite one." Greaves, Abram and Rees [l929, p. 116] adopted the same general view. They described the "hardened skin" on the surface of the cylinders that they fired in a 1fj-in. gun (see Secs. 72 and 73) as consisting of layers of different structure, as follows: (outside) coarse martensite, fine martensite, troostite, ferrite and troostite, narrow sorbitic region, and, finally, original structure. As a result of his chemical, microscopic and x-ray examination of a worn 16-in. gun (see Secs. 72, 7h, 77), Jones [19ц2] concluded that the hardness of the altered layer was caused by the formation of martensite, as evidenced by the observation that the alpha-iron lattice was distorted to an unusual extent and that this distortion
- 1Ц2 - Table XX. The microstructural oonatituent» of steel. Thia table ia baaed on data obtained from Epstein [1936, Chap. VIII], Evana [1?39, PP. 150-11 and a private communi- cation from the reaearoh laboratory of the U.S. Steel Corporation.
- Tl+3 - Notes on Table XX The allotropic transformation between the gamma and alpha forms of pure iron takes place at a slightly different temperature on heat- ing than on coolings The former temperature (910°C) is designated as Ac3 arid the latter as Ar3, which are abbreviations for the French ex- pressions arret sur chauffage and arret sur refroidissement, respectively, in reference to the breaks on the heating and cooling curves» The Curie point of alpha-iron (about 768°C), above which the iron is paramagnetic instead of ferromagnetic,- is designated as A2. The gamma-alpha transformation- of the iron in steel does not take place at a fixed temperature because it is accompanied by a re- action between the iron and the carbon. Thus on slowly cooling a hypoeutectoid steel — that is,- one containing less than O.87 per- cent C. — from above 91O°C, the transformation begins at Ar3/ As it proceeds, ferrite is formed and the austenite solid solution is enriched in carbon until the composition of the eutectoid between ferrite, austenite and cementite is reached at O.87 percent C. Mean- while the temperature falls until that of the eutectoid line is reached. If equilibrium is maintained, the remaining austenite then transforms to pearlite at this temperature, which is designated as Arx (723°C).- In the system Fe-C, when austenite is quenched the gamma-alpha ' transformation .takes place; but the precipitation of the excess carbon is inhibited. The presence of this excess carbon causes the crystal lattice of the alpha-iron in the resulting martensite to be distorted to tetragonal symmetry. Certain alloying elements that may be added to steel cause the transformation as well as the pre- cipitation to be inhibited, with the result that nonmagnetic austenite can be stabilized at ordinary temperatures. Then martensite is tempered by heating, the lattice distortion of the alpha-iron is removed by the formation of ferrite and the pre- cipitation of the .excess carbon as cementite. The secondary troostite or sorbite that is thus produced contains the same proportions of ferrite and cementite as the pearlite or primary troostite that might have been formed from the original austenite by slow cooling or rapid cooling, respectively; but they differ from it in microstructure. Bainite is a separately distinguished primary structure that is formed by the process of austempering, in which austenite is quenched to the intermediate temperature range 2OO-f>f>O°C and then isothermally trans- formed. The adjectives ’’primary” and ”secondary”' are frequently omitted in references to troostite, the context indicating which type of structure is meant. The troostite mentioned in the various papers on the altered layer is secondary. This constituent is now considered by the research laboratory of the U.S. Steel Corporation as merely a special variety of sorbifee and is- not separately named. Ap especially useful, equilibrium diagram of the system iron-carbon, containing much detailed information, was published by Sauveur [1935,Р»Ь6?
- 1ЦЦ - was removed by heating to ЦОО°С in a vacuum. X-ray examination of an eroded Bren machine gun that had been fired Ц000 rounds at 100 rnd/min also showed distortion of the alpha-iron lattice, but to a less extent than in the 16-in. gun. 76. Carburization as a factor in martenization Although de Sveshnikoff [1925] agreed that the altered layer consisted of martensite and troostite, he maintained that its formation involved not only quenching but also carburization of the steel at the surface of the bore by the powder gases. V/hereas Vieille [1901, p. 219], who had suggested this possibility to ex- plain the hardening of the surface of erosion vent plugs, had been merely speculating, de Sveshnikoff based his conclusion on analyses of chips reamed from six eroded machine-gun barrels made of dif- ferent steels.—' Two successive cuts 0.00Ц or 0.005 in. deep were made for a distance of 1.5 in. forward from the origin of rifling. The samples obtained by the first cuts showed increases in carbon content ranging from 30 to 183 percent, compared with that of cores drilled from the main body of the steel. The samples obtained by the second cuts showed increases of only 1h to 33 percent. Fay [1925], in commenting on the work of de Sveshnikoff, amplified his former theory by agreeing that absorption of nitrogen and of carbon accelerated the formation of the hard layer, since in the meantime he had found out [Fay, 1921 ] that ’’steels are nitrogenized 75/ These were some of the barrels made of special steels, the trial of which is described in Sec. 92.
- 11+5 - as well as carburized in casehardening operations, the absorption of nitrogen preceding that of. carbon." Graziani [1928] also supported the view that carburization is a cause, of the hardening of the sur- face of the bore.. Piantanida [l922], on the other hand, while he admitted that the altered layer consisted of martensite, denied that it was caused by carburization, because it disappeared on annealing. Later he examined guns (caliber not stated) that had been fired to the end of life. He annealed sections of the bore by heating them to 900°C in a slow stream of hydrogen or nitrogen in an electric furnace and allowing them to cool slowly in the furnace. Whereas unannealed ' sections showed a distinct altered layer, the annealed ones did not. Furthermore, the number of pearlite grains was no larger near the surface of the specimens than it was at a greater depth. Hence it was concluded that the altered layer had not been caused by carburiza- tion [Piantanida, 1935]. 77» The altered layer considered as a nitrided layer An entirely'different explanation of the altered layer was given by Wheeler [1922]. From a comparison of the appearance of the white layer found on the surface of steel nitrided in an atmosphere of ammonia with that found on the bore surface of a gun, he concluded that the latter was a locally cold-worked austenitic case due to penetration of nitrogen.. He argued that Since nitrogen under pressure and with heat penetrates and combines with steel as has been described, it would be strange indeed if the reaction did not occur on the bore of a gun where there is an atmosphere of from 35 to Li5 percent nitrogen,
- 1Ц6 - very hot and fresh from combination of another sort and therefore nascent and under a pressure of 30,000 to Ц3,000 Ib/in? Since even molecular nitrogen is known to react at 5>00°C and 15>00 lb/in?, the reaction must occur in a gun. By assuming that the white layer on the inside of a gun is the same as that which can be artificially produced by nitrogen, most of the loose ends of previous theories are picked up. The same samples of material from the surface of the bore of machine-gun barrels that he analyzed for carbon (Sec. ?6) , de Sveshnikoff also analyzed for nitrogen. Although the nitrogen con- tent, which was of the order of a few hundredths of 1 percent, was higher than normal for these crucible steels, it was ’’altogether too low for assumption that the white layer consists of iron nitride. ” It was considered that ’’nitrogenization may or may not be concomitant with carburization," depending on the presence of particular alloying elements, such as chromium, that have an affinity for nitrogen. Also, it was argued, the fact that it had been pos- sible to change the white layer to a dark-etching one by heating it to 320 or 3?0°C proved that it "could not be a nitride of iron be- cause iron nitride does not readily decompose below Ц5>0°С." He did finally concede tliat "the probability of absorbed nitrogen remaining in the solution should not be neglected, since in such a case the layer may have a martensitic character and change at temperatures below the decomposition temperature of nitride." In rebuttal to de Sveshnikoff, Iliheeler [l92$] pointed out that he had not claimed the presence of iron nitride in the altered steel but simply of a solid solution of nitrogen in steel. He i remarked that the nitrogen in solid solution would not have been determined by the analytical methods used by de Sveshnikoff. After
- Шт - he had suggested that the specimens analyzed might have included foreign matter Containing carbon or nitrogen, he Went on to argue that even if the altered layer contained more carbon than the un- altered steel, his nitrogen theory was not invalid. He considered that "whether partial carburization or decarburization takes place is a matter of no great consequence, dependent simply upon the con- centration of carbon and hydrogen in the gases." The suggestion that the altered layer might be the result of nitriding was supported by Lester [1929a] as a result of a micro- scopic and x-ray examination of rings from the same 12-in. gun that had been studied by Fay [ 1913; see Sec. 7h] compared with a similar examination of an eroded caliber .30 machine-gun barrel that had been made of nitralloy and then nitrided by the usual commercial procedure [see Lester, 1929b]. Polished cross sections of the two guns showed similar structures under the microscope. The x-ray diffraction patterns of sections from both guns showed the spectrum of alpha-iron with the lines displaced to a slight extent, indicating an enlargement of the lattice of 0.3 percent. The section from the 12-in. gun showed also a spectrum of gamma-iron that was almost as strong as that of the alpha spectrum (see Sec. 79); but the section from the nitrided machine-gun barrel showed no gamma spectrum. It was concluded, therefore, that the white layer in the 12-in. gun contained about Ц0 percent of austenite, "caused by the alloying effect of nitrogen with consequent lowering of the Arx point to such an extent that the solid-solution phase is formed on the surface," as explained by Wheeler [1923, p. 800]. The lack of a similar
- 1Ц8 - austenitic phase in the case of the machine-gun barrel was explained on the ground that there had been less opportunity for quenching. The hardness of the altered layer was claimed to have been the result of the lattice distortion, which in turn was ascribed to the presence of nitrogen in solution, both in ferrite and in austenite. Confirmation of the x-ray diffraction pattern described by Lester was reported by Snair and Wood [1939], who also found a dis- tortion of the lattice of alpha-iron of O.h9 percent in an exam- ination of the eroded surface of machine-gun barrels. Although they concluded that the white layer was a nitride, they had no other positive evidence than this agreement with Lester's previous deter- mination of the lattice enlargement of a nitrided barrel. They did find that wrhen a sample of the bore surface that showed a distinct white Layer was heated for 1 hr at ЦОО°С and cooled in the furnace to prevent quenching, the white layer remained white. They claimed that this result showed that the white layer was not martensite, bocause martensite would have broken down and would have appeared dark after the heat treatment, which was exactly what Bihlman [1 921] and de Sveshnikoff [1923] claimed that they had observed. The nitrogen contents of successive layers of steel 0.001 to 0.002 in. thick, removed with dilute hydrochloric acid from the bore of an eroded 16-in. gun, J—/ were determined by chemical analysis by Jones [ 1 9Ц2]. He found that in three sections ex- amined the amount of nitrogen decreased gradually from a maximum of 76/ Soo footnote 66.
- 11+9 - 0.11 to 0.18 percent in the surface layer to about 0.008 percent in the core metal at a depth of 0.01 in. The maximum amount of nitrogen in the surface layer was very much less than the 9 to 10 percent that Jones and Morgan [1932] had found in the surface layers of nitrided steels and considerably less than the 1 percent they had found at a depth of 0.011 in. In x-ray diffraction photographs of successive layers of similar sections from the same 16-in. gun Jones [19h2] observed a very faint set of lines, which indicated the presence of a small amount —/ of solid solution of nitrogen in gamma-iron; but he found no evidence whatever of Fe^N and Fe^N, which he claimed '’give particularly strong characteristic spectra when a steel is hardened by the process of nitriding.” He therefore concluded that the hardening of the altered layer was not caused by nitriding. Instead, as mentioned in Sec. 7^, he attributed it to the formation of martensite. 78. Relation of cracks and inclusions to the altered layer Photomicrographs published by Kosting [ 193 9b, 19Ц0] show that some of the deep cracks that penetrate into the unaltered steel are lined by the white layer along their upper* surfaces. Other photo- micrographs (Figi ^(cj, for example) show the birth of such a crack. To what extent the formation of the cracks was accelerated by the white layer is unknown. Inclusions, on the other hand, definitely 77/ In sections of an eroded Bren machine gun which he also examined by means of x-rays, Jones found relatively larger amounts of gamma-iron containing nitrogen, perhaps as Fe^N.
- 150 - seem to., accelerate the deterioration of the surface of the bore. Kosting [19Ц0, p. 8] reported that photomicrographs ’‘show that where a radial crack and inclusion intersect, the crack branched and followed around and finally enveloped the inclusion with a marked effect on the surrounding steel." Other photomicrographs "show that inclusions to which gases had access, and which were adjacent to or within the hard layer, facilitated marked change in the steel and loss of metal at the bore." 79. Significance of the outer white layers The work of Kosting that reveals the previously-unsuspected complexity of the altered layer may represent the means of bringing together the somewhat divergent viewpoints with respect to the nature of that layer. He agrees with the early workers that most of it is martensitic and is joined to the unaltered steel by a troostitic transition zone. At the same time he considers that the two white layers which he has found on top of the martensitic band indicate the occurrence of "a chemical change on the surface that results, in the case of the present steels, not only in the formation of a friable layer that spalls off in thickness of the order of magnitude of 0.000^ in., but also in the selective attack by some agent that penetrates the steels, resulting in cracks and leading to pitting and the loss of metal in large chunks" [Kosting, 1939а]. If this chemical reaction were a nitriding one, as Kosting himself has suggested it might be,—z the two principal points of 78/ Personal communication, August 19Ul.
- 131 - view concerning the nature of the altered layer would be recon- ciled. The identity of the two white layers is still unknown. ' Kosting,[ 1 93 9a ] reported that an x-ray diffraction pattern obtained from the bore surface of a V^-mm gun showed the spectra of both alpha- and gamma-iron, and also some unidentified lines (see Sec. ?6). The white layers do not seem, to be constant in their properties, the two layers being less sharply separated in some cases than in others [Fig. 7(f)]. The significance of variations in the white layer that Kosting [19h1, p. 19] reported he had observed in some other guns is not fully understood. It is possible that the white layer adjacent to the martensitic band represents a zone in which the reaction be- tween the steel and some constituent of the powder gases is still incomplete, and that the easily detached outer layer represents the end product of that reaction.
- 1$2 - CHAPTER X. EXPERIMENTAL METHODS OF STUDY OF GUN EROSION 80. The experimental study of the erosion of guns is performed most directly by actual firing under controlled.conditions. Such a test is both expensive and time-consuming; therefore the need has long been felt for a means of reproducing erosion on a small scale. Erosion plug tests have been performed in the past fox* this purpose, but without complete correlation of the results with those of firing tests. Tests with explosives in closed vessels give little information concerning erosion, Within the past year a laboratory "gun” has been developed at Battelle Memorial Institute, and there is some hope that it may more closely reproduce service conditions. 8l• general description of firing tests .A firing test for the study of erosion consists of firing a gun until erosion has proceeded so far that the gun is no longer serviceable. Usually several guns are fired in the same test, each.gun (or group of guns) differing from the others in only one particular,, such as the kind of steel in the tube, the type of projectile or the kind of propellant. Throughout the test, which lasts for several days with a major-caliber gun, all conditions of firing are kept as uniform as possible. At periodic intervals the velocity of the projectile is measured during a series of rounds and the accuracy of. .the gun during anpther series. In most tests the. gun is cleaned periodically and the bore is star-gaged in order to obtain a record of the progress of erosion. In some tests> other measurements, such as the maximum pressure or the temperature
- 153 - of the tube, are made from time to time. The final result of the test is the rating of the accuracy life of each gun. The differences among them are correlated with the variables being tested, with the help of the measurements made during the test. Frequently each variable in the test is represented by only one or two guns, in which case the interpretation of the results is somewhat uncertain, because the reproducibility of the measurements is not very high. The variation in the performances of two guns that are presumably the same may be almost as great or even greater than that between guns of different design. 79/ 82. Гetermination of accuracy life of guns-—z The accuracy of a gun is' determined by firing at a target. There is no complete agreement, however, as to the criterion to be used in measuring the accuracy. The mean deviation and the extreme deviation from the center of the target have both been used. The area of the smallest ellipse or circle that encloses all the hits gives another measure. The ellipse is preferable to the circle, because the deviations are likely to be arranged along some line as an axis. In the firing of small arms the end of life is often considered to be indicated by the occurrence of a. specified number of ''keyholes" or "tips" on the target — say three in twenty rounds. A keyhole, which is a hole of that shape in a paper target placed at fairly close range, indicates that the bullet passed sideways through the target. 79/ See Chap. XIV for a discussion of the life of guns under service conditions.
- 15Ц - е A. tip is an ellipsoidal hole in the target that indicates that the bullet was yawing—^ considerably (but less than 90°) at the moment when it passed through the target. A bullet that makes a tip would make a keyhole in a target placed farther away, at a distance depending on the rapidly with which the bullet is tumbling. A bullet that makes either a keyhole or a tip in the target would, of course, be very erratic in flight at normal range. The regular occurrence of tips or keyholes is a clear indication that the bore is so badly eroded that some of the projectiles can pass through it without being properly engraved by the rifling (Secs. 8 and 18) and hence that the barrel lias really reached the end of its usefulness [see Sec. 12£(e)]. 83. Cost <x£ .firing tests A firing test of ? special kind of tube or of some other special feature requires, in order to be reliable, a minimum of four tubes —• two regular tubes for control and two special tubes. For each additional variable at least two more tubes should be tested. The test of four tubes would require the services of about 9 men: 2 to operate the gun, 2 to measure the velocity, 3 to deter- mine the accuracy, 1 to star-gage the guns and 1 to supervise the test. The firing of four guns would occupy 15> to 20 days; but the over-all time of the test, including the compilation of the data, would be about three months. The cost of the ammunition required 80/ The' angle of yaw of a projectile is the angle made by its axis of symmetry with the direction of its motion.
- 155 - to fire a'tube to destruction by far exceeds both the cost of the tube itself and also the cost of the personnel*needed to carry out the test. Estimates of this cost for four sizes of.guns are given in Table XXI. Table XXI. Estimated cost of ammunition needed for firing guns to destruction. Gun Estimated Life (rounds) Projectile Ammunition for li guns (complete rounds)* Cost (dollars) Caliber .30 MG" 8,000 Ball Ц8,000 . 1600 37_mm, М3 3,000 HE . 18,000 75,200 l55~mm, Ml 2,000 Bnpty for sand loading 12,000 632,000 16-in.,М1919M3 175 AP, empty for sand loading 1,050 2,088,000 * The number of complete rounds required has been estimated by taking one and one-half times the figure given in the Ordnance Proof Manual for the estimated accuracy life, in order to insure that enough ammunition would be available to fire the guns to the end of life, even if some of them exceeded the nominal estimate. 8Ц. Erosion vent plug tests The erosion vent plug test was introduced by Noble in 1882 for the purpose of studying the relative erosibility of various steels by gunpowder, later it was used by Vieille [1901] to compare both the erosibilities of steels and the erosiveness of different smokeless powders. The method has continued to be used from time to time by other investigators. Their results will be discussed in Secs. 911 and Ю9.
- 15>6 - The explosion bomb— that is commonly used for erosion vent plug tests.is arranged with three openings. One is fitted with an electric connection for firing the powder by heating a wire to incandescence; a second is fitted with a pressure gage (Sec. 89); and the third is airranged so that a cylindrical test plug with a small axial hole can be inserted 'into it. The gases formed when a charge of powder is fired in the bomb escape through this hole with high velocity and enlarge its diameter. The extent of this "erosion" is usually determined by measuring the loss of weight of the plug. Sometimes the loss per unit weight of powder is recorded directly; at other times it is transformed into the corresponding loss of volume of material. Greaves, Abram and Rees [1929] expressed their results in terms of the slope of the (approximately) straight line obtained when the loss of weight was plotted against the maximum pressure. The erosive action in a vent plug test is so great that only a single firing is needed to obtain a measurable loss of weight. For a plug having a 2-mm vent 28 mm long, for instance, a charge of 3100 grains (20Ц gm) of cordite Li.D. fired at a maximum pres- sure of about 2300 bars (33,000 Ib/in?) caused a loss of $ to 13 gm, depending on the composition of the material of which the vent plug was made [Greaves, Abram and Rees, 1929, Table III]. 81/ One such bomb and the results obtained in testing a wide variety of materials are described in a forthcoming NDRC report by 0. H. Loeffler and G. Phair entitled Metals tested as erosion vent plugs.
157 - It seems certain that much of the erosion of a Vbnt plug occurs by reason of actual fusion of a surface layer of metal; but to what extent any other mechanism, such as the friction of the powder* gases; plays a part is not known. 85. Effect of conditions of vent plug test Vieille [1901] made a study of the effect of variation of the conditions under which an erosion vent plug test is carried out. He reached the following conclusions, which have been verified by later workers. (a) Diameter of vent. A given charge of powder produces less erosion the larger the vent, except for extremely small vents» The decrease is not entirely attributable to the lowering of the maximum pressure, because there is much less erosion of a 2-mm vent than of a 1-mm one, although the maximum pressure is nearly the same. Vieille [1901] and de Bruin and de Pauw [1931] used a vent 1-mm in diameter, but Greaves, Abram and Rees [1929] used a 2-mm vent, and « the plugs tested at the U.S.Naval Proving Ground had Ц-mm vents [Bu. Ord., 1931]. (b) Length of vent.- The erosion is greater the longer the vent. Thus the loss of weight of a plug having a 1-mm hole Ц0 mm long was two and one-half times as great as one having a hole of the same diameter only 10 mm long. (£) Quantity of gas. — By firing the same size of vent plug in bombs of different dimensions, it has been found that tho erosion per gram of powder is greater the smaller the charge.- This is explained on the ground that the increase of diameter of the vent
- if>8 - caused by the initial erosion slows down the rate of.erosion during the later stagesj so that the latter part of a large charge is less effective than the same-amount of powder would be if fired as a separate charge through a new vent, (d) Pressure. The erosion produced by a given mass of gas, which had been almost negligible, at a pressure of 100 bars, became considerable when the pressure was increased to 2000 bars; but then it scarcely increased any more when the pressure was raised to Ц000 bars. (e) Repetition of firing. — When the same vent plug is used for a number of successive firings, the amount of erosion decreases gradually, both because the diameter of the vent is in- creased and also because the maximum pressure is less. 86. Influence of turbulence on erosion of vent plugs The experiments of de Bruin and de Pauw [1931], in addition to confirming the foregoing conclusions of Vieille, showed that turbulence of the gas stream is a very important factor in the erosion of a vent plug. They used segmented plugs made up in three sections, in order that the turbulence might be modified by varying the diameter and the length of the middle section. In one experiment, for instance, the effect-of turbulence was clearly shown by using a plug that had a middle section $ mm long with a hole 7 mm in diameter, while the two end sections were 10 mm.long and had holes only 1 mm in diameter. When this plug was fired, using a l|0-percent nitroglycerin powder at a maxi- mum pressure of about 1Ц00 bars, the third section of the plug
- 159 - eroded more than the first, even though the gas had lost some of: its energy by the time it reached the third section. They also found that if the end sections of such a segmented plug were renewed after each round, repeated firing produced a cavity in the middle section that was of the same general shape even though the initial diameter and length of the middle section were changed. The characteristic form of the cavity was an elongated ellipsoid. Thus with end sections 10 mm long having 1-mm holes, the particular charge of powder that was used created a cavity about 20 mm long and 6 or 7 ram in diameter. at its widest part. It was found that after this characteristic cavity had been formed, the rate of erosion decreased considerably. Hence de Bruin and de Pauw concluded that the outline of the cavity represented the streamlines of the gas as it issued from the orifice of the first section of the segmented plug, and that once this form had been achieved, no more turbulence occurred and hence there was scarcely any erosion. 87. Applicability of erosion vent plug tests to gun erosion Various writers have pointed out that care is necessary in drawing conclusions concerning the erosion of guns from the results of erosion vent plug tests. Thus H. F. Leary [Bur. Ord., 1931], then Inspector of Ordnance in charge of the Naval Proving Ground, although he stated that "the Proving Ground believes that erosion plug tests offer the quickest and most economical method of deter- mining the suitability of a steel for gun use," nevertheless, on
- 160 - the basis of comparative tests against standard samples, sum- marized a memorandum on the relationship of erosion plug tests to gun erosion in the following words; The results of the erosion plug tests necessarily have to be considered, in relation to erosion in guns, as showing a tendency towards significance rather than as being sig- nificant in themselves. This is caused by the elements of difference between the two conditions. The period of application of heat, the fundamental significance of the pressure and the quantity of metal eroded in the two cases are radically different. On the basis of a critical comparison of the data from ero- sion vent plug tests published by Greaves, Abram and Rees [1929] with those obtained at Aberdeen Proving Ground and at the Naval Proving Ground, P. R. Kosting [1931b 1938} chemical engineer at Watertown Arsenal, concluded that This is not a sensitive test to differentiate metals. On the other hand, when used to rate gunpowders upon their erosive effects, parallelism between such tests and service tests appears to exist. Before discarding the test, its mechanism should be empirically studied, for it is believed that it has a bearing on the erosion of grooves. A metallographic study of the container, within which the explosions occurred, should be under- taken, for it is believed that more important data will be obtained than the weight loss of the plug. Even de Bruin and de Pauw [1931] concluded the account of their extensive experiments with erosion vent plugs with the following admonition: If one is to apply conclusions.from tests of this kind to the construction of guns, he must be very care- ful, inasmuch as these tests differ greatly in many re- spects from actual firing. Hence it is recommended that only the most general conclusions be drawn, for they are the only valid ones that these tests offer.
- 161 - In view of the results that have been obtained in the course of years with erosion vent plug tests, the following generalisation seems to be justified; namely, that no metal which is eroded badly in an erosion vent plug test would be satisfactory for a gun tube or barrel. Hence the test may have some value in the preliminary evaluation of new materials proposed for tubes or barrels. Even if it would not distinguish the best, at least it would eliminate the worst. 88. Battelle Institute laboratory erosion gun At Battelle Memorial Institute a gun lias been developed to furnish a convenient means of producing erosion in a small test barrel, which is a cylindrical tube of whatever material is being tested, 1 in. long and 0.^0$ in. in inside diameter, with a wall thickness of 0.27 in. The projectile consists of two parts. One piece is a carefully machined solid cylinder about 2 in. long that fits into the barrel. The exposed end of this part of the projectile is screwed into a 19-lb cast-iron weight which slides in a frame between two vertical oak strips that act as a brake by being par- tially compressed by heavy coil springs. The projectile is fired by a solenoid-operated firing pin which sets off 1 gm of 37-пий powder contained in a 12-gage shotgun cartridge case that has been cut down to a length of about 3/h in. The powder is held in place by one of the regular shotgun wads. Under these conditions the maximum pres- 'l . * sure is about 11,£00 Ib/in? and the projectile,attains a muzzle velocity of h0 ft/sec. The gases are in contact with the test barrel for about O.OO36 sec.
162 The standard procedure adopted after preliminary experiments •was as follows. The barrel and its projectile were ground until the diametral clearance was 0.005 in. They were fired a total of Ц00 rounds at a slow rate of fire — 1 rnd/min or slower. After each group of 25 rounds., the barrel was cleaned, weighed and photo- graphed on the inside. At the end of firing, measurements of the wall thickness were made at a considerable number of points^, from which a contour map of the inner surface was drawn. With this gun a wide variety of metals and of steel plated with different metals have been tested. The results have been described in a series of monthly progress reports [Russell,. 19h1a] which contain photographs of the eroded areas. The materials tested have been classified [Russell, 19Ц1b] as follows, on the basis of the loss in weight of the barrel that resulted from the firing of Ц00 rounds in the laboratory gun; (a) Bad materials, loss 0.23 to 0,li3 gm. — Austenitic steels, including the particular steels selected for their re sistr- ance to heat checking in hot-dye work; massive nickel; massive copper; nickel plate; copper plate; gold plate; ajid some chrome plate. (b) Good materials, loss 0.10 to 0.16 gm. — Gun steels; plain carbon steel; most alloy steels of the SAE, machine grade; diffused molybdenum coating; diffused tungsten coating; and nitriding steels. (c.) Excellent materials, loss less than 0.05 gm. •— High chromium (2? percent) steel; chrome plate, when made very adherent; and diffused tantalum coating. Each of these materials, howeveir, has some drawback to its application to gun tubes. Some of the most recent experiments with this laboratory gun have shown that in reality it is a form of erosion vent plug in which the vent is annular in cross section. The movement' of the
- 163 - projectile contributes nothing, apparently, -fco the test of a material, for the same weight loss and heat-check pattern were obtained in an experiment in which the projectile was held in place as in one in which it moved. Consequently the test is being modified to employ a vent of rectangular cross section, formed by two opposite plane surfaces separated by a fixed distance. It is felt that this modi- fied test will give essentially the same results as the one with the moving projectile, with much less trouble in the preparation of the specimen. It has been suggested, furthermore, that one face of the vent can be formed of the material being tested and the other of a standard material, in order to provide continuous control of the conditions of the test [Russell, 19h-2]. 89. Pressure gages—/ In test firings of guns and in closed-chamber experiments the pressure of the powder gases is frequently measured. The crusher gage has long been employed for the measurement of the maximum pressure in a gun. The piezoelectric gage, although it is more accurate, is limited in use to special investigations, because 'a hole has to be made into the chamber of the gun for its insertion, and also because it is a laboratory instrument that requires special care in its use. (a) Crusher gage. — A form of crusher gage was invented by General Thomas J. Rodman of the U.S. Army, in 18^7> and then was . > 82/ The design and response characteristics of gages suit- able for the measurement of the stresses generated by propellants and high explosives are discussed in detail by Goranson, Garten and Crocker [19Ц2].
- 16Ц - modified by Noble [187$; p. 107 in 1906 reprint] and by various later investigators [for example, Vieille, 1906; Desmazidres, 1922; and Burlot, 1923] ♦ Such a gage usually consists of a small metal cylinder at the bottom of a hole in a metal housing. The cylinder is compressed by a piston that is pushed into the hole when the pressure is applied. A gasket at the top of the piston prevents leakage of gas into the hole. The permanent change of length of the cylinder caused by compression is a measure of the maximum pressure. Annealed copper§^/ has been the metal most frequently used for the cylinders; but lead is used in some minor-caliber gages, especially for the measurement of pressures less than 10,000 Ib/in? (700 bars). In order to prevent undesirable dynamic effects that would be produced if the travel of the piston were very great, copper cylinders intended for the measurement of pressures above h0,000 Ib/in? (2800 bars) are pre- comprbssed within 20,000 Ib/in? (1Ц00 bars)* of the expected pres- sure. One or more of the gages are placed in the powder chamber at the base of the charge, after the several parts have been care- fully cleaned and oiled. Although copper crusher gages indi- cate pressures some 10 to 1$ percent lower than those registered by a piezoelectric gage, their results are self-consistent, and hence they furnish a convenient means of measuring the maximum pressure in a gun without altering the gun itself. 83/ The maximum pressure in a gun is sometimes said to have been measured by ’•coppers," which means by copper crusher gages, 8Ц/ Detailed directions concerning the use of these gages are given in Ord.Dept.[1 91$]; Ord.Dept.,[1932] and Ord.Dept.[1936], No. Ц0-16. ~
— 165 — A recently suggested improvement to the technic of pressure measurements with crusher gages [Hickman, 19Ш] is the use of 5/32- or 3/16-in. annealed copper balls instead of cylinders. Because the ball presents an increasing area with increasing com- pression, a much wider range of pressures may be covered. Further- more, the balls of a given lot from a commercial source are so nearly of the same diameter that it is not necessary to measure each one of them before use. The calibration curve is essentially a straight line throughout the range 0 to 15,000 Ib/im (0 to 1000 bars) under static load. The use of a crusher gage in a gun slightly affects the ballistics. Thus it was found in the firing of a 37_mm gun M1A2 that a certain charge of powder would produce a velocity of 2600 ft/sec with the gages in the gun, whereas the velocity was I 5o to 60 ft/sec lower when this charge was fired without the gages [Ord. Dept., 19Ц1]. , This result is naturally to be expected, because the removal of the gages decreased the density of loading. (b) Piezoelectric gages. —A piezoelectric pressure gage consists of several crystals of quartz or tourmaline so arranged that when they are compressed, the charges of electricity that are developed on opposite faces of each crystal are added together. This total charge is very small; the piezoelectric constant for ' quartz at room temperature, for instance, is -6.90 (esu/cm2)/(dyne/cm2) [International Critical Tables, vol. 6, p. 209]; hence a pressure of 1 kilobar produces a charge of 1 microcoulomb when exerted on a
— 166 - total surface of approximately 100 cm? The charge is a linear function of the pressure above a certain limiting pressure. The piezoelectric gage originally devised by Sir J. J. Thomson [1919; also Keys,1921 ] for measuring explosion pres- sures used tourmaline crystals, the charge on which was measured by means of a cathode-ray oscillograph. The gage that has been developed at Aberdeen Proving Ground especially for the measure- ment of powder pressures [Kent, 1927] contains a stack of quartz plates. At first the charge was measured by recording the deflec- tion of a ballistic galvanometer on moving film [Karcher, 1922]; but later measurement was made by amplifying the charge with a vacuum-tube amplifier and then recording it with a General Electric Company oscillograph and the high-speed spiraling-drum camera des- cribed by Eckhardt [1922]. In its present form [Kent, 1938b] this gage is connected through a vacuum-tube amplifier to a cathode-ray oscillograph having a fluorescent screen that is photographed by an external camera. It can be used as far away from the gun as half a mile [Kent and Hodge, 1939]. The especial merit of the piezoelectric gage is that it is capable of recording rapidly fluctuating pressures, because the response of an oscillograph, especially a cathode-ray oscillograph, is so nearly instantaneous. Another important advantage is that by a proper choice of electric circuits, it is possible to obtain directly the time derivative of the pressure as a function of either the time or the pressure (see Sec. 31, Part I).
- 16? - (с) Other pressure gages. — The spring manometer developed by Peteval [190$; also Bone, Newitt and Townsend, 1929, pp. 93-96] is slightly more accurate than a piezoelectric gage, but it is not suitable for use in a gun because its action is affected by the recoil. It has been used extensively in closed-chamber experiments. Another form of mechanical pressure gage that can be used in a gun has been developed at the Naval Proving Ground, Dahlgren, Virginia [Webster and Thompson, 1919]» It consists of a stiff metal girder so mounted that the slight distortion caused by the pressure of the powder gases moves a contact on a variable electric resistor. The resulting change in resistance, and hence the motion of the girder, is recorded by having the resistance form part of a bridge circuit. Although this gage is not as sensitive as a piezo- electric gage, it is much more rugged. The diaphragm pressure gage has been used for the measure- ment of explosion pressures amounting to a few hundred pounds per square inch, but it does not seem to have been used for measuring . pressures as high as those in a gun [Bone, Hewitt and Townsend, 1 929, pp. 99-WO; Caldwell and Fiock, 19hl ]. A recent development is the use of a strain gagq for the measurement of powder pressures, such as the one described by DuMond [19Ul] for measuring and recording the pressure of powder gases in rocket chambers.
- 168 - CHAPTER XI. RELATION BETWEEN PROPERTIES OF THE GUN METAL Al© EROSION 90. Composition and physical properties of the gun metal Various alloy steels have been used for gun barrels and tubes; but a quantitative comparison of the rates of erosion cannot be obtained from the ordinary service records because the guns are not fired under the same conditions. The normal variations in the rate of fire, in the amount and kind of powder, and in other conditions introduce variations of about + £0 percent in the average life of a gun of a given model. Within these limits the variations in the properties of the gun steel do not seem to affect the rate of erosion. 91. Nickel steel Kosting, in his review of erosion [193h], had suggested that steels containing a high percentage of nickel erode more rapidly than some other alloy steels. His conclusion was based on a com- parative firing at Aberdeen Proving Ground [Lane, 1933] of two radially expanded 3~in. antiaircraft-gun tubes, one of nickel steel (C, 0.U5 percent; Mn, 0.7h percent; Si, 0.2Ц5 percent; Ni, 2.Ц1 percent; Cr, 0.13 percent) and one of carbon steel (C, 0.Ц1 percent; Si, 0.217 percent; Mn, 0.65 percent). After 2000 rounds the increase in diameter across the lands was 0.208 in. for the nickel-steel tube and 0.170 in. for the carbon-steel tube at the origin of rifling and 0.02Ц in. and 0.002 in., respectively, at the muzzle. The loss of muzzle velocity, however, was about
- 169 - 1^0 ft/sec for both. guns. In a letter to Datertown Arsenal comment- ing on this report, Rear Admiral Defrees said that Kosting*s con- clusion whs substantiated by the experience of the U.S. Naval Gun Factory. 92. Special alloy steels in machine-gun barrels One of the most extensive series of tests of steels for gun barrels was that conducted jointly by the Springfield Armory and the National Bureau of Standards during the five-year period from 1919 t-o '192Ц, on the effect of chemical composition and other factors on the erosion of Browning caliber f30 machine-gun barrels. Ovep 20 separate reports and memoranda were issued from time to time. Fortunately, a comprehensive report [Bur. Stand., 192ЦЬ] summarizes the results and gives complete references to the indi- vidual reports, some of which contain more details. The compositions of the steels that were tested are listed in Table XXII (A). —For steel of each composition, five full-length barrel blanks were rolled from a i|-in. square ingot that had been made by the crucible process by the Bethlehem Steel Company. The thermal expansion to 900°C and the critical temperatures, which were determined by the method of thermal analysis, were measured at the National Bureau of Standards [see Table XXII (B)], after which a heat-treatment was planned to develop a limit of proportionality 85/ These compositions were chosen by G. K. Burgess, then chief of the Metallurgical Division of the National Bureau of Standards, after consultation with representatives of the Tech- nical Staff of the Ordnance Department, U.S, Army, and with other persons interested in the projectJ
- 170 - Table XXII. Properties of special steels tested for Brov.-ning machine-gun barrels. (A) Chemical composition (weight percentage). Barrel . No. c Mn Cr V p S Si Other 0 o.hh 0.69 0.011 0.017 0.21 1 .h3 1.22 —— —— .0h6 .0h9 .12 2 .38 O.67 —— — .0h6 .055 .07 3 .ho .55 — 0.15 .006 .031 • 13 li .55 .92 — — .020 .026 .26 5 0.32' 0.11 1.20 0.11 0.005 0.028 0.12 6 .27 .hh 0.69 .ih .005 .031 .09 7 .10 .008 2.5h .hi .005 .033 .12 Ni h.5 8 .21 .77 — —— .058 .09h .05 Ni O.h9 9 ,16 .08 1.1h .21 .016 • 0h7 .11 Cu 1.88 10 0.22 0.075 —— 0,008 0.028 0.02 11 • 1h .060 0.93 0.25 .009 .017 .13 Mo 0.60 12 .05 .030 1.38 .21 .011 .019 .91 13 .29 .81 — —— .013 .038 .09 Ni 3.71 ih • 35 1 .h5 1.11 J3 .012 .056 .20 15 0.32 0.28 0.80 0.28 0.017 o.oh6 0.15 Cu 2.76 16 .33 .26 .009 .057 .°5 W 3.82 17 .36 1.21 — —- .053 .077 .09 Ni 0.81 18 .ho 1.30 — .011 .060 1.10 Ni 3.70 19 .50 0.39 — — .009 .013 0.17 W 1.59 20 0.25 o.ho 13.98 0.006 0.010 0.77 22 .32 .66 —— — .023 .028 .37 U 0.22 23 .19 1.05 —— .022 .031 . 16 Ni 36.33
- 171 - Table XXII. (Continued.) (В.) Thermal properties. Barrel ' No.' Average expansion coefficient, x 10° Critical .1 'emperatures 25° to 300°C 300° to 6oo°c 600° t0 900°C . Acx Max. (°C) Arx Max. (°C) Manganese 0 12.1 16.2 16.8 (a) (a) 1 12.7 16.5 16.7 72Ц 65o 2 12.7 15.8 16.6 (a) (a) 3 12.7 16.1 16.8 7b9 680 h 12.9 16.1 16.6 73h 666 10 12.5 16.0 16.9 U) (a) Manganese, chromium, vanad: .um $ 12.9 16,0 16.2 76h 718 6 13.1 15,9 16.8 750 692 11 12.5 .15.9 16.1 — — 12 12.7 15.2 15.8 801 735 1h 13.3 15,6 16.2 753 659 Manganese, nickel - 7 12.1 1Ц.З ——— (b) (b) 8 12.6 16.1 16.Ц 731 670 13 12.1 15.3 ——— ' (a) (a) ' 17 12.9 16.1 * 16.Ц (a) (a) 18 . 12.3 - 15.0 716 . Manganese, chromium, vanadium. copper 9 12.7 15.7 16.7 771 688 15 12.9 16.1 16.9 759 682 Tungsten, manganese 16 12.5 15.7 16.5 7U1 681 19 12.2 15.9 ——— 7U0 680 Chromium (stainless) 20 11.0 13.3 13.7 • —— — (a I No thermal miWe taken. (b i No definite transformation. -
- 172 - Table XXII. (Continued.) (C) Heat treatment and tenaile properties. Barrel No. Temperature(°F) Brinell Hard- ness Tensile Strength (lb/in?) Prop. Limit (lb/in?) т Elonga- tion (percent) Reduction in Area of Cross Section (percent) Quencha Temper13 Manganese 0 1600-0 900-N 2Ц1 12h,65O 75,000 20 Ц6 1 i5oo-o 1250-F 2hh 1ih,65o 85,000 23 59 2 1600-0 950-N 228 112,800 70,000 20 Ц6 3 1575-w 1200-F 217 103,100 80,000 26 65 Ц 1500-0 1250-F 223 110,750 67,000 2Ц 58 10 1650-w 500-0 177 81,100 55,000 25 6Ц Manganese, chromium, vanadium 5 1600-W 1200-F 196 96,200 60,000 26' 65 6 1600-W 1250-F 196 95,100 73,000 26 69 11 1650-w 1325-F 286 12h,050 105,000 23 66 12 175o-w 500-0 235 110,000 53,000 19 Ц2 111 1625-0 1250-F 2h1 115,600 100,000 23 58 Manganese, nickel 7 1650-w 11 75-f 2Ц2 118,100 80,000 20 66 8 . 1550-w 700-N 185 90,Ц00 63,000 28 68 13 ili75-o 1125-F 235 112,300 92,000 26 66 17 1525-0 900-N 225 113,500 90,000 2h 5U 18 1600-0 1150-F 277 1U2,5OO 65,000 27 Ц6 Manganese, chromium, vanadium, copper 9 1650-w 1300-F 235 108,100 98,1iOO 23 ' 6Ц 15 1650-w 1125-F 2li1 1O7,U5O 70,000 21i 66 Tungsten 16 19 1550-W 1550-w 1100-F 1250-F 2Ц1 2Ц8 9h,500 1 oh,85o 70,000 90,000 28 29 61 56 Chromium (stainless) 20 1800-0 125o-fc 269 120,600 75,000 22 60 Nickel (invar 23 . None None 1h9 9h,000 6t,25o 33 63 a0, oil; W, water. bp, open flame; N, nitrate bath; 0, oil. cHeat-treated after machining.
- 173 - Table XXIII. Firingtests of Browning machine-gun barrels made of special steels. First. Test Second Test No. Accuracy Life ... (rds) Condition No. Order of Accuracy Muzzle Velocity 2000 rds 6000 rds 7000 rds Initial (ft/sec) Loss, 6000 rds (percent) Manganese 0 1 6000 8000 Bulged Bulged 1B 1C 8 8 * 7 J 2618 2622 li.Oh 3.05 2 8000 Bulged 2B 1 # — 265U 5.06 2C 3 * - 26ЦО 3.30 3 10000 3B 6 * — 2655 3.1U 30 1 11 6 26hU U.3U U 7000 Bulged 1»B 5 * — 2618 2.93 UC 5 * - 2616 3.60 10 2317 1 Manganese, chromium, vanadium 5 9000 Bulged 5b 9 5 — 2612 2.75 5c u 6 1 2629 2.62 6 6000 Bulged 6B 13 7 , - 2623 3.16 11 UOOO Bent 12 8000 12B 11 2 — 2601 1.97 12C 6 10 - 2660 3.31 Manganese, nickel 7 8000 8 8000 8B 2 10 — 2635 3.32 8C 10 13 - 2665 U.70 13 9000 13B 10 8 - — 2605 2.10 13C. 7 12 - 2613 : 1.61l 17 8000 17B 7 3 — 2612 3.95 17C 2 5 Il 2622 3.00 Manganese, chromium vanadium, copper - 9 8000 9B 12 2 2668 5.19 9C 9 8 - 2665 . Л.62 15 1705 Bulged Tungsten, manganese , 16 8000 16B 15 .11 - - 2620 2.71 19 9000 19B 3 1 — 2609 2.53 19C 11 U 2 26Ц0 2.65 Chromium (stainless) 20 Barrels spoiled in 20B 1U 12 1 2632 3.18 machining operations 20C 9 2 5 26U3 5.15 Uranium 22 22B 22C U 6 13 9 3 2620 2631 2.79 2.70 * Not rated.
- 17h - of at. least 75>,OOO lb/in>, coupled with a minimum elongation of 20 percent in a 2-in. gage length. After some trial quenchings, each group of barrel blanks was given.the heat-treatment best adapted to it, and then the microstructure was examined. The heat- treatment and resulting thermal properties of each steel are listed in Table XXIl(C). One barrel was made from each group of blanks and fired for erosion at Springfield Armory. The rate of fire was not stated, except that 5>0-round bursts were used for the targets. All barrels were fired 11,000 rounds regardless of their accuracy life, except for several barrels that became unserviceable because of bulging. The end of life of the barrels, which is. listed in the second column of Table XXIII, was taken as the point at which keyholing (Sec. 82) occurred on a screen 100 ft from the muzzle. • Two more barrels were made up from each of most of the re- maining, original ingots. These barrels were fired for erosion, and the accuracy was measured after each 2000 rounds. The order of accuracy that is listed in columns p, 6 and 7 of Table XXIII was determined by the relative degree of clustering of the shots. It should be noted that the order after 6000 rounds was consider- ably different from that after 2000 rounds. Part of the reason for this variation was the smallness of the differences among the various barrels, but part of it also reflects the general order of uncertainty of such a test. As mentioned by Hatcher [ 1920a], ''experience with service machine-gun barrels has shown that several barrels of identical heat-treatment and material will vary in accuracy life from I4.OOO to 8000 rounds."
. -175- . The conclusions [Bur<Stand. ,1 92lib] were.as.followb: It is very evident from the. results obtained that there are noticeable differences in the behavior of. the different barrels. 'Just how much is to be attributed to the different variables in composition is somewhat problematical. The results available are .entirely too few to permit any very / definite or wide-sweeping.conclusions being drawn. The wide discrepancies found in some cases for barrels of the same steel in the two firing tests or even between two barrels of the same steel in the second firing series emphasizes this fact. In ranking the barrels with respect to erosion, it is most convenient to compare them with barrel No. 1 [the nominal, composition of which was essentially that of the "standard1* used for machine gun barrels]'. Of the barrels tested, Nos. 22, 20, 3 and 17 are superior to the "standard barrel," the first two being distinctly so. They all showed less wear on the lands and grooves at the breech section of the bore. For many of the others, including, the "standard," the rifling was completely worn away for a distance of several inches at the breech end wdth a considerably smaller number of rounds fired. In accuracy all the barrels listed above surpassed the standard, the comparison being based upon the order in the arrangement of the targets. The depth of penetration of the erosion cracks was also considerably less in most cases and the decrease in velocity noticeably less. In this respect, the conclusions expressed here are not in complete agreement wdth that reached in the final report from the Springfield Armory which was to the effect that "none of these barrels have shown any marked improvement over the regular service barrel in regard to accuracy life." The magnitude of the difference between these conclusions depends, of course, upon how "marked" the improvement must be in order to receive recognition. While it is evident from the tests that superiority of the different steels over the standard is associated with differences in composition, no definite statements can be made upon this point. However, concerning tho correlation of "erosion11 and mechanical properties, it appears from the. results available for comparison that erosion resistance cannot be predicted, a priori, by a determination of the mechanical properties, at least as carried out ordinarily at room temperatures. By way of example, steel No. 5 (chromium- vanadium), which is here ranked as superior to the "standard," showed in the mechanical tests a proportional limit of only 60,000 and an. ultimate tensile strength of 96,200 lb/in?, whereas another chromium-vanadium steel No. 9 (containing copper), which did not show superiority to any extent in the firing tests, had a proportional limit of 9В,Ц00 and an ultimate tensile strength of 108,100 lb/in?, the ductility
- 1?6 - in the two cases being not materially different (No. 5, elongation, 2 in. or 26 percent, reduction of area, 6$ per- cent; No. 9, 22.5 and 63.5 percent, respectively). The malfunctioning of a gun on account of bulging or similar causes appears to be more closely related to the mechanical properties of the steel than is the erosion resistance. Although bulging was not experienced with all the steels having the lower tensile properties, the cases which did occur were confined, for the greater part, to such steels. However, it should be borne in mind that the results of these tests indicate that one cannot pre- dict with any degree of assurance the malfunctioning of a gun by reason of bulging or similar causes on the basis of the low tensile properties of the steel. Three barrels that were made of invar steel were tested as a supplement to the Springfield firings just described. They have been listed as No. 23 in Table XXII(A), The metal was so tough that drilling was difficult [Ames, 1921]. The three barrels were fired until pronounced keyholing of the target occurred at 100 ft. The numbers of rounds fired were: barrel No. 1, 3500; No. 2, 17ОО; No. Ц000. The lands were entirely removed for a distance of 6 to 12 in. from the origin of rifling [Hatcher, 1920b]♦ In addition to the conclusions already quoted, it was sug- gested [Bur. Stand., 192hb] that the steels that had superior erosion resistance showed very small permanent dimensional changes as a result of thermal transformation. It was admitted, however, that not all steels which showed small change were erosion re- sistant. 93. Nonferrous metals used for gun barrels Metals other than steel that have been tried for machine-gun barrels have included beryllium-copper, monel and molybdenum. The
- 177 - barrels of beryllium-copper, which were tested at Springfield Armory oz / about 1938, had extremely short lives.—' The U.S. Navy has tried some, monel-metal guns in an effort to g У / reduce sea-water corrosion. The erosion— of a 3-in./5O-caliber unplated monel gun (No. 5533) was 0.06l in. at the origin after z, 88/ 6Ц1 E.S.R.,— which was about the same as that of a standard 3-in./5O-caliber Mk. 20, unplated steel gun (No. 2956) firing the same ammunition at a muzzle velocity of 2700 ft/sec. After 1255 E.S.R. the erosion of the monel gun had increased to 0.085 in. A molybdenum liner was tested in a caliber .30 machine gun at Aberdeen Proving Ground in 1936 [Lane, 1936]. The liner, which was 6 in. long, was shrunk into a caliber .30 barrel blank and then bored, chambered and rifled. Because the liner was pushed back- wards by the powder gases while the bullet was in the forward part of the barrel, the barrel had to be cut off -jj in. in front of the liner. For comparison a standard М2 barrel was similarly altered. Each barrel was fired Ц6Ц9 rounds at-such a rate that a thermo- couple placed on the barrel registered a temperature of 900°C. Although the molybdenum liner did not have a homogeneous structure and had not had a smooth inner surface because of difficulty in machining it, the lands of this liner eroded less than those of the steel barrel — 0.007 in. and 0.011 in., respectively, after 86/ Capt. С. E. Balleisen, U.S. Army, personal communication. 87/ Lt. В. B. Mills, U.S. Navy, personal communication. 88/ Equivalent service rounds; see Sec. 12Ц.
- 178 - Ц000 rounds — and the grooves did not erode at all. Microscopic examination of a longitudinal sectipn of the molybdenum liner showed that there were no heat checks in front of the chamber, •whereas the standard steel barrel had the customary heat checks. These results were considered to support the gas-washing theory of erosion advocated by Greaves, Abram and Rees [1929], the ero- sion of the lands having been ascribed to the abrasion of the bullet and the lack of erosion of the grooves having been ex- plained by the high melting point of molybdenum (2бОО°С). 9h. Erosion vent plug tests (a) Pure metals. — Other investigations of the erosion resistance of metals have been made by means of erosion vent plugs. Vielle [1901], for instance, measured the erosion of a number of pure metals and alloys in the form of vent plugs fitted to a closed chamber that had a volume of 17-8 cm3 and in which was fired a charge of 3.55 gm of ballistite containing 50 percent nitro- glycerin. Vieille claimed that for pure metals the erosion in- creases in inverse proportion to the melting temperature. This con- clusion is only partially substantiated by the data that he gave, which is reproduced in Table XXIV. In terms of volume, which was the basis of comparison that Vieille used, zinc eroded less than aluminum, and one sample of copper eroded less than iron or platinum. (b) Alloy steels (U.S. Army). — The effect of alloying elements on the erosion resistance of steel was investigated by Ritchie [ 192 8] by means of erosion vent plugs. He made up a series
- 179 - Table XXIV. Erosion of pure metals as a function of melting point.' Metal Melting Pointa(CC) Specific Gravity^1 Erosion0 Volume (mm3) Mass (gm) Platinum 1755 21.5 59 1.27 Iron 1535 7.8 68 0.53 Copper 1083 8.9 998 .88 Silver 960 10.53 231 2.U5 Aluminum 660 2.60 2238 • 5.82 Zinc Li19 7.15 1018 7.28 *After Vieille [1901]. aThe melting point given is that listed in the International Critical Tables. The values given by Vieille had differed somewhat from them. ^The specific gravity is that given by Vieille, which does not differ much from current values. cThe mass of metal eroded has been computed for this table from the volume eroded and the specific gravity given by Vieille. d-The value given is the mean of two for "red copper for crusher- gage cylinders." Vieille also listed four values for commercial red . copper, ranging from Ц8.7 mm3 (O.h3 gm) to 90.8 mm3 (0.8l gm). This wide variation was perhaps due to" impurities. of Ц1 alloy steels and subjected them to the following tests: , (i) proportional limit, (ii) tensile strength, (iii) elongation, (iv) re- duction, (v) Charpy impact. These tests were conducted at 7O0, 600° and 1000°F on annealed and heat-treated specimens and also at 70°F on both annealed and heat-treated specimens (a) after cold-working and (b) after cold-working followed by soaking at h80°F. The cold- working was obtained by cold-drawing to a 6-percent increase in length. Photomicrographs were taken of all specimens, both annealed and heat-treated. The two steels which had the most all-round
- 180 - desirable physical properties were one containing 0.25-0.30 per- cent C, 1.25 percent Mn and 0.35 percent Mo, and another contain- ing 0.25-0.30 percent С, 1.50 percent Mn, 0.35 percent Mo and 0.075 percent Zr. Specimens of the annealed and heat-treated bars (not cold- worked) in tho form of vont plugs were tested for erosion at Aberdeen Proving Ground [Kent, 1926]. Each erosion test bar, which weighed 19.Ц oz, had a vent 0.156 in. in diameter. It was fitted to a closed chamber having a volume of 1000 cm3, which had been constructed from a 5-in. gun. A charge of 7 lb l5 oz of powder was fired, all the gas from which escaped through the vent in slightly over sec. After erosion the hole ranged in diameter at the breech end from-0.35 to 0.53 in. and at the muzzle end from 0.32 to О.ЦЦ in. The enlargement at the breech end was always .the greater. The maximum, minimum and average losses in weight for the three classes of steel—are given in Table XXV(A). Each chrome or chrome-vanadium steel was tested in both the annealed and the heat-treated conditions; and the same was true of all but three of the manganese steels. Three of the stainless steels were annealed, the fourth was heat-treated. 89/ P. R. Kosting (personal communication, Aug. 22, 19112) has suggested that "the data collected by the Army by means of the erosion -vent plug can be divided into two subgroups; one of low-alloy steel and the other of high-alloy steel, such as 12 per- cent Or stainless alloy. Yihen sensitivity and reproducibility of the results are taken into consideration it is believed that Greaves, Abram and Rees* conclusions that low-alloy steels have the same erosion rate in vent plugs will be substantiated.”
f - 181 - Table XXV(A). Loss of weight of erosion vent plugs. Type of Steel No. of Specimensa Loss in Weight (percent) Max. Min. Avg. Manganese 5i (27) 30.7 20.8 27.3 Chrome and Gr-V 22 (11) Ц3.1 21.7 28.9 Stainless 12 percent Or, 0.Ц percent Ni h (Ц) _ 5 36.2 38.2 *After Ritchie [ 1 92 8 ]. aThe number in parentheses indicates the number of different compositions of each type of steel, some of which were tested in duplicate. It wafe observed that heat-treatment seemed to have little effect on erosion, and' that '*the apparent average depth of change in struc- ture from the bore outward on the manganese steels was 0.031 in. as against 0.025 in. for the chrome and chrome-vanadium steels. In other words the depth effect on the manganese steels is greater while at the same time they suffer slightly less erosion than do the other steels"[Ritchie, 1 928] (see Sec. 73)- (с.) A^°y steels (U.S, Navy). — The Naval Proving Ground has also tested a large number of erosion vent plugs. A summary'[Bur. Ord., 1931] of one series of these tests was enlarged [Bur. Ord., 1938a] to include subsequent results. Each plug-was tested by fitting it in the. base of a 12-in. armor-piercing projectile, volume 2h9 in? and firing a charge' of 2.25 lb of pyro powder through its orifice, .after which the loss of weight was measured. The wires attached to the electric primer ("squib") that was used to set off
- 182 - the powder passed through the hole in the plug, the diameter of which was not stated. The tungsten steels, nitralloys, stainless steels and monel were all inferior to gun steels, as is shown in Table XXV(B). Both the plain carbon and the nickel gun steels Table XXV(B). Loss of weight of erosion vent plugs. Type No. Loss of Weight (percent) Gun steels, chrome plateda 10 9.7 “ 11-7 Gun steels, unplateda 38 10.1 - 12.3 Tungsten steels (0.6, 1.1, and 2.3 per- cent w)- 3 11.8 - 12.5 Nitralloys, nitrided h 12.6 - 13.1+ Stainless steels (12.2 to 20.6 per- cent Cr) 7 13.1 - 16.Ц Inconel (79.6 percent Ni, 13.5 percent Cr, 5‘. 7 percent Fe) 1 13.7 Konel (66.9 percent Ni, 1.Ц percent Fe) 1 20.0 К-Konel (6^.6 percent Ni, 3«5 per- cent, Al, 29.1 percent Cu, 1.3 per- cent Fe) 1 22.3 "'^After Bur. Ord. [193da] aGun steels include carbon steels, nickel steels (2.2 to 3.3 percent Ni) and low alloy steels containing 1 percent or less of Ni, Cr, V and Lio. that were chrome plated eroded less than the unplated plugs of the same steel; and in both cases the least loss was suffered by the plug plated with the thinnest coating (0.0005 in.) 95. Effect of structure of steel on erosion The structure of the steel may also affect its erosion re- sistance. Rockwell [1919], for instance, suggested that slag in
- 183 - the steel nay accelerate erosion. Photomicrographs made by Kosting [19й0] support this conclusion (see Sec. 78). Kosting [193h, p. 11 also suggested that in guns, ’’steels of uniform sorbitic micro- structure with absence of much free ferrite tend to have longer life than steel not treated for uniformity of structure.” 96. Effect of radial expansion (cold-working) on erosion The effect of radial expansion on the erosion of a nickel-steel gun was investigated in the case of a 3_in. gun [No. 1(B-1)] that was compared with a built-up gun (No. 91). After having summarized the results, Lane [1933] concluded that the radially expanded gun was superior to the built-up gun; but the reason fbr this conclusion was not stated except by reference to the curves showing the enlarge- ment of the bore as a function of the number of rounds. These curves, however, do not substantiate the conclusion, for the enlargement at the origin after 2000 rounds was 0.15 in. on the lands and 0.08 in. in the grooves for the built-up guns, whereas it was 0.21 and 0.15 in., respectively, on lands and grooves of the radially expanded gun. The star-gage records show also that after 2000 rounds the lands of the radially expanded gun had been enlarged 0.02Ц in. at the muzzle, whereas those of the built-up gun had been enlarged 0.011; in. On the other hand, the accuracy life of the radially expanded gun was 10 percent greater than that of the built-up gun — 2220 rounds instead of 1990 — and the loss of muzzle velocity was only 95 ft/sec for the radially expanded gun after 2000 rounds compared with 165 ft/sec for the built-up gun. Also the mean deviations in velocity, were somewhat less and much more regular for the radially
- 18Ц - expanded gun than for the built-up one. The comparison was com- plicated by the fact that several different kinds of projectile were used in the firings of the built-up gun. Also the gun steels, made by different manufacturers, were different in chemical com- position and heat treatment. All in all, therefore, the comparison was inconclusive. 97. Effect of condition of bore surface on erosion The nature of the surface of the bore in a mechanical sense may also influence erosion. Thus both Fleming [1919] and de Sveshnikoff [1922] considered that tool marks represent ’’incipient erosion,” for the reason that the powder gases would be expected to react more quickly with a roughened surface than with a smooth one. Also recent researches in connection with such processes as Chrysler's "Super-finish” have shown that the deformation of the surface of steel during finishing varies with the type or process used [Wulff, 19Ul ]; therefore it might be expected that the sur- face of a gun bore finished by machining would have been deformed more than one finished by a less drastic process, and would con- sequently be more reactive to the powder gases. The bores of 90/ some guns are now being honed,—• but records are not yet avail- able to show whether such a surface resists erosion any better than the usual machined surface. If any improvement did occur it would only apply to the lands, because the grooves are cut after the honing operation. 90/ Personal observation, Watervliet Arsenal, 19Ц1.
- 185 - 98. Chrome-plat ed. Navy guns Since the middle thirties U.S. Navy guns have been chrome plated to protect them against sea-water corrosion. After such plated guns had been in use, it was discovered that they eroded less than unplated guns fired under the same conditions — especially at the muzzle. A quantitative comparison of a plated gun with un- I plated guns was made during the.experimental firing of 1 Ц-in./50-cali- ber guns in 1938. Gun No. 110L2, Mark C, was fired 207 equivalent service rounds during 209 actual rounds, which was somewhat beyond the end of effective life. The behavior of the chrome plating was markedly different at the two ends of the gun. At the muzzle, 91 / ... flaking was first noticed at 52 E.S.R.— but no appreciable amount of plating was gone until 62 E.S.R., when plating had worn rapidly on top of the lands (60 per-. . cent gone for 5 ft at 62 E.S.R.) so that at 73 E.S.R. plating was gone from on top of the lands for 6 ft aft. From then on the plating wore off farther aft slowly until at 12Ц E.S.R. it wras worn off for 18 ft aft of the muzzle. Wear on the side of the lands and in the grooves was first noticed at 83 E.S.R. Q5__p_ercent gone), but wear here was slow and at 12Ц E.S.R. amounted only to 30 percent for 18 ft aft (wear on driving sides being greatest). [Bur. Ord,, 1938b, par.11,] At the breech end, on the other hand, erosion was scarcely affect- ed by the chromium plating. The plating on the slope of the origin was gone after 15 E.S.R.; the plating from the origin to 18 in. forward of the origin wore rapidly and at 36 E.S.R..was 65 percent gone. At Ц.1 E.S.R. the plating at the origin was all gone, de- creasing to 55 percent gone Ц ft forward, then to 1 percent gone 8 ft forward. The plating then wore with regularity to 73 E.S.R., when it was 90 percent gone 6 ft forward., . 91/ Equivalent service rounds; see Sec. 12Ц for method of com- putation.
- 186 - decreasing to 1 percent gone 9 ft forward. From then on the wear was slow until at 12Ц E.S.R. the plating was 90 percent gone 8 ft forward and 1 percent gone 11 ft forward. [Bur. Ord., 1938b, par. 11.] In the intermediate section of the gun, from 11 ft forward of the origin to 18 ft aft of the muzzle, a distance of 21g ft, the plating appeared intact at 12Ц E.S.R, [Bur. Ord., 1938b, par. 11]. The life of a 1 Ц-in./ЗО-caliber gun was considered as the number of rounds at which enlargement at the origin, AD, reached 0.Ц in,, or enlargement at the muzzle, AM, reached 0.1 in., or at which [Bur. Ord., 1939a, par. 6] Z [ x i ( AD/O. Ц j + |(AM/0.1)]'- 1. Inasmuch as AM reaches 0.1 in. before AD reaches 0.Ц in. in un- plated ЛЦ-in, guns, the end of life is caused by muzzle erosion. Therefore the retardation of muzzle erosion by chrome plating increased the life of gun No. 110L2 about Ц0 equivalent service rounds, from 113 to 133 E.S.R. [Bur. Ord., 1939a, par. ЦЬ]. The enlargement at the origin of chrome-plated gun No. 110L2 after 207 E.S.R. was О.О38 in. more than the average for unplated guns; but inasmuch as the total enlargement amounted to only 0.3h8 in., it was not enough to decrease the life of the gun [Bur. Ord., 1939a, encl. 0]. A slight decrease in the loss of velocity of chrome-plated gun No. 110L2 up to 80 E.S.R. was attributed to decreased erosion forward of the origin of rifling [Bur. Ord., 1939a, par. Це]. This decrease was manifest at first, but then the rate of erosion at a point 1 in. forward of the origin increased
- 18? - with respect to the rate for unplated guns, so that at about 130 E.S.R. the total erosion at this point -was the same as for unplated guns, and after 207 E.S.R. it -was 0.028 in, greater. 99. Experimental chrome-plated Army guns The interest of the Ordnance Department of the Army in chrome 92/ plating—' -was originally directed toward small arms. The first experiments with chrome-plated rifle barrels at Frankford Arsenal in 1927-’28 indicated that a 5>0-percent increase in accuracy life was obtained, but that the process cost Ц0 percent of the price of a new barrel. More extensive tests of rifle, pistol and caliber .30 machine guns confirmed the conclusion '‘that chromium plating very materially increases the resistance of small bores to both erosion and corrosion.’* [Ord. Dept., 1929» 1 As a result of subsequent erosion tests of caliber .30 plated machine-gun barrels at Springfield Armory, it was reported that plating had doubled the life [Whelen, 1930]. A microscopic examination at Watertown Arsenal [Lester, 1929c] of two of the rifle barrels plated at Frankford Arsenal showed that the plating adhered strongly, and ’'was found on the most eroded sections even after the firing tests.” However, it was observed that some particles of chromium torn from the plating, had been driven into the softer base metal,- which may have accounted for the presence of the chromium in the heavily eroded sections. 92/ A resume of this subject, including extracts from the original reports, was prepared by Vincent [19371» His report was the source of the information in Sec. 99 unless otherwise credited, i
- 188 - During antiaircraft firings in the fall of 1928, two chrome- plated 37-ШП1 Browning automatic guns Ml and two chrome-plated liners for 3~in. AA guns M1 were tested. The plating, which was 0.000!? in. thick in the 3-in. liners and of unknown thickness in the 37-mm guns, began to flake off after a few rounds. No difference was observed in the increase of diameter at the origin of rifling of these guns com- pared with unplated ones; but in neither case did erosion extend as far. forward in the plated as in the unplated guns. It was recommended that plated guns be retested after the technic of plating had been improved, The guns subsequently tested comprised a 3~in. liner plated to a thickness of 0.001 in., fired in 1929> and four 3_in. liners plated to thicknesses of 0.001!?, 0.0020 (2), and 0.0030 in., fired in 1930. The plating also flaked off the surface of these guns, but not as quickly as from the earlier ones. It was again observed in the plated guns that, although erosion at the origin of rifling was not affected very much, the forward extent of erosion was markedly decreased. Be- cause none of these guns was fired to the end of life, the effect on accuracy life is not known. It was observed further that the thicker coatings seemed less satisfactory than the thin one. Exact comparison of the plated and unplated guns was not possible, because the star- gaging was not done systematically. The next trial of a .chrome-plated gun was that of liner No. 138 for a 3_in. AA gun М3» in 1936. This liner, after having been machined at Watervliet Arsenal, was chrome plated to a thick- ness of approximately 0.000!?' in. at the Naval Gun Factory according to the standard procedure used there.
- 189 - The gun- was fired in conjunction with tests of pressure gages. Star-gage records were made frequently. The first evidence of erosion was noted after 2Ц7 rounds, at which time the diameter of the lands had increased ;0.002 in. at the origin of rifling. By the time the gun had been fired 6Ц0 rounds the increase in diameter of the lands at the origin amounted to 0.060 in., and the increase across the grooves was 0.019 in. (In the other chrome-plated 3“in. liners tested in 1929-30 the increase in diameter at the origin had been only 0.03 to 0.0Ц in.' at 700 rounds.) The eroded region extended about $ in. forward of the origin. Beyond that point- the chromium plating remained intact. Measurements of muzzle velocity and of pressure made at inter- vals with a standard charge showed a steady decrease. By round 677 the muzzle velocity had changed from the original value of 28j?O ft/sec to 27U3 ft/sec and the pressure had dropped from about 35>jOOO lb/in? to 31,600 lb/in? In view of these changes, according to Vincent [1937, p.36], "it was decided that the cost of chromium-plating liners was not warranted inasmuch as incre sjd accuracy life was not obtained. At any rate, as Vincent reported, a project for the testing of 8 additional liners was abandoned, and on March 12, 1936 the 93/ The justification for this conclusion is not clear, inasmuch as an unplated 3-in. antiaircraft gun liner (No. 1183) fired at a-different time showed an increase of 0.06Ц in. at the origin after only 326 rounds, as contrasted with 6Ц0 rounds. A careful comparison of the conditions of firing of these two guns and of other unplated ones needs to be made; but unfortunately the writer has not been able to find the records.
- 190 - Ordnance Committee directed ’’that the development of chromium- plating technic be closely followed by the Ordnance Department} with a view to its possible consideration for gun-bore treatment when the progress of the art appears to warrant such considera- tion.” Up to the time of writing, no new trials of chrome-plated cannon have been made, although early in 19Ц2 Watervliet Arsenal furnished three oversize 37~ram tubes to the Van der Horst Corpora- tion of America for experimental plating. Also, recent experiments at the Battelle Memorial Institute have indicated considerable im- provement in the technic of chrome plating [Russell, 19Ц2], tests of which were made with their laboratory erosion gun (Sec. 88). 100. Partially chrome-plated machine-gun barrels A new idea in chrome-plating was developed during 19h1 at Springfield Armory, namely, to plate only the breech end of the bore where most erosion occurs. Use of a short anode for plating only the first 7 in. of the bore of a caliber .30 machine gun permits better control of the plating, and it also decreases the cost. The plating is only O.OOO25 in. thick. One interesting result of these experiments is that not only is the erosion reduced, but also the muzzle velocity is increased, in one case frbm 2610 to 27h7 ft/sec [Benson, 19h1]• 101. British nickel-plated guns As an economy measure, eroded gun tubes are now being plated with nickel and then rerifled. This process, which originally had been suggested by de Sveshnikoff and Haring [192h],
- 191 - has been developed by the Research Department of Woolwich Arsenal [A.C., 191i2a ]. The process, which involves the deposition of a layer of nickel 0.1 in. or more thick at the breech end of the bore, has about reached the stage at which it can be used on a production basis for 3-7“ and li.3-in. antiaircraft guns [A.C. 19112b]. One of the plated 3-7“in. guns has been found to have an accuracy life as long as that of a similar steel gun. After 7^0 rounds, nearly one-third of which were bursts of 25 rounds each, the increase of diameter at 1 in. from the origin of rifling was 0.1511 in. as com- pared with 0.191 in. for the average of normal service guns. Some flaking of the nickel plating had occurred. Ranging was normal and accuracy was still satisfactory. The loss of muzzle velocity, however, was greater than with a service gun [O.B.P., 19112a]. A subsequent trial of two more barrels gave less favorable results. 9h/ One barrel showed poor accuracy after 100 E.F.C.—' and definitely bad results after 1150 E.F.C. The. accuracy of the other was fair up to about 1100 E.F.C. It was reported that "the complete trial shows that satisfactory results may be achieved by this method of repair, but that it is not yet reliable" [U.B.P. 191i2e]. 911/ E.F.C. = equivalent full charge, which is the British term corresponding to equiv-alent service round (see Sec. 12li).
- 192 - CHAPTER XII. CONDITIONS' OF FIRING THAT INFLUENCE THE RATE OF EROSION ot/ 102. In previous sections—' the effect of various features of the design of parts of the gun on the rate of erosion have been considered. Of equal or even greater importance are the condi- tions under which the gun is fired. Although the records of the erosion of guns under varying conditions are not complete enough 96 to evaluate all the factors that conceivably might be suggested,— they do indicate qualitatively the effect of some of them. The present chapter gives a resume of the influence of a number of the conditions of firing, illustrated by examples of data concerning the rates of erosion of actual guns. 103. Rate of fire - It has long been known among artillerists that a rapid rate of fire wears out a gun faster than a slow rate. This conclusion has been confirmed frequently in proving-ground tests. Thus Tolch [1936a, p. h] reported that the erosion produced in a 9$/ In Sec. 9 Part I, on Effect of design of rifling on ero- sion, mention should have been made of the work of Jeansen L1936'j, in which the results of calculations of the pressure on the sides of the lands are given for different riflings in a 12-in./3O-cali- ber gun. They show that although much of the pressure on the lands at the origin of rifling can be eliminated by the use of a rifling that starts with zero twist and then increases, a very large pressure is then exerted on the lands during the forward two- thirds of the travel of the projectile. 96/ Both Cranz [ 1926, vol. 2, p. 1 j and Schwinning [193b, Chap. IV] have given extensive lists of such factors.
- 193 - 105-mm antiaircraft gun by 169 rounds of rapid fire was much greater than that produced by 77O rounds of prior routine firing. The rate of erosion at the origin of rifling during the first 770 rounds was 0.00018 in./rnd, whereas during the next 169 rounds it was 0.00025 in./rnd. The increase in erosion in the forward half of the tube was even more marked. This gun was fired at its normal muzzle velocity of about 2775 ft/sec in four bursts, three of which were at the rate of 8 to 9 rnd/min and the fourth at half this rate. A recent test, in which eight 13~lb barrels and three 9j-lb barrels were fired under different conditions, resulted in the con- clusion [Grant, 19Ц1 ] ’’that the erosion in the bore of a caliber .50 machine-gun barrel varies directly as; (a) length of the burst; . (b) number of bursts; (c) cyclic rates; (d) lightness of the barrel; and (e) velocity of the ammunition." This conclusion is in agree- ment with those of earlier tests [Fleming', 1918]. One of these tests illustrated the effect of continuous fire. Two caliber .30 Lewis aircraft automatic-rifle barrels were fired in bursts,of approximately 500 rounds. At the end of the first 500 rounds ero- sion was slight; but after 1500 rounds the rifling wore away rapidly. The barrels became a bright cherry red, and the temperature of the outside of the chamber as measured with an optical pyrometer was 978°C. The two barrels bulged just beyond the chamber after 2Ц25 and 2885 rounds, respectively, and a bullet passed through the wall of one of them at the bulge. The steel used in these barrels had the following composition: C, 0.20 percent; S. 0.91 percent; Mn, О.ЦЦ percent; Or, 1.Ц7 percent; and V, 0.15 percent. Its tensile Strength was 110/500 lb/in?, and its elastic limit was 75>OOO lb/in?
- 19b - ' ЮЦ. Temperature ’bf the barrel The increase in rate of erosion with an increase in the rate of fire is usually considered to be a result of the correspondingly higher temperature of the surface of the barrel. A direct measure- ment of this factor was made by Tolch [1935] who fired six caliber .30 Ml 919 Browning machine-gun barrels to destruction at different tem- peratures, using two lots of single-base and two lots of double- base powder. The temperature of the barrel was measured at first by resistance thermometers and later by thermocouples welded 3«7 and 11.2 in. from the muzzle. After 100 to 200 warming rounds, during which the temperature rose 1.3° to 1.9°C per round, 500 (or 1000) rounds were fired in bursts of 10 rounds at such frequency as to maintain a predetermined temperature at 11.2 in. from the muzzle. The accuracy life — corrected for warming and velocity rounds — which was determined on the basis of the extreme spread on targets at 1000 yards (see Chap. XIV, as given in Table XXVI. Table XXVI. Accuracy life of machine-gun barrels fired at different temperatures. Temperature (°c). Accuracy Life (rounds) Single-base Powder Double-base Powder 200 19 050 21 000 300 12 180 10 830 hoo 9 060 6 650 Although with the same ammunition there was considerable variation in the heat effects of the different barrels, the conclusion is
- 195 - unmistakable that the life of the barrels fired at a- high tern-, perature was shorter than that of the barrels fired at a lower temperature.. 105. Muzzle velocity The muzzle velocity of a gun is not an independent factor in the rate of erosion, for it is dependent on the weight of the powder charge, the weight of the projectile, the length of the gun and the maximum pressure. Data are not available for an evaluation of the separate influences of the length of the gun and the weight of the projectile. For a given gun firing the same projectiles, of course,, these factors are eliminated, and any increase in the rate of erosion caused by an increase in muzzle velocity is due to the combined effect of the increased powder charge and the resulting increased pressure. It is convenient, therefore, to refer comparison of the rates of erosion to this combined effect as expressed by the muzzle velocity, although actually most of the increased erosion is probably due to the increased charge, for the pressure itself, which is discussed in Sec. 110, apparently has little effect on the rate of erosion. Even for different guns, such as the three listed in Table XXVII, the apparent inverse relation between the life of the gun and its muzzle velocity probably is merely a reflection of the correspond- ing variation of the weight of charge. The interpretation of this comparison is complicated by the fact that the three guns were not fired at the same pressures or with the same charge. (a) 3~in. antiaircraft guns. — In a comparison of two nickel- steel radially expanded 3-in. antiaircraft guns, one gun, No. 1(B-1),
- 196 - Table XXVII. Comparison of life of guns and muzzle veloc- ity. . Gun Length (calibers) Tift, of Charge (lb) Pressure (lb/in?) Pro j.Wt. (lb) Muzzle Vel, (ft/sec) Est.Life* (rounds) 75-mm pack howitzer 15.93 1.0 26 000 15.0 1270 12 000 75-mm M1897A2 ЗЦ.5 2.0 36 000 ih. 7 1950 10 000 3-in. AA gun М2 55.0 6.0 111 000 12.8 3000 2 500 *Ord. Dept., 1936, Ii0-15, Table I, pp. 2-Ц. was fired initially at 2600 ft/sec with an average "copper pressure”— for 5 rounds of 26,800 lb/in? and the other gun, No. 2(B-2), was fired initially at 3000 ft/sec with an average copper pressure for 5 rounds of 38,200 lb/in? The two guns were fired with the same powder charges, whereupon the pressures and velocities decreased gradually. .After 631 rounds, gun No. 2(B-2), which had been fired with 6.5 lb of powder, had a velocity of only 276O ft/sec at a pres- sure of 31,000 lb/in? The charge in this gun was increased pro- gressively during the next 12 rounds, until it blew up at a pressure of 113,200 lb/in? The corresponding decrease of velocity and pres- sure in gun No. 1(B-1), which had been fired with 5-h5 lb of powder, resulted in a velocity of 2Ц26 ft/sec at an average pressure of 23^700 lb/in? during rounds 723-727• The increase in diameter at the origin of rifling of the gun fired at higher velocity amounted to 0.181 in. on the lands and 97/ See Sec. 89(a).
- 197 - 0.126 in. in the grooves after 6Ц0 rounds. The corresponding en- largements of the lower-powered gun after 6Ц0 rounds were O.O87 and O.056 in. It required 1530 rounds to enlarge it to the same extent as the higher-powered one had been enlarged after only 6Ц0 rounds. At the muzzle after 6J4O rounds the higher-powered gun had increased 0.0$0 in. in diameter on the lands; the other not at all. Even after 2200 rounds it had increased only 0.030 in. The rate of loss of velocity was 0.32 ft/sec per round for the higherrpowered gun and only 0.05 ft/sec per round for the other [Lane, 1933]» (b) h«7~in. guns. — The report of a recent endurance firing of an experimental model of the U.S. Army’s new Ц.7-in. antiaircraft gun not only illustrates the rapid erosion that occurs- at a muzzle velocity of about 3Q00 ft/sec but also furnishes an opportunity to compare this rate of erosion with that of another gun of the same caliber fired at almost the same pressure with different charges and hence different muzzle velocities. The muzzle velocity of the 50-lb projectile of this Ц. 7-iri. gun, T2, when the tube was newz (rounds 16-20, inclusive) was 31 h2 ft/sec with a charge of 23.5 lb of single-base powder, which gave an average maximum copper pressure of 37,220 lb/in?; but with this same charge the muzzle velocity de- creased, rapidly at first and then more gradually, to 2975 ft/sec at the end of Ц00 rounds. The maximum pressure had dropped corre- spondingly to 29,000 lb/inT The erosion that had caused these changes in pressure and muzzle velocity had been such as to increase the diameter across the lands at the origin of rifling by 0.175 in. at the end of Ц18 rounds. The erosion of the grooves had been about
- 198 - half as great and, consequently, the distinction between the lands and the grooves was lost at the origin of rifling. From an extra- polation of the behavior of the gun for Ц63 rounds, the ’’economic accuracy life'* [Sec. 125(f)], calculated on the basis of the proba- bility of hitting an airplane under certain specified conditions, was ihOO rounds. However, it was admitted in the report of this calculation [Altschuler, 19hl] that, although no stripping of the rotating band had been observed during the rounds fired, it is probable "that the rotating bands of the projectile would be stripped before the predicted accuracy life would be reached and, therefore, that the stripping of the rotating band would determine the actual life of the gun.’’ In contrast to the rapid erosion of this high-velocity h.7“in. gun is the record of an earlier model, the U.7“in. gun M19O6. The complete history of one of these guns is not available, but a record [Ord. Dept., 1922c] has been found of the charges and copper pres- sures during one period of the life of gun No. 808. Between rounds 22I4I4 and 2312, inclusive, lj.1 rounds were fired with a charge of 6.5 lb of powder at a mean-pressure of 30,050 Ib/inf The mean velocity was 1703 ft/sec. The remaining 28 rounds were fired at lower velocities, in the range 1529 to 1700 ft/sec. Thus, if this sequence of firings may be considered representative, the pressures in this gun were just about the same as in the h.7-in. gun T2, but the powder charges were only about 30 percent as large. The muzzle velocities were correspondingly some 1Ц00 ft/sec lower. The erosion of tho lower-powered gun was so slight that a star-gaging after
- 199 - 3386 rounds showed an increase in diameter at the origin of rifling of only 0.023 in. on the lands and 0.002 in. in-the grooves.. (c) 3-in./105-caliber gun. — A striking example of the ex- tremely short life of a hypervelocity gun was furnished by the 3-in./l0£-caliber gun, Mark’ XVI, No. 61-L, tested by the U.S. Navy in 1922. This gun was fired Ц6 actual rounds, in nearly all of which a charge of 17 lb of nitrocellulose powder was used. With the 10-lb projectile that was used in most of the rounds, the average muzzle velocity was Ц7^0 ft/sec at an average maximum pressure of about Ц8,000 lb/in? The increase of diameter at the origin of rifling amounted to 0.1Ц0 in. in this gun after Ц6 rounds, whereas in 82 3-in./5O-caliber guns fired at a muzzle velocity of 2700 ft/sec, the average increase was only 0.028 in. after rounds 5 that is to say, the erosion of the higher-velocity gun was five times as great after only one-tenth as many rounds [Naval Ord., 1926]. (d) The Paris gun. — The German long-range guns that were used to bombard Paris during World War I were 21-cm guns — one was rebored later to 21д cm — made by inserting a lining in a 38-cm gun and having it extend beyond the muzzle of the original tube-. The total length was 36 meters (17h calibers). The muzzle velocity was 1^00 to 16OO meters/sec (Ц92О to 32Ц8 ft/sec) at a maximum pressure of ЦЦ,000 lb/in? The probable life of one of these guns was not more than 5>0 rounds [Miller, 1921]. 106. Type of propellant The effect on erosion of difference in the kind of propellant was early recognized, for it was the change from black powder to
- 200 - smokeless powder that made the erosion of guns a serious problem. After the experience of the Boer War, the British decreased the proportion of nitroglycerin in cordite in order to reduce erosion [Marshall, 1917* p. 30Ц]. According to Greaves, Abram and Rees [1929, p. 122], ’’the change from cordite Mark I (calorific value, 1150 cal) to cordite K.D. (950 cal) trebled the life of big guns. Similarly the use of nitrocellulose (810 cal) in place of cordite 98/ M.D. leads to increased life.”— 107. Double-base versus single-base powder ( The comparison between double-base and single-base powder is illustrated by some recent data obtained in the endurance tests of the 37-mm gun M1A2. For the first 500 rounds the increase in diameter of the lands at the origin of rifling was the same (0.00Ц in.) for a gun fired with du Pont single-base powder as for another gun fired with Hercules double-base powder. Then the ero- sion of the gun fired with double-base powder began to increase rapidly, so that after 800 rounds, the lands had increased in diameter a total of 0.01Ц in. while the lands of the other gun had increased only 0,005 in. The increase in diameter of the grooves was 0.009 in. after 800 rounds for the gun firing double-base powder and only 0.001 in. for the other gun. The expected life of the gun that had fired double-base powder was only 1800 rounds [Ord. Dept., 19Ы> Report 13] compared with 3000 to 3500 rounds for the one 98/ See the quotation from O.B.P. [19h2d] in Sec. 108.
- 201 firing single-base powder in bursts of £0 rounds each with external cooling between rounds [Ord. Dept., 19Ц1, Report 11]. In the test of machine-gun barrels already referred to (Sec. 10Ц), half of the barrels were fired with single-base and half with double- base powder. As is shown in Table XXVI, those fired with single-base powder at 300° and ЦОО°С had the longer accuracy lives. The com- parison at 200°C was complicated by the fact that one barrel was fired with both single-base and double-base powder and had an actual accuracy life of 27,770 rounds. The life of 21,000 rounds listed in the table was an estimate of what would have been expected if the barrel had been fired only with double-base powder. Furthermore, a lower rate of fire was required with the double-base powder for the maintenance of a given barrel temperature. Hence, as Tolch [193$} p. 20] pointed out in his report on the test, ”if the barrels had been tested at constant rates of fire, the differences in accuracy life would have been considerably greater, other conditions being equal, because of the greater heating effect of nitroglycerin-type powder." An important observation reported by Greaves, Abram and Rees [1929, p. 123] was that cordite and tubular nitrocellulose (NOT) powders produced the same sort of eroded surface. Two li.£-in. howitzers of nickeL-steel and two 6-in. howitzers of nickel-chromium- steel that had been worn out by firing one of each type with cordite and one with NCT were examined. ’'There was no characteristic dif- ference in the type of crack produced by either propellant. The cracks in all instances were sharp at the bottom.! Their maximum
- 202 - depth was about 0k05> in. in the tubes worn out with NOT and 0.03 in. in those worn out with cordite, the apparent depth of the crack being reduced by the ..more rapid erosion of the surface of the tubes worn out with cordite. The character of the hardened skins and of the surface cracking was the sane in each case, the difference be- tween cordite and NOT being confined to the more rapid removal of steel from the surface by cordite/* 108. Picrite powders The опЗу information at hand concerning the erosion produced by picrite powders is the statement that "the erosion produced by cordite N is about one-half that produced by American nitrocellulose cannon powders. The erosion caused by cordite NQ is about equal to that produced by American nitrocellulose powders.” [O.B.P. 19h1a.] No quantitative data were included. Another report [O.B.P. 19h2c] suggests that in Q.F. ЦО-mm (Bofors) guns "the use of NH propellant may bo expected approximately to quadruple the life of the gun," compared with its life when firing cordite W.T. In a later action the Ordnance Board recommended "that in all guns one cordite full charge (ll.D., S.C., W or W.l-i.) should be reckoned 99/ as one E.F.C.,— nitrocellulose full charges (NH or FNH) and cordite NQ as half E.F.C. and cordite N as quarter E.F.C." [O.B.P. 19h2d]. 99/ See footnote 9h.
- 203 - 10?. Erosion vent plug tests The relation between the adiabatic explosion temperature (Sec. 36) of the propellant and the rate of erosion was investigated by de Bruin and de Pauw [1931J» In two series of experiments at pressures of 1^,500 and 20,000 lb/in?, erosion vent plugs of "Silberstahl’* (0, 1.28 percent; Mn, 0.31 percent; Si, 0.20 percent; P, traces; S, 0.029 percent)^ having axial holes 1 mm in diameter and 25 mm long, showed losses of weight that were linear functions of the temperatures of combustion as computed from the composition of the smokeless powder. These computations were based on the authors' previous measurements [de Bruin and de Pauw, 1928] of such temperatures as a function of composition. The temperatures ranged from 2б750 to 35?5°C. It was also found that two powders of very different composition but having the same combustion temperatures produced the same amount of erosion. 110. Pressure The effect of pressure on erosion is difficult to evaluate, because in guns of the same design any marked increase of pressure must be obtained by an increase of charge. Hence any variation of the rate of erosion may be ascribed to either^ cause. Vieille [1901] suggested that the pressure affects erosion indirectly by deter- mining the amount of gas that flows through a crack. This sugges- tion, which is bound up with Vieille*s theory of the mechanism of erosion (see Sec. 11Ц), has been neither proved nor disproved. (a) 6-pounder guns. — One occasion on which two guns of the same caliber were fired at different pressures with the same charge
- 20Ц - of powder was when two. new 6-pdr. guns — Mark III, No. 18, American Ordnance Co., and Model 1900, No. 3h* Driggs Seabury R. F. — were fired a total of 995 rounds each., in two series, at a muzzle velocity of approximately 21j00 ft/sec. During the first series of 600 rounds, gun No. 3h was fired with slow-burning powder and gun No. 18 with quick-burning, and during the second series of 395 rounds, the powders were interchanged. The slow- burning powder was U.S. Navy 6-pdr. S. P. 592, web 0.03h in., and the quick-burning powder was U.S. Navy 3-pdr. S. P. 53h, web 0.023 in. The mean pressures were about 32,500 lb/in? with the former and 37*000 to 38*000 lb/inf with the latter. After 600 rounds the erosion in the rear portion of the rifled bore of the gun firing the quick powder was very slightly greater than that of the other gun. The difference was described as 11 inconsiderable.11 After 995 rounds the wear on both guns, which was not near the limit of endurance, had been approximately equalized by the interchange of powder [Ord. Dept., 1907]» From this experiment we may judge that the difference of 15 percent in the maximum pressure had practically no effect on the rate of erosion of these guns. (b) Ц.7-in. guns. — The comparison of the erosions of a ! h.7_in. gun and а Ц. 7-in. howitzer, already cited in Sec. 105(b), shows clearly how little influence the maximum pressure per se has on the rate of erosion. It seems that only with reference to a particular gun, firing a certain kind of powder, is the maximum pressure related to the rate of erosion, and then only as an index of the amount of charge.
- 205 -' (c) Erosion vent plug tests of propellants. — The corre- sponding effect in the case of erosion vent plugs is shown by the results of two series of experiments at pressures of 1050 and 1350 atmos and at different temperatures ranging from 2675° to 3575°C, made by de Bruin and de Pauw [1931] The extent of the erosion of the plug was a linear function of the temperature for each pressure, that is, the extent of the erosion was increased by the same absolute amount at all temperatures; and hence at the low temperatures, where the erosion was slight, the proportional effect of the increase in pressure was-much greater than at the higher temperatures. 111. Lubrication of the bore Lubrication of the bore of a rifle or handgun firing plain lead-alloy bullets is essential to prevent the barrel from becoming ’’leaded,'1 as described by Sharpe [ 1 9Ц2, p. 81]. The lubricant ordinarily is applied to the bore by putting it in the cannelures cut in the body of the bullet. It consists of a mixture of several of the following constituents, the proportions of which vary accord- ing to the whim of the handloader: carnauba wax, ”Japan” wax, bees- wax, tallow, ozocerite, mineral oil, castor oil, graphite, and J sodium aluminum stearate (dictaphone-records shavings). 112. Lubrication of bullet by base wad Another method of lubricating a rifle bore that was in vogue in the days of black powder was to place a wax wad on the base of the bullet before inserting it into the cartridge case. The wad
- 206 - was cut from a thin sheet of wax of the desired composition. This practice, which is still used by some riflemen when firing lead bullets, has been extended recently to the firing of high-velocity metal-case bullets. Sharpe [ 19Ц2, p. 8Ц] claims that the wax wads he has used with both the .220 Swift and the .257 Roberts cart- ridges ’’not only improve accuracy, but practically eliminate ero- sion when light bullets are driven at velocities above Ц000 ft/sec.11 He goes on to quote [Sharpe, 19U2, p. 85] J. B. Sweany, one of the originators of the .220 Swift cartridge, as having stated that "all Swift loads require grease wads for best results, both in reducing erosion and producing maximum accuracy. ц A wax wad especially recommended by Sharpe for this use contains colloidal graphite in the form of Oildag, which he claims [Sharpe, 19h2, p. 85] penetrates "the pores of the barrel metal under the pressures developed during firing, thus sealing these pores and greatly re- ducing wear on the barrel." Lubrication of small arms bullets for military use has never been adopted by the Ordnance Department of the U.S. Army on the grounds that it would be a useless practice. 113* Lubrication of artillery projectiles Artillery projectiles have been lubricated at times. Thus during World War I the Italian army used an asbestos band impreg- nated with wax and tallow. The French greased the outside of shells; and instructions were issued to the A.E.F. to grease the 155-mm howitzer projectiles, apparently in acceptance of tho current French practice. The Ordnance Committee in 1922, however, approved a report by a subcommittee recommending that this practice
- 20'7 - be not followed on the grounds that the grease may cause the intro- duction of dirt into the bore, that the use of grease is an unneces- sary expense and inconvenience, and that tests at Aberdeen Proving Ground with greased projectiles in 75>_™. guns had shown no decrease in erosion. Details concerning these tests have not been located. Limited trials by the U.S. Navy of experimental projectiles having cannelures filled with graphite and grease did not indicate any improvement with respect to either erosion or coppering.122/ Similarly, Greaves, Abram and Rees [1929* p. 123] reported that ”a comparative examination was made of two 60-pdr. guns, each of which had fired about h5>00 equivalent full charges of cordite M.D. One had been lubricated with wool grease throughout the last two-thirds of its life and the other had been worn out dry. The main differences observed were a slightly greater maximum depth of scoring in the nonlubricated gun, and rather more coppering in the lubricated gun. The character of the surface cracking was almost identical at similar positions throughout the tubes." In spite of these negative results, the use of a bore lubri- cant as a cure for erosion is suggested probably more often than any other remedy. The a .priori reasoning is sound, being based on the proven value of a lubricant such as graphite in reducing wear on metal dies. The lack of similar success with the use of a lubricant in the bore of gun may be due either to the operation of other more powerful causes of erosion or else to improper 100/ Lt. B. D. Mills, personal communication
- 208 - application of the lubricant. In this latter respect the ”J” band having cannelures at the front end, which is discussed in Sec. 122, would seem to represent about the best means of apply- ing a lubricant to the bore surface at the point of contact be- tween the rifling and the rotating band.
- 209 - CHAPTER XIII. EFFECT ON EROSION OF LEaKAGE OF GAS PAST THE PROJECTILE 11Ц. Theory of gas leakage Although doubts have often been expressed as to how well, the rotating band performs its function of obturating the bore, there seems to have been no systematic experimentation to determine whether it does or does not. Consequently, most of the writings concerning the influence of gas leakage on erosion have been speculative. Lanfroy [1885] and Bourgoin [1913] supposed that this leakage takes place through a crescent-shaped opening between the projectile and the bore, caused by dilation of the bore by the pressure of the powder gases. Vieille [1901], who was the leading exponent of the theory of gas leakage, supposed that it takes place principally through the small openings that result from failure of the rotating band to mold itself into the fine cracks on the surface of the bore (Sec. 70). He argued that the particular openings through which leakage occurs change from moment to moment as the rotating band moves forward; and that therefore it is only at the breech end of the bore that the rotating band is moving slowly enough for a given set of cracks to remain in action long enough to cause much erosion. Charbonnier [1922, pp. .100Ц-10С5], on the other hand, who opposed Vieille*s theory, stated that projectiles fired without rotating bands gave the same velocity and pressure as was obtained with the' same kind of projectile fired with bands. He claimed that
- 210 - the same was true of a tubular projectile. He suggested that these phenomena indicated that as the projectile moves forward it carries with it an annulus of gas that acts as an obturator. The best exposition of the objections to the theory of gas leakage was given by Alger [1910] in a paper written in reply to the one by Yarnell [1910] mentioned in the next section. He pointed out five objections to the theory, which may be summarized as follows: (i) the rate of wear of the smooth bore just at the rear of the rotating band is the same as that of the bottoms of the grooves at the origin of rifling; (ii) the lands erode twice as rapidly as the grooves; (iii) the opportunity for gas leakage should be greater halfway down the bore, because of wearing of the copper rotating band, than at the beginning of travel; (iv) the erosive effect of the gas that escapes past the projectile depends solely on the temperature and pressure of that gas, which conditions are the same in low-powered and high-powered guns firing the same kind of powder at the same maximum pressure, although these guns erode at different rates; and (v) the rotating bands of fired projectiles show no erosion. 115. Vent plug experiments The erosion vent plug experiments performed by Vieille [Sec. 9h(a)] were offered as evidence of the erosive effects of hot powder gases rushing through a small opening with a high veloc- ity. Neither he nor Yarnell [1910], who espoused Vieille's theory and who reported additional erosion vent plug experiments, performed
- 211 any other experiments that more closely simulated the conditions in a gun. A special vent plug experiment reported by Yarnell consisted of placing a can containing water on 'the inside of the chamber in such a manner that the water was forced out through two pinholes into the top of the chamber near the vent. He found that in this way the erosion of the vent plug was reduced to a third of its former value. He suggested that in a gun, erosion similarly might be reduced by having a circular receptacle containing water — or other liquid — at the rear of the projectile so that it would burst upon the shock of firing and would permit the water to seal the bore before and shortly after the projectile begins to move. Yarnell suggested that "as the necessity for such a seal exists for an exceedingly small fraction of a second, not much liquid would be required, and its high latent heat would effectively cool such small percentage of the total volume of gas as would escape past the projectile." There is no record that Yarnell’s sugges- tion has ever been tried, 116. Photography of gas leakagel^/ High-speed photographs of a large projectile coming out of the muzzle of a gun were cited by Dr. L. ,T. E. Thompson—as sup- porting evidence that leakage of gas is not an important factor in 101/ Photographs of the gas' that leaks past,a rifle bullet when fired into a vacuum are being taken at the Geophysical Labora- tory. 102/ Personal communication, September 19U1.
- 212 - erosion. These photographs show that only a ’’small” amount of gas escapes ahead of the projectile in comparison with what comes out behind it; but inasmuch as there is no way to estimate the actual quantity of this gas, the evidence is not convincing. 117. Escape of gas through holes drilled in the shot Greaves, Abram and Rees [1929, p. 119] described an experi- ment that caused them to conclude that leakage of gas past the projectile was a negligible factor in erosion. Inasmuch as it represents about the most thorough investigation of the subject that has been made, their description is quoted in full: It has been suggested by Vieille that not only is scoring due to escape of gas past the driving band, but that normal erosion is also due to such escape through the minute channels existing in the cracked surface layer. It is a matter of importance to determine the minimum size of orifice through which gas can pass without under- going sufficient cooling to render it inactive as an erosive agent. Two 18-pdr. proof shot were drilled longitudinally, one with a central axial hole 0.$ in. in diameter, and the other with four holes each 0.25 in. in diameter, spaced symmetrically relative to the axis of the shot and to each other. The two shot were then fired under the normal conditions adopted for solid 18-pdr. proof shot, with, the following results: 18-pdr. Proof Shot Weight of Projectile Muzzle Velocity (ft/sec) Maximum Pressure (tons/in?) Solid 18 lb,8 oz 1677 13.65 Drilled with hole 0.5 in. in diam. 18 lb,1| oz 1602 12.35 Drilled with h holes 0.25 in. in diam. 18 lb,l£ oz 1607 12.65 There was thus a decrease in recorded pressures and muzzle velocities relative to those obtained for solid shot.
- 213 - The fired shot was sectioned. Microscopical exam- ination of the metal in the neighborhood of the holes showed the presence of a hardened-slowly-etching sur- face layer. The structure at a position near the nose of the shot drilled With four holes is shown in Fig. 16 (Plate IV). The depth of this hardened skin appeared to be constant throughout the length of the shot, andwas 0.003^ in. for each shot. Brinell hardness measurements made with a 1-mm- ball and 30-kg load on the bore surface of the drilled holes gave ЗЦ1 in comparison to 220, the normal hardness of the material of the proof shot. It was thought desirable to employ a smaller ori- fice, but a hole much less than 0.25 in. in diameter could not be bored, through the full length of a proof shot. Two 18-pdr. proof shot were therefore drilled with axial holes 0.5 in. in diameter. The forward end of these holes was then fitted with a plug of annealed 0.8-percent carbon steel. This steel (the same type as was used for the cylinders subjected to chamber conditions during firing) was chosen because it is sensitive to the formation of a hardened skin and has poor resistance to erosion. The plug in one shot was drilled with an axial hole 2 mm in diameter, while the other was drilled longitudinally with Ц holes each 2 mm in diameters 18-pdr. Proof Shot • Weight of Projectile Muzzle Velocity (ft/sec) Maximum- Pressure (tons/in§) Solid shot 18. lb,8 oz I672 13. Ц0 Shot fitted with plug with 1 hole, 2 mm in diam. 18 lb,L|. oz 1668 13.35 Shot fitted with plug with Ц holes, 2 mm in diam. 18 lb,3loz 1659 ... 13Л5 There was thus no diminution in maximum pressure, and only a slight loss in muzzle velocity. The plugs were weighed before and after the proof shot had been fired; the loss of weight on firing.was 0.180 gm for the plug containing one hole, and 0.233 gm for that with four holes. As the plugs suffered some slight abrasion in the sand bay, these figtires may bo ' rather too high, but evidence of the passage of pro- pellent gases through the plugs was afforded by the fact
- 21Ц. - that along all the holes there was a slowly-etching hardened layer, which appeared to be of a constant depth of 0.0022 in. throughout the length of the plugs. In order to ascertain whether there had been any escape of gas past the driving band, the surface of the shot was examined by etching, but there was no trace of a hardened skin due to gas-washing. ^Thus, these ex- periments, while they show that gas passes through a 2-mm orifice in the proof shot at some tine during its passage down the bore, fail to indicate (in thg case of a new 18-pdr. gun, at least) that there is any escape of propellent gases past the driving band during the initial stages of motion. 118. Gas leakage in a worn gun As a continuation of the experiment just described, Greaves, Abram and Rees sought for evidence of gas leakage in a worn gun. They reported as follows: Some Ц-in. proof shot fired in a worn Ц-in. Mark V gun (mean wear at 1 in. from commencement of rifling, 0.039 in.) were examined. One of these showed a hardened skin (maxi- mum thickness, 0.0010 in.) produced by gas-washing for- ward of the driving band. The driving band itself showed some signs of gas-washing. Four other shot fired in the same gun showed no hardened skin on the side of the shot. It seems, therefore, that the formation of a hardened skin, indicating gas-washing, on the projectile forward of the driving band, is not a necessary consequence of the worn condition of the bore (except possibly when scoring is already very pronounced), but is fortuitous and due to occasional faulty ramming or bad centering of the projectile in the gun. In a worn quick-firing gun firing fixed ammunition, however, the same effi- ciency of obturation could not be obtained. 119. Undersize rotating bands In a recent experiment with rotating bands, projectiles having bands that differed in diameter were fired with single-base powder from two new 37“ьтп guns. After 800 rounds, each gun had eroded only very slightly; there was an increase in diameter of 0.005> in.
- 215 - across the lands at the origin. The advance of the forcing cone was also essentially the same '— 0.035 in. for the large bands and 0.0Ц5 in. for the small ones. This result, which by itself would have been inconclusive, assumes new significance upon com- parison with the result obtained when some of the same projec- tiles having rotating bands of normal diameter were fired from another new gun with double-base instead of single-base powder. At the end of 800 rounds this gun had eroded O.Olh in» at the origin and the forcing cone had advanced 0.20 in. Hence it may be concluded that the use of the small-diameter rotating bands, the diameters of which were slightly less than that of the grooves, had far less effect on the erosion of one of these 37-mm guns than the use of double-base powder compared with single-base [Ord. Dept., 19Ы]. 120. Obturation by bullets The obturating action of a small arms bullet is different from that of a banded projectile, for the whole jacket of the bullet acts as the rotating band. In the case of a lead-core bullet, obturation is aided by the ’’upsetting" of the bullet, that is, the increase of diameter caused by the sudden application of the accelerating force of the powder gases. The increased erosion that armor-piercing bullets are supposed to cause may be due to the lack of upsetting and hence to poorer obturation. The use of a bullet that is a force-fit in the throat of a rifle just behind the forcing cone is reputed to eliminate erosion.
- 216 - Mr. Niedner, of the Niedner Rifle Manufacturing Company, once told Mr. J. C. Gray, Ordnance Engineer of the Ordnance Depart- ment, that Mr. Mann, the amateur rifle expert, fired such bullets about 191$, and found nb erosion of a caliber .2$ rifle after 1f>,000 rounds.!^/ It is not known whether a control test was made with normal bullets, and so it is questionable whether the erosion of a rifle of this small caliber could have been detected after that number of rounds. 121. Gas checks Some experimental firings have been made with projectiles having special obturators or gas checks attached to them. Thus different modifications of a device invented by J. H. Brown were tested at Aberdeen Proving Ground from 1907 to 1911 л without any marked decrease in erosion having been noted. Another obturator that was tested at Aberdeen was reported [Feltman, 1929] to have increased rather than decreased the erosion. The exact design of the device was not given in the report of the firing, but it seems to have consisted of some sort of cardboard wad at the base of a 3-in. projectile. The increased erosion was ascribed to the 11 sandpapering effect’* of the bits of cardboard that were blown down the bore. 122. Rotating band with forward gas seal If obturation by the rotating band is important, 'then a very sensible suggestion concerning the position of the gas seal is 103/ J. G. Gray, personal communication, 19Ц1.
- 217 - that made Recently by Commander F. G. L. Johnson, (British) Royal Navy (retired). In the ''J” band that he has patented, Johnson [19Ц2] claims that the lubricated seal formed by the front end of the rotating band is unaffected by the dilation of the bore of the gun by the gas pressure. This seal is obtained by having several narrow cannelures which are cut in the front end of the\band filled with a heavy grease ("Tropilene E," melting point 32O°F). The cannelures are close together, so that ’'when the shell is loaded, the front tiro or three ribs [of copper that separate the cannelures] are deformed to the rear due to contact with the band seating, and grease is extruded, thus creating a gas-tight joint. As the lands cut through the ribs and grooves during engraving, they force the grease in a circumferential direction, which action has the bene- ficial effect of forcing the portions of ribs located in the rifling grooves to the front, thus keeping them in close contact with the bottom of the groove to maintain the seal." Commander Johnson stated that firing tests, of which he gave no details, have sub- к stantiated his claims, and that the British Ordnance authorities are giving serious consideration to his proposal. Somewhat the same end is achieved by certain features of the so-called ”R. D. System of Rifling” (Sec. 126), which has been developed recently by Colonel G. 0. C. Probert and others in the Research Department of Woolwich Arsenal. A projectile having two bands is used and the forcing cone is moved forward slightly. The rear band, which fits tightly in the smooth cylindrical throat of the gun, is supposed to seal the bore while the front band is being
- 218 - engraved. Then, after the rear band has been engraved, the front band can continue to make the gas seal even if the bore is dilated at the rear band. The R. D. System of Rifling has been tried recently with considerable success in the reduction of erosion. Thus, in two quick-firing 3. 7“in« antiaircraft guns with muzzle velocities of about 3h00 ft/sec the maximum erosion in the one of conventional design after 80 rounds was O.O73 in., whereas in the modified gun it was only 0.0Ц8 in. after 80 rounds. The other feature of this design, as will be described in Sec. 123(d), is that the effect of erosion on ballistic performance is mini- mized [Naval Intell., 19Ц2].
- 219 - CHAPTER XIV. LIFE OF GUNS 123. The life of a gun is not a fixed mathematical quantity but is a statistical average, analogous to a life expectancy for a human being. Because of variations in the manner of their use, several guns of the same kind will have widely different lives. Some of these variations, such as rate of fire, amount of charge and kind of powder, have already been discussed in Chap. XII, but it is likely that there are other less well-defined conditions that also affect the life of guns in service. 12Ц. Formula for equivalent service rounds One of the most important variations of firing conditions is the amount of charge, because scarcely any erosion occurs when a gun is fired with a charge considerably below the "service charge” for which it was designed. Sometimes at the beginning of a firing a reduced charge is used for the first round or two, just for the purpose of warming the gun. In the U.S. Navy, for the purpose of tabulation, such warming rounds or any other round fired at other than the standard charge are converted into equivalent service rounds — E.S.R. — by means of the formula, E.S.R. = (W/We)^ (XlV-la) Ь where W is the weight of the charge actually used and W_ is the weight of the service charge.
- 220 - Another formula for the same purpose was developed by Feltmanl2k/ empirically on the basis of the values of E.S.R. for various charges of nitrocellulose powders as given in various British firing tables and then checked against the values of E.S.R. used in the U.S. Navy. This formula, which is intended to take into account the effect of the reduced pressure that accompanies the decrease in weight of charge, is p E.S.R. = (V/V_), (XIV-1b) where V is the muzzle velocity at which the gun was fired, V_ is the standard muzzle velocity, and g is an empirical constant taken equal to 7 when V/V„ >0.6 and equal to 5.5 when V/V_ < 0.6. Both these formulas indicate a large increase in the life of a gun when it is fired with reduced charges. For example, during the accept- ance test of some 37-mm powder a charge of Ц.00 oz gave a maximum pressure of 21,500 lb/in? and a muzzle velocity of 2339 ft/sec, whereas a service round required Ц.70 oz of the same powder to give the standard velocity of 2596 ft/sec (nominally 2600 ft/sec) with a maximum pressure of 28,600 lb/in? The round fired at reduced charge, therefore, was 0.Ц E.S.R. according to Eq. (XIV-1a) and 0.5 E.S.R. according to Eq. (XIV-1b). Correspondingly, a round fired in the same gun with a charge of Ц.oz of the same powder, which gave a muzzle velocity of 268Ц ft/sec at a maximum pressure of 31jU50 lb/in?, was equivalent to 1.35 E.S.R. according to 10h/ S. Feltman, Ballistics Section, Ordnance Department, personal-communication.
- 221 Eq. (XIV-1a) and to 1.26 E.S.R. according to Eq. (XIV-1b). Thus Eq. (XIV-1b) indicates less of a variation from unity than does Eq. (XIV-1a), for charges either more or less than standard. 125. Criteria for end of life Another complication in considering the life of a gun is that there is no unambiguous point in its history that is considered the end of its life. As erosion proceeds during use, the muzzle velocity gradually decreases, because of the advance of the forcing cone [see Sec. 125(d)], and also the dispersion of the. shots at a given target increases. However, the decision as to just how much of a decrease in muzzle velocity or how large a dispersion can be tolerated is purely arbitrary. Moreover, this decision depends to a certain extent on the tactical purpose of the gun. (a) Loss of muzzle velocity. — Some writers have suggested that the time when a gun has lost 10 percent of its initial muzzle velocity should be considered the end of its life. This criterion, according to Lintes [1935] corresponds to a loss of 50 percent in precision and an appreciable diminution of range. A ‘somewhat more severe criterion--—/ is actually used now by the Small Arms Division of the Ordnance Department with respect to the- caliber .50 machine gun. When the muzzle velocity of this gun — which is 2800 ft/sec when the gun is new — has decreased 200 ft/sec, the gun is con- sidered to be at the end of its accuracy life. This usually occurs ЮЦа/ Capt. С. E. Balleisen, personal communication, August 19h2.
- 222 - after about 8000 rounds. Further firing would be possible, but the sights would have to be adjusted for the decreased velocity, which causes the gun to hit below the target. (b) Dispersion in range. — In the U.S. Navy the mean dis- persion of five successive shots fired at a target at a given range is usually used as the basis of the criterionl2£/ for determining the end of life of a gun. Yilhen this mean dispersion exceeds 0.5 percent of the range for a major-caliber gun — 8 in. and larger — or 1 percent for a smaller gun, the end of accuracy life is usually' considered to have been reached. There is no regula- tion, however, on this question, and not all guns are rated in this way. (c) Enlargement of the bore. — The loss of muzzle velocity, and hence the life of a gun, has frequently been correlated with the extent of erosion as expressed by the increase in diameter of the lands at the "origin of rifling" (really at the front end of the forcing cone) or at a point 1 in. in front of the true origin of rifling. The decrease in muzzle velocity for most guns is a linear function of the increase in diameter, the constant of propor- tionality being a function of the caliber and muzzle energy of the gun. Manning [1 919], for instance, found that in the expression, AV = 100 Ky, (XTV-2) where AV is the loss of muzzle velocity in feet per second and £ is 105/ Lt. B. D. Hills, personal communication, August 19h2.
- 223 - the increase in diameter of the lands in inches at a point 1 in. in front of the origin, the empirical constant К varied from 2.3 to 12.1 for eight guns of different calibers, being larger for small calibers. Practical use is made of this relation at proving grounds in connection with so-called "work guns," which are used for acceptance tests of powder. The end result desired is the relation between muzzle velocity and weight of charge when the powder is fired in a new gun. A given weight of charge fired in an eroded work gun gives a somewhat lower muzzle velocity than would have been obtained in a new gun. The muzzle velocity characteristic of a new gun is obtained by an extrapolation based on the measured erosion of the work gun and the relation between muzzle velocity and the increase in diameter at the origin. The end of life of some U.S. Navy guns is considered to be reached when the enlargement at 1 in. in front of the origin of rifling has reached a certain magnitude. For 1 Ц-in./50-caliber guns the permissible maximum enlargement is 0. Ц in. For such high- powered guns the application of this criterion is complicated by the fact that they are subject to considerable muzzle erosion, which also affects the accuracy .life; and therefore, as already mentioned in Sec. 98, a duplex criterion involving erosion at both ends of the bore is used. (d) Advance of the forcing cone. — Another direct measure of erosion that has been correlated with gun performance is the.advance of the forcing cone (Sec. 10) as measured with a plug gage. This
- 22Ц - measurement is much simpler than that of the increase in diameter across the lands and therefore is more readily performed in the field. The ЦО-mm Bofors gun now being used by the U.S. Services, for instance, is supplied with such a gage. The advance of the forcing cone is reflected in the perform- ance of the gun. In an eroded gun using separate-loading ammunition the projectile can be seated correspondingly farther forward in the bore. Thus for some worn 155-mm guns, G.P.F., when the forcing cone had advanced h.O in., the seating had advanced about 2.5 in., and when the forcing cone had advanced 8.0 in., the seating had advanced 6.75 in. [Weyher, 1939)• The decrease in the density of loading that results from the advance of the seating of a separately loaded projectile causes a decrease in the pressure and velocity. The re- duced travel of the projectile also decreases the velocity slightly. In a case gun the projectile remains in the cartridge case until the powder is ignited; but as soon as a little pressure is developed, the projectile is able to move forward nearly the whole distance that the forcing cone is advanced. This free run-up has about the same effect on the maximum pressure and the muazle velocity as the advance of seating in a bag gun. In addition, it tends to cause stripping of the rotating band [Sec. 125(e)]. The relation between loss of muzzle velocity and advance of the forcing cone is nonlinear for most guns; but in spite of this, the advance of the forcing cone can be used as a criterion of the end of life. This was done in World War I by the French and sub- sequently by the A.E.F. when using French guns. For the 155-nm gun,
- 225 - for instance, the criterion was an advance of the forcing cone of 150 mm. It had been found that beyond this point, which was reached after 3500 + 500 rounds, a maximum of Ц00 more rounds could be fired [Ord. Dept., 1920]. This criterion was confirmed by a later study of the characteristics of a number of 155-mm guns at different stages of erosion (see Table XVI, Sec. 61). (e) Stripping of the rotating band. — Any arbitrary criterion of the life of a gun, such as the enlargement at the origin or the advance of the forcing cone, has the disadvantage that frequently during actual service a gun can still be fired after the measured erosion indicates that it has reached the end of life. The gun will not perform as satisfactorily as a less worn one, but it will be much better than no gun at all. This situation occurred during World War I, when the French twice found it expedient to relax the regulations concerning the replacement of 75-nim guns in the field. Such continued use of an eroded gun cannot go on indefinitely, however, when the gun fires fixed ammunition. Eventually the fbrcing cone is advanced to such an extent that the projectile attains so high a velocity by the time its rotating band hits the forcing cone that the upper surface of the band is sheared off by the rifling and the projectile is not rotated. When this occurs the projectile tumbles in flight. This stage of erosion of the- gun truly represents the end of its life, for accurate firing is no longer possible. It has frequently been used as the criterion of the end of accuracy life in erosion tests of small arms, the occurrence of keyholes in a close target giving the indication of when it happens (see Sec. 82).
- 226 - (f) Economic life. — An artificial criterion of the end of life of a gun has been considered in at least two recent erosion studies of army guns, namely, those of the 155-mm gun (Sec. 61) and of the h.7-in. gun [Sec. 105(b)]. This criterion, which is called the economic life, is defined as the maximum number of rounds that can be fired at a minimum cost per round, taking into account the replacement cost of the gun and the increasing cost of ammunition per hit as the accuracy decreases because of erosion. This criterion is without real significance, because it presupposes an abundant supply of guns. In effect, the replacement cost of a gun .is infinite during an engagement when the replacement gun.is not available. 126. The R.D. system of rifling and banding^-^^ The basic feature of the special design of rifling and of pro- jectile banding which is known as the '*Probert!* or !,R.D. System’* is an attempt to prolong the accuracy life of a gun by minimizing the effects of erosion as well as by reducing the erosion itself. To this end the.original contour of the bore is made to approximate that of a somewhat eroded barrel and then the projectile is designed to function well in such a bore, by reducing the extent to which its resistance to movement at the beginning of travel changes as erosion proceeds. 106/ See Sec. 136 for the decrease of muzzle erosion observed with the R.D. system.
- 227 - This is accomplished by a combination of (i) making the initial resistance to Movement as small as possible in a new gun and (ii) making the proportional decrease of resistance for a given increase of wear as small as possible. The former end is achieved by making the slope of the forcing cone as small as is consistent with .the need for sealing the gas as quickly as possible and by reducing to the minimum the quantity of copper that has to be displaced when the shot starts to move. The latter end is achieved by making the diameter of the.upstanding copper in the rotating band exceed that of the bore by the greatest practicable amount in order to dwarf the effect of the wear in the bore on the rate at which this copper is displaced after shot travel^ An attempt is made to reduce the erosion itself both by improving the gas seal (Sec. 122) and by placing the origin of rifling farther from the mouth of the cartridge case. The few trials that have been made of this system have shown very favorable results. Thus, in the two quick-firing 3.7“in. guns mentioned in Sec. 122, there was a marked difference in the effect of erosion on muzzle velocity. Whereas in the conventional gun the velocity had decreased ft/sec after 81 rounds, in the modified gun it actually increased 63 ft/sec after 80 rounds, even though the bore of the latter gun had been increased two-thirds as much as that of the former
- 228 - 127. Navy-formula for the life of guns-— A general formula for erosion in U.S. Navy guns was proposed by Captain G. L. Schuyler [Bur. Ord., 1928]. As a preliminary step to developing tabulated numerical functions of the several variables on which the life of a gun depends, he expressed the average bore enlargement per round, E, during the first 100 rounds by the equation log E =log А- 1.5Ц log 1 -16.Ц log d+12.0 log V+6.0 log M, (XIV-3) where E is measured in inches at a point 1 in. in front of the origin of rifling, 1 is the length of the gun in calibers, d is the diameter of the bore in inches, V is the muzzle velocity in feet per second, M is the projectile weight in pounds, and A is an empirical con- stant. This constant and the other four in the equation were evalu- ated from data for 12 guns that ranged in caliber from 3 to 16 in., in length from 23 to 103 calibers, in muzzle velocity from 2200 to Ц800 ft/sec, and in projectile weight from 10 to 3100 lb. The observed bore enlargement of naval guns at the end of life was combined [Bur. Ord., 1935] with Eq. (XIV-3) to yield the following formula for the accuracy life of a gun in E.S.R.: N = 0.1080 d2/3 E"1 (XIV-U) This formula was subsequently modified [Bur. Ord., 1939b], in order to take into account recently acquired experimental evidence of the 10?/ See Sec. 130 for a comparison of this formula with others.
- 229 - lack of effect of the weight of the projectile on erosion when the muzzle velocity is constant and to permit the calculations to be - performed by means of equations, instead of tabulations. The modi- fied formula is log N = 6.35 + 0.03 1 - 0.82 log d - 0.001611 Vs, (XIV-5) where Vs = Из = ’И3, and the other symbols have the same meaning as in Eq. (XIV-3). Values of the accuracy lives of U.S. Navy guns of various calibers computed according to this formula agreed within about 15 percent on the average with estimates made by the Naval Proving Ground, based on firing experience. Values of N calculated from Eq. (XIV-5) were arranged in the form of an alinement chart [Bur. Ord., 1939b] by means of which the calculation can be carried out quickly with an accuracy of about 2 percent. 128. Kent1 s formula for the life of guns-^^/ The following formula for the life of guns, which is dis- tinguished from the Navy formula in that it involves the maximum pressure, was developed by Kent [1939]г N = AdaKbPc, (XIV-6) ' 108/ See Sec. 130 for a comparison of this formula with others.
- 230 - where N is the accuracy life in rounds, d is the caliber in inches, k[= iH2] is the muzzle energy in foot pounds, P is the maximum pressure in pounds per square inch, and A, a, b and c are empirical constants. These constants were determined by the method of least squares from data for a number of Navy guns. When the values thus computed are used, the formula becomes N = 10.26 x Ю1? d3«$7S p-1 «761. (XIV-7) Kent used this formula to calculate the accuracy lives of a number of Army guns and then compared the values so determined with those given in the Ordnance Proof Manual [Ord. Dept., 1936]. The cal- culated values were consistently lower than the latter estimates. Kent [1939] suggested that ’’this discrepancy arises mainly from failure to allow for the rounds fired with reduced charges in the estimated lives of the guns given in the Proof Manual, ’* 129. Other formulas for the life of guns (a) Jones1 formula?-—— One of the earliest attempts to develop a formula for the life of guns was made by Jones [1911] on the basis of an analysis of the heat transferred from the powder to the bore surface in the region of the position of maximum pres- sure. He proposed the relation, A (XIV-8) V2 d(d- 2)p1‘7 108а/ See Sec. 130 for a comparison of this formula with others.
- 231 - , where the symbols have the same moaning as in Eq. (XIV-6). Jones proposed the value A s 1,2Ц x 10^ for guns firing cordite M.D.; Kent [1939] has remarked that the French have used a value of A that is 5>0 percent greater for guns firing nitrocellulose powder. These values of the constant are for the purpose of giving an estimate of the probable life before relining in peacetime, Jones having suggested that the actual life of serviceability would be 30 to ЦО percent greater. Even the lower value of A, however, gives high estimates of life, as is shown in Table XXVIII, Sec. 130. (b) Justrow* s formula. — As a result of a detailed considera- tion of the many factors' that affect the life of guns, Justrow [1923, Eq. 20] proposed the following formula for the number of shots that can be fired from a gun before the initial muzzle velocity is de- creased by 10 percent: N = (xy/d2)(Av71)(G€/Ak), (XIV-9) where x is an empirical function of the maximum pressure, £ is an empirical function of the caliber, d is the diameter of the bore, A is a factor of proportionality for expressing the weight of the projectile in terms of the cube of the caliber, V is the muzzle velocity, and 1 is the length of the tube in calibers; G6 is propor- tional to the tensile strength of the tube material and дек to that of the driving band, but the exact physical significance of these symbols was not stated. Justrow published graphs for x — which varied from 0 to 1.2J? — and for — which varied from 0 to 1.0 — but did not tell how they
- 232 - were derived. He gave a table [Justrow, \-923, Table VI] showing calculated values for the lives of guns of different calibers and lengths, having different maximum pressures, velocities and weights of projectiles; but he gave no indication of how closely these values agreed with experience. (c) The formula of Lintes. — Another attempt to include the physical properties of the metal of the tube as factors in the life of a gun was made by Lintes [1935, Eq. (28)]. Adopting the criterion that the end of life is characterized by a 10-percent loss of muzzle velocity, he deduced that N = A(^ Ao)(d2/K)(R2/e), (XIV-10) where A is a factor of proportionality, w/qo is the ratio of the weight of charge to the initial surface of the charge, d is the diameter of the bore, K[= isMV2'] is the kinetic energy of the projec- tile, R(kg/mm2) is the elastic limit of the metal of the tube, and e is its modulus of elasticity. Lintes found empirically that A is 700 for bag guns and 350 for case guns; and since w/"l0 usually is 0.5, the formula is simplified to N = 175 d2R2/Ke for case guns. He stated that in rapid fire, N is reduced in the ratio no/n, where n0 is the normal rate of fire and n is the increased rate. The examples of the application of- this formula which Lintes cited showed reasonably good agreement between the calculated and the observed values for the length of life.
- 233 - 130» Comparison of estimates of lives of guns The lives of a few typical unplated guns firing nitrocellulose powder have been computed for the present report by Eqs. (XTV-5), (XIV-7) and (XIV-8), with the results listed in Table XXVIII. The Navy formula [Eq. (XTV-£)] was not intended to be used for calibers smaller than 3-in.; and Jones formula [Eq. (XIV-8)] obviously cannot be used- for small calibers, because of the term (d-2). Similar calculations using Eqs. (XIV-?) and (XIV-10) were not made because of lack of data concerning the physical properties of the gun steels. The remarkably close agreement between experience and the estimates for the life of the Ц.7-in. gun T2 given by Eq. (XIV-5) and (XIV-7) is probably a coincidence, inasmuch as the two equations show systematic deviations from actual values for different calibers. Both of them tend to give somewhat low estimates, those computed by Kent's formula [Eq. (XIV-7)] being in general lower than those com- puted by the Navy formula [Eq, (XIV-5)]. Except for the l5>3-mm, 1Ц-in. and 16-in. guns, the estimates by the Jones' formula [Eq. (XIV-8)] are much too high.. All three formulas indicate the general inverse dependence of the life of a gun on the muzzle energy. However, the exact functional relation between them and the extent to which it is modified by other factors are not adequately expressed by any of these formulas.
Table XXVIII. Calculated accuracy lives of guns. Gun Velocity V (ft/sec) Pressure P (lb/in?) Caliber d (in.) Length 1 (calibers) Proj.Wt. M (lb) Muzzle Energy, E (ft lb) Accuracy Life, N (rds) Equation Number Exp.* xiv-5 XIV-7 XIV-8 Caliber . 50 2800 Ц8 000 0.5 72 0.101 1.2Ц x 10U — 51 7Ц — 8000 37-mm M1A2 2600 36 000 1.U57 53-5 1.3U 1.38 x 105 — 6161 — 3000 to 3500 75-mm M1897 1950 36 000 2.95 ~ 3U.5 1Ц.6 8.67 x ю5 U365 31190 20850 10000 14.7-in. T2 3150 38 000 U-7 6o 5o.o 7.76 x Ю6 398 I4OO 1613 500 to 600 5-in./25 Mk X 2175 38 000 5.0 25 51.7 3.96 x 106 2203 1669 2862 2500 155-mm G.P.F. 2Ц00 28 000 6.1 38 95.8 8.62 x 106 1995 Ш52 2370 . I4OOO 8-in. Mk VI 2750 38 000 8.0 U5 261.8 3.09 x 107 518 226 507 — 8-in. Mk IX 2800 38 000 8.0 55 260.0 3.18 x 107 269 253 560 14OO lU-in./50 Mk IV 2800 38 000 H4.O 5o H4OO.O 1.71 x 108 230 91 II4O 175 16-in. M1919M3 2700 38 000 16.0 5o 23U0.O 2.67 x 108 166 77 12Ц 175 w This column lists the life based on experience, according to the following sources of information: caliber .$0, Capt. 0. E. Balleisen, personal commtmication; 37-mm, Ord. Dept. [ 19U1]; 75-mm, Ord. Dept. [1936 J; Ц.7-in., S. Feltman, personal communication; 5-in./25 Mk X, Bur. Ord. [19h2]; 155-mm, Weyher [1939] 8-in. Mk IX and ТЦ-in., Bur. Ord. [191|2]; 16-in., Ord. Dept. [1936].
- 235 - CHAPTER XV. THEORIES OF THE MECHANISM OF EROSION 131. The foregoing account of the erosion of guns reveals that it is a complex phenomenon. The various theories of the mechanism of егпяпрп''09/ are attempts to explain in specific terms the following generalized description of that phenomenons during the firing of a gun a physical or chemical agent, or a combination of such agents, loosens some of the steel on the Surface of the bore and then either the same or some other agent removes it from the bore. Whereas each theory has emphasized the alleged effectiveness of some particular agent or agents, the convincing evidence-adduced in partial confirma- tion of each of them lends weight to the opinion, already expressed by earlier writers [for instance, de Bruin and de Pauw, 1929; Kosting, 1939a] that the true explanation of erosion must take account of more than one causative factor. In the present chapter!-^/ the several theories are classified with respect to the agents proposed, in an attempt to summarize some of the facts presented in the earlier chapters and to point out deficiencies in our knowledge that need to be supplied before the mechanism of erosion will be understood. This classification is most conveniently made with respect to the agents responsible for 109/ A convenient review of the principal theories was given by de Bruin [1930]• 110/ At some time in the future it is intended to prepare a more complete analysis of these theories, with consideration being given to the additional information acquired in recent investiga- tions of erosion.
- 236 - loosening the material from the surface of the bore. After the material has been loosened, it is considered to be either blown out of the bore by the powder gases or else rubbed off the surface of the bore by the rotating band. 132. Melting of the bore surface The quantity of heat liberated inside a gun is a major factor in determining the amount of erosion, as is demonstrated by the fact that in a given gun the amount of erosion is greatly increased either by an increase in the weight of the powder charge or by the use of the same weight of a powder having a higher heat of explosion. It has not yet been possible to express quantitatively either how much of the heat liberated inside a gun is absorbed by each portion ' of the surface of the bore or to what temperature that surface is raised by the heat absorbed. Hence there is yet no proof that the surface is melted — which is the simplest hypothesis to account for the loosening of metal from the surface of the bore — for raising the temperature of the bore surface may simply accelerate some other change. The hypothesis that the surface of the bore is melted has usually been combined with the idea that the loosened material is swept out of the bore by the powder gases. Considerable difference of opinion, however, has arisen as to whether this supposed melting of the surface and removal of the melted layer is performed by small jets of gas escaping past the rotating band or by the main stream of gas moving down the bore behind the projectile. As
- 237 - mentioned previously (Sec. 11Ц), the gas leakage theory has had a long history, beginning with Lanfroy [188$] and continuing with Vieille [1901]> Yarnell [1910,], Bourgoin [1913] and Justrow [1923]» Charbonnier [1908, 1922], Alger [1910] and Jones [1911]j among others, opposed this point of view very strenuously. Greaves, Abram and Rees [ 1929], who also considered that during erosion a layer of molten metal is blown off the bore surface, implied that this was done by the main stream o^ gas. They did not, however, go as far as Charbonnier [1908, 1922] had gone in attributing ' \ erosion to a vena contracta effect of the stream of gas following the projectile. Letang [1922] and also Justrow [1923] calculated that a major part of the heating of the bore surface to its melting point was caused by the heat resulting from the work of engraving the rotating band. Johnson [19Ц2] combined the action of both gas leakage and the main stream of gas in his theory of erosion, according to which the escape of gas past the band heats the skin surface of the bore almost to the melting point before the projectile moves, so that after it has moved, the main stream of gas is able to erode a short section of the bore by fusion. Farther up the bore the surface is not eroded by the main gas stream because it has not been subjected to preheating by gas leaking past the band. This theory has the appeal of a compromise, but it has no more basis in terms of observational evidence than either of the separate theories relating to fusion of the bore surface; as yet there is not even proof that the bore surface reaches the melting’ point.
- 238 133. Abrasion by unburned powder. As part of his theory of the vena contracta, Charbonnier [1908, 1 922] suggested that some of the erosion is caused by the scouring effect of unburned grains of powder which are swirled about by the gases. It is admitted that the powder grains are hard enough to act like a sand blast, but there is no experimental evidence that they do so. Kent [19Ы] pointed out that the escape of gas from the burning grains might form a cushion which would prevent violent con- tact between the grains and the bore surface. He suggested that a test of this idea might be obtained by comparing the rate of ero- sion of a gun fired with strip powder with that of one fired with grain powder of the same composition. 13h. Cracking of the bore surface The severe thermal stresses developed on the surface of the bore during the sudden heating followed by sudden cooling that occurs during firing have been analyzed by Tschernoff [1913], by Howe [1918], by de Sveshnikoff [1922] and by Justrow [1923] (see Sec. 70). Tschernoff in particular considered these stresses as the main cause of loosening material from the bore surface, whereas Howe and Justrow attributed to them merely secondary in- fluence. Similarly, Vieille [1901] had ascribed to the cracks on the bore surface only a contributory effect that made leakage possible.
- 239 - « 139. Chemical reaction between the bore surface and the powder gases Intimately related to the cracking of the surface of the bore is the question of chemical reaction between the powder gases and' the surface. Vieille [1901]} on the basis of a microscopic examination by Osmond [1901 ] of an eroded gun, had concluded that the surfa.ce layer of steel was hardened by carburization, and that it cracked more readily because of this condition. The various investigations by Fay [1916, 1 929], Howe [ 1918Wheeler [1 922], de Sveshnikoff [1929]> Greaves, Abram and Rees [1929], Lester [1929a], Snair and Wood [1938] and Kosting [1939a, 19hQ] on the possibilities of the formation of a hardened layer on the surface of the bore by reaction with the powder gases have been summarized in Secs. 79 to 79. In spite of those extensive investigations, we are still ignorant of the nature of the altered layer on the surface of the bore, and . therefore we cannot say definitely how great an influence chemical reaction between the bore surface and the powder gases may have on erosion. An entirely different type of chemical reaction was suggested by Cole [19Ц1]. He assumed "that gases under extreme pressure and high temperature approach the characteristics of dilute solutions" and that they are highly ionized, both because of thermal agitation and, because of collisions. The alloying elements in the steel complete a galvanic couple, whereby iron is converted into an oxide film when the highly ionized gas comes into contact with the sur- face of the bore. An earlier suggestion of the possibility of oxidation had been made by Crawford [1937] ? who supposed that the
- 2 ЦО - surface of the bore is covered by an adsorbed ’’film of one or more molecular layers of oxygen, which, when the surface reaches a critical temperature, is released and forms an oxide with the sur- face layer of the metal.'1 The layer of oxide, according to both hypotheses, is swept away by the gases moving with high velocity. 136. Muzzle erosion The foregoing discussion of causes of erosion has been limited to erosion in the region of the forcing cone. The cause of erosion at the muzzle is equally unknown. Three principal theories have been advanced to account for it. According to the first, which has already been mentioned (Sec. h7)j muzzle erosion is simply the result of frictional wear between the projectile and the bore, although why the erosion decreases progressively away from the muzzle is not explained. This point of view seems to be current among students of the subject in the U.S. Army today. The other two theories explain muzzle erosion as caused by the powder gases. Charbonnier [1^22] pointed out that his vena contracta theory of erosion at the forcing cone applied equally well to erosion at the muzzle, where another discontinuity in the gas stream occurs. Thompson [1?Ц1], on the other hand, advanced the theory that muzzle erosion is caused by the jet of high-velocity gas that rushes through the annular orifice formed at the moment the rotating band breaks contact with the bore. The remarkable reduction of muzzle erosion in guns using the R. B. system of rifling (Secs. 122 and 126) seems to discredit both the theories of frictional wear and of the vena contracta effect.
- 2h! - In the R. D. system [Naval Intell., 19Ц2] the grooves decrease gradually in depth toward the muzzle, and for the last three calibers of its length the bore is smooth. The original purpose of this change was to erase the rifling marks from the rotating band in order to decrease the air resistance of the projectile in flight. Limited tests of guns rifled in this way, however, have given the unexpected result of almost complete absence of muzzle erosion. Thus the modi- fied 3.7-in. quick-firing gun mentioned in Secs. 122 and 126 showed no erosion at the muzzle after 80 rounds, whereas the conventional gun fired under similar conditions showed an increase of diameter beginning about 27 calibers from the muzzle and increasing almost linearly to a maximum of 0,018 in. at the muzzle. 137« Summary The one single feature of the mechanism of erosion to which all investigators have subscribed is that it is a dynamic phenomenon. Erosion occurs only when the bore surface is attacked by hot powder gases in motion, as was pointed out by Sir Andrew Noble [1906], one of the earliest students of the subject. Therefore, in order to achieve a true explanation of erosion, it will be necessary to under- stand in detail the physico-chemical changes that the material of the bore undergoes as it reacts with a rapidly changing environment during firing.
2h3 - BIBLIOGRAPHY AND AUTHOR INDEX— for PART TWO A. 0. [19U2a] (Sec. 101) "Repair of guns by electrodeposition. Summary of positions, March 31, 19Ц2," Adv. Council Sci. Res. Tech. Dev. to British Ministry of Supply, A. C. 1965 (Apr. 16, 19h2). A. C. [19U2b] (Sec. 101) "Repair of worn-out gun barrels by electrodeposition. Note on position on May 20, 19Ц2," Adv. Council Sci. Res. Tech. Dev. to British Ministry of Supply, A. C. 217h. Alger, P. R. [1910] (Sec. 11Ц, 132) "Gun erosion," reprint U.S. Naval Institute Proc. 36, 697-705 (1910)5 Ordnance Dept. Library. Altshuler, Bernard [ 19111 ] (Sec. 105b) "Report on the firing test to determine the limits of accuracy life of a h.7~in. gun T2," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 25Ц (Sept. 10, 19hl). Ames, T. L. [1921] (Sec. 92) "Preliminary report on erosion tests of Browning machine-gun barrels. Rept. No. 9 — The manufacture, physical and chemical properties of machine-gun barrels made of invar steel," Springfield Armory; manuscript (Feb. 8, 1921), 2 p.; Ordnance Dept. Library. Benson, A. E. [19U1] (Sec. 100) Remarks at conference on gun erosion, Watertown Arsenal, Watertown, Mass. (Oct. 15, 1911.1)- Report of conference, Watertown ArseYial, p. 11. Bihlman, V. W. [1921] (Sec. 71, 7Ц, 75, 77) "Preliminary report on erosion of Browning machine-gun barrels. Rept. No-. 8 — Metallographic examination after firing of Browning machine-gun barrels made of twenty-one special steels," manuscript (Jan. 19, 1921), Ц p.; Ordnance Dept. Library. Bone, W. A., D. M. Newitt and D. T. A. Townsend [1929] (Sec. 89c) Gaseous combustion at high pressures, London: Longmans, Green (1929), 396 p. 111/ The section number given on the same line as the author's name refers to the present report.
- 2ЦЦ - Bourgoin [1913] (Sec. 11Ц, 132) Mem. artillerie navale 7, 3h3 (1913)- Browne, C. A. [1939] (Sec. 73) ’•Charles Edward Munroe: 18Ц9“1938," J- Am. Chem. Soc. 61, 1301-1316 (1939). Bruin, G. de [1930] (Sec. 131) l,La poudre sans fumee et 1*erosion des bouches a feu. I. Court apergu historique et critique du sujet," Communique de la Ste. Anme^ "Fabrique Neerlandaises d’Explosifs" No. 10 (1930), 39 p. English translation by Lt. Col. Aiken Simons, Mil. Sales Div., E. I. du Pont de Nemours and Co. (Mar. 1931), Ц9 p.; Ordnance Dept. Library. Bruin, G. de and P. F. de Pauw [1928 ] (Sec. Ю9) ’•La temperature de combustion de la poudre sans fumee en fonction de sa composition chimique. II. Partie experi- mentale" [See Moesveld, 1928, for Part I.] Communique de la Ste. Anme. "Fabrique Neerlandaises d*Explosifs" No. 8 (1928); Ordnance Dept. Library. Bruin, G. de and P. F. de Pauw [1929] (Sec. 131) "La temperature de combustion de la poudre sans fumee en fonction de sa composition chimique. III." Communique de la Ste. Anme. "Fabrique Neerlandaises d'Explosifs" No. 9 (1929); Ordnance Dept. Library. Bruin, G. de and P. F. de Pauw [1931] (Sec. 83,86,87,109,110c) "La poudre sane fumee et 1'erosion des bouches a feu. II. Partie experimentale," Communique de la Ste. Anme. "Fabrique Neerlandaises d'Explosifs" No. 11 (1931), 27 p, English translation by M. S. P., Mil. Intell. Div., War Department; Ordnance Dept. Library. Bur. Ord. (Navy Department)[1928] (Sec. 127) "Erosion, a general formula for, in U.S. Naval guns," Memo. from G. L. Schuyler to Chief of Bureau of Ordnance, 372-^/11/77 (Dec. 29, 1928). Bur. Ord. (Navy Department) [1931] (Sec. 83, 87, 9hc) "Erosion plug tests — relationship to gun erosion," Naval Proving Ground letter to Chief of Bureau of Ordnance, S72-U (9)/L3-2(2) (Mar. 30, 1931). Bur. Ord. (Navy Department) [1933] (Sec. 127) "Formula for life of guns," Section "K" Memo, to Chief of Bureau of Ordnance, S72-h(22/39) (Nov. 20, 1933).
- 2Ц5 - Bur. Ord. (Navy Department) [1938a] (Sec. 9hc) ’•Erosion plug tests *-summary of tests 1 929-31,dnd later," Naval Gun Factory letter S7h-1(3137)(K) (Feb, 10, 1938).' Bur. Ord. (Navy Department) [1938b] (Sec. 98) "Experimental firings of 1800- and 20OO-lb projectiles' in 1h-in. guns; partial report covering the firings scheduled in 1h-in./ЗО-caliber Mark C gun, No. 110L2," Naval Proving Ground letter to Chief of Bureau of Ordnance, S78-I(ЗЦ-Th-in.) L5-2(B6h37) (Apr. 27, 1938). Bur. Ord. (Navy Department) [1939a] (Sec. 98) "Life of gun program with 1h-in./30-caliber Mark C gun, Nd. 110L2," Naval Proving Ground letter to Chief of Bureau of Ordnance, S72-h(m-in.) (B86h7) (Mar. 31 , 1939). Bur. Ord; (Navy Department) [1939b] (Sec. 127) "Suggested modification for generalized erosion formula," Bureau of Ordnance Memo., T11-3 (Sept. 29, 1939). Bur. Ord. (Navy Department) [19h2] (Sec, 130) "Table of estimated lives of U.S. naval guns," Washington, D.C.: Bureau of Ordnance, O.D. No. h398 (Apr. 23, 19h2), Ц p. Bur. Stand. [192ЦЬ] (Sec. 92) "Report on Investigation of machine-gun barrels manufactured from special steels," Bureau of Standards and Ordnance Department joint investigation; manuscript (June ,192Ц), about 100 p.; Ordnance Dept. -Library. , Burlot, E. [1923] (Sec. 89a) "Note sur la mesure des resistances opposees par les metaux a des deformations rapides. Application a la mesure des pressions developpees en vas clos par la combustion de la poudre," Mem. poudres 20, 233-322 (1923); Ch. Abs. 18, 2077. Caldwell; F. R. and E. F. Fiock [19Ul] (Sec. 89c) "Some factors influencing the performance of diaphragm indicators of explosion pressures," J. Res. Bur. Standards 26, 173-96(19^1). Charbonnier, P. [1908] (Sec. 132, 133) Balistique interieurs, Paris: Doin(l908). Charbonnier, P. [1922] (Sec, 11Ц, 132, 133, 136) "La veine gazeuse," Mem. artillerie francaise 1_, 1001-11 (1922).
- 2Ц6 Cole, T. S. [19Ы] (Sec. 135) "Electrochemical theory of erosion of gun barrels," manuscript (no date), Ц p.; manuscript (no date), 2 p.; comments by Ordnance officers; Ordnance Dept. Mail and Record Section, File 0.0. Ц00.111/Ц161, T. S. Cole. Crawford, D. J., Jr. [1937] (Sec. 135) "The applicability of photography and spectroscopy to some problems of interior ballistics," Watertown Arsenal Rept. No. 122/9 (1937), 18 p. and bibliography, 5 p. Cranz, C. [1926] (Sec. 102) Lehrbuch der Hallistik, 3 vols., Berlin: Teubner (1913). de Bruin, G. See: Bruin, G. de de Sveshnikoff, W. W. See: Sveshnikoff, W. IV. de Desmazi&res [1922] (Sec. 89a) "Mesure des pressions absolues au moyen des crushers," Mem. artillerie fran^aise 1_, 777-826(1922). DuMond, J. W. M. [19U1] (Sec. 89c) New equipment for measuring and recording the pressure of powder gases in rocket chambers and the thrusts exerted by blocked rockets, each as a function of time, NDRC Memos. A-28M to A-30M (OSRD No. 30), bound as one (Oct. 9, 19L.1)• Eckhardt, E. R. [1922] (Sec. 89b) "A precision high-speed oscillograph camera," J. Franklin Inst. 19Ц, 1x9-67(1922). Epstein, Samuel [1936] (Sec. 71) The alloys of iron and carbon. Vol. I. Constitution, New York and London: McGraw-Hill (1936). Fay, Henry [1913] (Sec. 7U, 77) "Tests of trepanned rings from 12-inch gun," Tests of metals, Watertown Arsenal (1913)} p. 132-6 and 6 plates; also published as Appendix II of Report of Chief of Ordnance (1913). Fay, Henry [1916] (Sec. 70, 71, 73, 75, 135) "Erosion of guns: the hardening of the surface," Bull. Am. Inst. Mining Engrs. 56, 2237“51(1916); reprinted in Journal U.S. Artillery h7, 392-ЦО5(1917)j abstracted in Iron Age 98, 1290-3.
- 2Ц7 - - Fay, Henry [1921] (Sec. 76) “Nitrogen and case-hardening," Chem. and Met. Engr. 2-Ц, 289-90 (1921). Fay, Henry [1925] (Sec. 76,’ 135) "The cause of the white layer," Army Ordnance 5, 798-9(1925). Feltman, S. [1929] (Sec. 121) "Effect of obturators and distance wadding," Report of Project KW 366, Ordnance Dept. (Nov.16, 1929). Fleming, J. P. [1918] (Sec. 103) "Erosion tests of high silicon-chrome-vanadium steel in Lewis aircraft automatic-rifle barrels," manuscript, Ordnance Dept. (June, 1918); Ordnance Dept. Library. Fleming, J, P. [1 91 9] (Sec.,97) "Erosion in machine guns and automatic-rifle barrels," letter to Major F. F. McIntosh, Technical Staff, Metallurgical Branch (June 12, 1919)* h p. and bibliography, h p.j Ordnance Dept. Library. Goranson, R. W., Wm. Garten, Jr. and J. A. Crocker [19h2] (Sec. 89a) The measurement of large transient stresses, NDRC Report А-Ц5 (OSRD No. Ц98) (Apr. 6, 19h2), 33 p. ' Grant, G. P. [19Ш ] (Sec. 103) "Erosion of ,50-caliber machine-gun barrels. First report of Ordnance Program 537h,” Aberdeen P. G. (June 30, 19hl), 220 p.; Ordnance Dept. Library. Graziani, M. [1928] (Sec. 76) "Alterazioni dell1 aciaio dei tub! d'anima delle bouche da fuoco," Rivista d'Artiglieria e Genio Ц5, I4.OI4.(1 928). Greaves, R. H., H. H. Abram and S. H. Rees [1929] (Sec. 65, 70, 71, 72, 73, 7h, 75, 8Ц, 85, 87, 93, 106, Ю7, 113, 117, 118, 132, 135 ) "Erosion of guns," J. Iron and Steel Inst. 119, 113-167(1929); includes а Ц-р. bibliography. Guion, J. L., H. G. Carter and N. L. Reed [1937] (Sec. 70) "Miorographic and macrographic studies of a 1.1-in. Navy machine gun No. 73, centrifugal casting C533," Watertown Arsenal Report No. 3lh/533(Mar. 2, 1937). Hardcastle, J. H. [190U] (Sec. '68) "Remarks on rust, wear and erosion in gun barrels, with especial reference to small arms," Proc. Roy. Artillery Inst. XXXI, No. 2 (July 190Ц).
- 2Ц8 - Hatcher, J. L. [1920a] (Sec. 92) "Report on test of 21 Browning machine-gun barrels of dif- ferent steel compositions," manuscript, Springfield Armory (Oct, 9, 1920), Ц p.; Ordnance Dept. Library. Hatcher, J. L. [1920b] (Sec. 92) "Report on test of Browning machine-gun barrels made of invar steel," manuscript, Springfield Armory (Nov. 1920), 2 p. Ordnance Dept. Library. Hickman, C. N.,[1981] (Sec. 89a) The use of copper balls for measuring pressures in combustion chambers, NDRC Report A-21 (OSRD No'. 28) (Nov. 28, 198l), 9 p. Howe, H. M. [1918] (Sec. 68, 70, 7b 75, 138, 135) "The erosion of guns," Trans. Am. Inst. Min. Met. Engrs. 58, , 513-96(1918). Howorth, H. G. [19Ю] (Sec. 70) "The cracking of inner tubes (carbon steel)," Research Dept., Woolwich; R. D. Report No. 11 (19Ю). Inglis, N. P., and W. Andrews [1933] (Sec. 73) "The effect on various steels of hydrogen at high pressures and temperatures," J. Iron Steel Inst. 128, 383~8O8(1933)• Jeansen, 0. F. [1936] (Sec. 102) "A treatise on rifling of guns," Washington, D.C.: Bureau of Ordnance, Navy Department (June 1936), 53 p. Johnson, F.G. L. [1982] (Sec. 122, 132) National Inventors Council Suggestion No. 527, consisting of: (a) manuscripts, "Short outline of Commander Johnson's theory of erosion" and "Description of the 'J* band" (liar. 19, 1982); (b) Correspondence relative to the submission of this sug- gestion. Jones, Brymor [1982] (Sec. 72, 78, 75, 77) "The nitrogen content and constitution of the hard white layer in eroded guns," Woolwich Research Dept, and University College, Cardiff; extra-mural Rept. No. 8(Feb. 1982). Jones, B.-, and H. E. Morgan [1932] (Sec. 77) "Investigations into the nitrogen-hardening of steels," J. Iron Steel Inst. 21, 51(1932).
- 2Ц9 - Jones, H. J. [1911 ] (Sec. 129a, 132) ’’The erosion of gun tubes and heat phenomena in the bore, of a gun," Engineer 111, 29h,. .317, 380, 399(1911). Justrow, Capt. [1923] (Sec. 129, 132, 13U) "Theoretische Betrachtungen uber die Lebensdauer, usw.," Charlottenburg (1923); translation, Ordnance Dept. Library. Karcher, J. C. [1922] (Sec. 89b) "A piezoelectric method for instantaneous measurement of high pressures," Sci. Papers of the Bureau of Standards (Aug. Ц, 1922), 8 p. Kent, R. H. [1926] (Sec. 9hb) "Closed bomb erosion test of bars of special steels," report for Ordnance Program Ц7ЗЗ (Dec. 28, 1926), 2 p. Kent, R. H. [1927] (Sec. 89b) "Time-pressure records by a piezoelectric method," Ordnance Tech. Notes No. 10(1927), 2I4. p.; Ordnance Dept., U.S.A. Kent, R. H. [1938b] (Sec. 89b) "The piezoelectric gage: its use in the measurement of gun-pres- sures," Army Ordnance 18, 281-Ц(1938). Kent, R. H. [1939] (Sec. 128, 129a) "A formula for the accuracy life of a gun," Aberdeen P. G., 1 Ballistic Res. Lab. Rept.. No. 133(1939), 3 p. Kent, R. H. [19Ы] (Sec. 133) Remarks at conference on gun erosion, Watertown Arsenal, Watertown, Mass. (Oct. 13, 19111). Report of conference, Watertown Arsenal, p. 63. , Kent, R. H.,and A. H. Hodge [1939] (Sec. 89b) "The use of the piezoelectric gage in the measurement of powder pressures," Trans. Am. Soc. Meeh. Engrs. 6l, 197-2ОЦ(1939). Keys, D. A. .[1921 ] (Sec. 89b) "A piezoelectric method for measuring explosion pressures," Phil. Mag. Ц2, 1+73-88 (1921). Kosting, P. R. [1931+] (Sec. 87, 91, 93) "Gun erosion," Watertown Arsenal Rept. No. 731/2 (1 931+), 19 p. and bibliography, 8 p.
- 25о - Kosting, P. R. [1936] (Sec. 66) "Pastilles: Partial Report I," Watertown Arsenal Rept, No. 731/1/1 (Jan. 15, 1936), 18 p. Kosting, P. R. [1938] (Sec. 87) "Erosion II. Plug erosion tests," Watertown Arsenal Rept. No. 73l/37(June 15, 1938), 16 p. Kosting, P. R. [1939a] (Sec. 79, 131, 135) "A metallographic and diffraction study ,of erosion of guns," manuscript, 3 p.j Ordnance Dept. Library. Kosting, P. R. [1939b] (Sec. 71, 7h, 78) "The effect of service upon 75-mm gun M1897 No, 1537," //, , I Watertown Arsenal Rept. No. 73T/hU(Aug. 29, 1939). [ ' Kosting, P. R. [19Ц0] (Sec. 70, 71, 72, 7h, 78, 95, 135) "37-mm Browning automatic gun M1, No.-1 barrel," Watertown Arsenal Rept. JJo^JT34-/l;5(Oct. 17, 19h0), 16 p. 7 Kosting, P. R. [19Ц1] (Sec. 79) Remarks at conference on gun erosion, Watertown Arsenal, Watertown, Hass. (Oct. 15, 1 9h1)• Report of conference, Watertown Arsenal, p. 19- Kosting, P. R., H, G. Carter and N. L. Reed [1937] (Sec. 72) "Machine-gun barrel No. 12 from Wright Field," Watertown Arsenal Rept. No. 731/31(Sept. 28, 1937), 18 p. Kosting, P. R. and R. E. Peterson [19h2] (Sec. 65, 70, 72) "37-mm gun M1A2, No. 109," Watertown Arsenal Rept. No. 731/59 (May 29, 19U2). Lane, J. R. [1933] (Sec. 91, 96, 105a) "Memorandum report on test of six 3-in. antiaircraft guns, M1925 — radially expanded," report for Ordnance Program Ц615 (Apr., 2h, 1933), 6 p. Lane, J. R. [1936] (Sec. 93) "The influence of the melting point of the barrel material on erosion, Project RB 152, firing with molybdenum barrels," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 55 (July 16, 1936). Lanfroy, A. [1885] (Sec. 11Ц, 132) "Sur la cause des erosions dans l'Sme des bouches a feu," Mem. artillerie Marine 13, 173(1885).
-251 - Langenberg, F. 0. [1916] (Sec. 71) "The examination of rings from the.bore of 10-in. guns,. No. 21, M1895, M1," Tests of metals, Watertovm Arsenal, ,(1916_), pp. 105-108. Lester, H. H. [1929a] (Sec. 71, 77, 135) "The white layer in gun tubes and its relation to the case of nitrided chromium-aluminum steel," Trans. Am. Soc. Steel Treating 16, No. 5(1929). Lester, H. H. [1929b] (Sec. 73, 77) ".50 caliber nitrided rifle barrel," Watertown Arsenal Rept. No. 3hl(Jan. 21 , 1929), 16 p. Lester, H. H. [1929c] (Sec. 99) "Chrome plated . 30 caliber rifle barrels," Watertown Arsenal Rept. No. 351(Nov.20, 19290, 22 p. Letang, G. [1922] (Sec. 132) "Sur l’usure des bouches a feu," Mem. artillerie francaise 1, 1013-1035(1922). 5 Lintes, J. [1935] (Sec, 125a, 129c) "Etude critique sur l’usure des bouches a feu," Mem. artillerie francaise 1Ц, 175-223(1935); translation in Ordnance Dept. Library. Manning, H. G. [1919] (Se'c. 125c) "Loss of muzzle velocity due to erosion," Aberdeen P. G.; manu- script (Feb. 10, 1919); Ordnance Dept. Library. Marshall, Arthur [1917] (Sec. 106) Explosives, London: Churchill; Philadelphia: Blakiston vol. I and II, (ed, 2, 1917); 795 p.; vol. Ill, (1932), 286 p. Marshall, Arthur [1920] (Sec, 73) "The detonation of hollow charges," J. Soc. Chem. Ind. 39, 35t(192O). Miller, H. W. [1920] (Sec. 66) "Preliminary report on character, cause, rate and effect of the various phenomena of the wear and erosion of guns," manuscript (1920), 180 p.; Ordnance Dept. Library. Miller, H. W. [1921] (Sec. 105d) Railway artillery: a report on the characteristics, scope of utility, etc., of railway artillery. Appendix I. "The German long-range gun," p. 723“U5; Ordnance Dept. Doc. No. 2O3h; Washington: Government Printing Office (1921).
- 252 - Naumann, F. К. [1937] (Sec. 73) "Einwirkung von -Wasserstoff unter hohem Druck auf unlegierten Stahl," Stahl und Eisen 57, 889-899(1937)• Naval Intell. [19h2] (Sec. 122, 136) "'Probert' gun and projectile design," Report of Naval Attach^, London, No. 838(Apr. 6, 19h2). This report contains a memo- randum by Col. Probert, giving his interpretation of gun ero- sion and suggesting a remedy. Naval Ordnance [1912] (Sec. 6Ц) "Erosion and life of guns," Naval Ordnance Pamphlet No. Ц12, Washington! Government Printing Office (Oct. 1912). Naval Ordnance [1926] (Sec. 105c) Reports of special board on Naval Ordnance, Oct. 1 and 15, 1926, concerning 3”in./lO5-caliber gun. Quoted in letter from Capt. G. L. Schuyler to J. S. Burlew, NDRC, Mar. 28, 19^2. Noble, A. [1906] (Sec. 137) "Researches on explosives, IV," Trans. Roy. Soc. (London) 2O6A, Ц$3(190б). Noble, A. [1906 reprint] (Sec. 89a) Artillery and explosives: essays and lectures written and de- livered at various times, New York; Dutton(19^6), 5h8 p. Noble, A. and F. Abel [187$] (Sec. 89a) "Researches on explosives, I," Trans. Roy. Soc. (London) (1875). Reprinted in Artillery and explosives (1906), p. 99-230. Norton, M. R. [1935] (Sec. 71) "Metallographic polishing," Watertown Arsenal Report No. 1ЗО/З (Aug. 19, 1935). Norton, M. R. [1937] (Sec. 71) "Metallographic polishing at Watertown Arsenal," Watertown Arsenal Report No. 130/7 (Apr. 1937)» 0. В. P. [19Ma] (Sec. 108) Ordnance Board (Great Britain), Proceedings No. 136О3 on "Flashless propellants." The substance of this report is found in Naval Intelligence Report, Serial 2681, Monograph Index Guide No. Ц09-Ю0, a report from Naval Attach^ (London) (Nov. 2h, 19Ы).
- 253 - 0. В. Р. [19t2a] (Sec. 101) "Repair of bore by electro-deposition," O.B.P. No. 17,255 (Mar. 11, 19Ц2). О. В. P. [19h2b] (Sec. 69) "Use of tinfoil in ammunition of 37“ппп M.5 tank gun," O.B.P. No. . 17,508; 18,059; 18,060 (May 13-and June 19, 19Ц2). 0. В. P. [19U2c] (Sec.,108) "Q.F. h0-mm guns. Relative lives with cordite W.T. and NH charges," O.B.P. No. 16,363 (Feb. 23, 19Ц2). 0. В. P. [19U2d] (Sec. 106, 108) "Computation of charge values in E.F.C. Formula to' be used for flashless and NH. Application to 6-pr. 7cwt.," O.B.P. No. I8,3h3 (July 6, 19U2). 0. В. P. [19U2e] (Sec. 101) "Repair of bore by electro-deposition. Trial in Q.F. 3.7“ln. barrels," O.B.P. No. 18,832 (Aug. 5, 19Ц2). Ord. Dept. [1907] (Sec. 110a) "Erosion tests of 6-pounder, using slow- and quick-burning powders," Ordnance programs 386, 398 and Ц10; Ordnance Department, Mail and Record Section, File 0.0. 38961/5 and 0.0. 38961/17 (May 10, 1907)» Ord. Dept. [1915] (Sec. 89a) "Pressure gage outfits for determining pressures in cannon with descriptions and instructions," Ordnance Pamphlet No. 1Ю9 (Rev. 1915), 10 p. 4s Ord. Dept. [1920] (Sec. 125d) "Erosion of guns," miscellaneous file, Ordnance Department; 3 vol., pages numbered consecutively (1920); Ordnance Dept. Library, Ord. Dept. [1922c] (Sec. 105b) "h.7-in. experimental gun, No. 808, fired in a program to develop a higher grade of steel for gun construction," firing record of this gun for rounds 22ЦЦ-2312, Ordnance Program Ц282 (June 19-23, 1922). Ord. Dept. [1926] (Sec. 68) "Test of gilding metal rotating bands," second report of Ordnance Program Ц6Ц2 (Aug. 3, 1926).
- 23h - Ord. Dept. [1929] (Sec. 99) ’’Test for determination of corrosion and erosion in chromium- plated barrels for small arms,11 first report of Ordnance Program 1911(Oct. 16, 1929). Ord. Dept. [1932] (Sec. 89a) "Small arms pressure gages (radial crusher type)," Eng. Reprod. Plant, U.S.A.; Doc. No. 722? (Mar. 15, 1932). Ord. Dept. [1936] (Sec. 89a, 10$, 128, 130) Ordnance proof manual, Aberdeen P. G.; 1936 edition; issued from Oct. 193б to Oct. 1938; looseleaf. Ord. Dept. [1911] (Sec. 89a, Ю7, 119, 130) "Endurance test of 37-mm gun M1A2 and powder," eleventh and thirteenth reports of Ordnance Program Ц879 (Feb. 10-23, 1911). Osmond, F. [1901] (Sec. 71, 75, 135) "Note sur la trempe superficielle de l'acier soumis a 1*action des explosifs," Hem. poudres salpetres 11, 211(1901). Paquelier, R. [1929] (Sec. 69) "Usure et desencuivrage des canons," Mem. artillerie francaise 8, 707(1929). Peteval, J. E. [1905] (Sec. 89c) "Pressure of explosion," Trans. Roy. Soc. (London) 203A, 33 7-98(1903). Piantanida, E. [1922] (Sec. 76) "Sulle alterazioni subite dall* acciaio dei tubi anima du cannoni usurati," Rivista Karittima (1922); translation: "Sur les alterations subies par l'acier de l'ame des canons uses," Mem. artillerie francaise 2, 1021(1923). Piantanida, E. [1933] (Sec. 76) "Le alterazioni dell* acciaio dei cannoni usurati dal tiro," Rivista Marittima 15 (1935); translation: "Les alterations de l'acier des canons usee par le tir," Mem. artillerie francaise 17, 137-31(1938). Ritchie, S. B. [1928] (Sec. 73, 91b) "Development of a nonfragmenting gun steel, best suited for cold-working," Watertown Arsenal Report No. 326 (June 6, 1928), 18 p.
- 255 - Rockwell, S. P. [19191 (Sec. 95) "Erosion in small arms," letter to Major McIntosh (Apr.. 18, 1919) t 3 p(; Ordnance Dept., Library. Rosenhain W. and Beliew [1913] (Sec. 75) "Erosion in gun tubes," Proc. Int. Assoc. Testing Materials II, 127(1913)» Discussion at 6th Congress of the Association in New York on Sept. 3, 1912 of paper by Tschernoff [1913]« Russell, H. 17. [19Ц1а] (Sec. 88) "Erosion in gun-barrels," monthly reports of work under contract with Watertown Arsenal; Battelle Memorial Institute; Feb. 19U1 to June 191x2, incl,; p. 1-h30. Russell, H. W. [19Mb] (Sec. 88). Remarks at conference on gun erosion, Watertown Arsenal, Watertown, Mass. (Oct. 15, 19M). Report of conference, Watertown Arsenal, p. 6. . . . Russell,-H. W. [19М] (Sec. 88, 99) "Erosion in gun barrels," monthly reports on Contract 6183 with Watertown Arsenal; first report, July 31, 19Ц2; Battelle Memorial Institute; p. 1-17. Sauveur, Albert [1935] (Sec. 75) The metallography and heat treatment of iron and steel, ed. Ц, Cambridge, Mass.: The University Press (1938)* 531 p. Schwinning, W. [193h] (Sec. 102) Konstruktion und Werkstoff der Geschutzrohre und Gewehrlattfe, Berlin; VDI-Verlag GMBH(193M, 167 p. Sharpe, P. В. [19Ц2] (Sec. 111, 112) Complete guide to handloading, New York and London: Funk and Wagnalls(l9h2), U65 p. and supplement, 72 p. Shumate, P. W. [1938] (Sec. 71, 73, 7h) "Properties of nitrided gun steels," thesis (M.S.), Massachusetts Institute of Technology (1938); Watertown Arsenal Report No. 326/2, 33 p. Siwy, P. [1908] (Sec. 6Ц) “Die Abnutzung der Geschutze und deren Ursachen," Z. ges. Schiess u. Sprengstoffw. 3* h2,66(1908). Snair, W. H., and W. P. Wood [1939] (Sec. 72, 73, 77, 135) "The white layer structure in the erosion of machine-gun barrels," Trans. Am. Soc. Metals 27, 608-21;(1939) •
- 256 - Stefanides, V. See: Snair and Wood [1939) p. 621]. Sveshnikoff, IV. IV. de [1921a] (Sec. yO, 72, 73) '•Some factors affecting the life of machine-gun barrels,4 Bureau of Standards Technical Papers No. 191(1921), 27 p« Sveshnikoff, IV. IV. de [1921b] (Sec.. 73) ’’Life of machine-gun barrels. Part I. Abrasive action of bullet and gases,'1 Army Ordnance 2, 16l (1921). Sveshnikoff, IV. IV. de [1922] (Sec. 70, 97, 13k) ’’Life of machine-gun barrels. Part II. Cracking of the sur- face of the bore,'1 Army Ordnance 2, 3011(1922). Sveshnikoff, IV. IV. de [1925] (Sec. 71, 711, 76, 77, 135) "Carburization as a factor in erosion of machine-gun barrels," Arm-у Ordnance 5, 7911(1925). Sveshnikoff, IV. IV. de [1933] (Sec. 61i, 70) "Data on some special steels for machine-gun barrels," Am. Soc. for Metals 21_, 652-62(1933). Sveshnikoff, IV. IV* de, and H. E. Haring [19211] (Sec. 101) "Electroplating worn machine-gun barrels," Army Ordnance 5, 503-11, 51111(1921). Thompson, L. T. E* [19111] (Sec. 136) Reply to proposal for elimination of leakage past rotating band, letter to P. 17. Bridgman, Harvard University (June 26, 1 9111). Thomson, J. J, [1919] (Sec. 89b) "Piezoelectricity and its applications," report of a lecture delivered on Apr. 11 at the Royal Institution, Engineering Ю7, 5113(1919). Tolch, N. A. [1935] (Sec. 10Ц, Ю7) "The rate of erosion of the Browning caliber .30 tank type machine gun M1919 as dependent on the temperature of the gun and other factors," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 13(July 8, 1935), 21 p. Tolch, N. A. [1936a] (Sec. 67, ЮЗ) "On heating effects and endurance properties of the 105-mm antiaircraft gun M1 as determined by rapid fire tests," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 35(Feb. 12, 1936), 16 p.
- 237 - Tschernoff, D. К. [1913] (Sec. 70, 13h) "Erosion of gun tubes," Proc. Int. Assoc. Testing Materials IT, 121-2(1913)- Written discussion read to sixth Congress of the Association in New York on Sept, 3, 1912 by Capt. Beliew, after paper on "The effect of high temperatures on, the physical properties of some alloys" by I, H. Bregowsky and L. W. Spring., Vieille, P.. [1901] (Sec. 76, 8Ц, 83, 9ha, 110, 11Ц, 132, 13h, 133) "Etude sur les phenomenes d'erosion produits par les explosifs," Mem. poudres salpetres 1J_, 137~21 9(1 901). Vieille, P. [1906] (Sec. 89a) ["Arrangement of crusher gages in measurement of explosion pres- sure,"] Comptes Rendus 1ЦЗ, 1218(1906). Vincent, T. K. [1937] (Sec..99) "Development of chrome-plating of guns," Aberdeen P. G., Ballistic Res. Lab. Rept. No. 87(Nov. 1937), 38 p. Walther, H. H. [1938] (Sec. 73) "Heat checks in brake drums," Brake Service (Aug. 1938). Webster, A. G., and L. T. E. Thompson [1919] (Sec. 89c) "A new instrument for measuring pressures in a gun," Proc. Natl. Acad. Sci. 3, 239(1919). Weyher, T. A. [1939] (Sec. 60, 61, 63, 123d, 130) "Tests- to determine limits of accuracy life of 133-mm guns, G,P.F.," report on Ordnance Program 3h!2; Aberdeen P. G. (July 10, 1939). Wheeler, H. E., [1922] (Sec. 73, 77, 133) "Nitrogen in steel and the erosion of guns," Trans. Am. Inst. Min. Het. Engrs. 67, 237“3O6, discussion 306-16(1922). Wheeler, H. E. [1923] (Sec. 77) "The nitrogen theory of erosion," Army Ordnance 3, 800(1923). Whelen, Townseiji [-1930] (Sec. 99) "Erosion test of caliber .30 chromiurf-plated machine-gun barrels," Springfield Armory (Jan. 29, 1930), Ц p.; Ordnance Dept. Library. Wulff, John [19M ] (Sec. 97) "Surface finish and structure," Metals Technology(l 9Ц1)'. Yarnell, H. E. [1910] (Sec. 11Ц, 113, 132) ' "Gun erosion," J. Am. Soc. Naval Engrs. 22, 33U(191O); Journal U.S. Artillery ЗЦ, 181(1910). '
- 259 - 112/ SUBJECT INDEX for PARTS ONE AND TWO Accuracy life, 10h. See also Life of guns. Adiabatic flame temperature, 36, 38, ЦО, Ц1, 109, 110(c), VII, III, Advance of forcing cone, 63, 125(d)., Alloy steels, tests of, 92, XXII, XXIII. Altered layer on bore surface, 71—79, 107, 13^ on erosion vent plugs, 73, 9h(b); hardness of, 7U, XIX; theories concerning, 75-77, 79; thickness of, 72, 9U(b), XVIII; on various steel surfaces, 73, 117, 118. •• Ammonia, action on steel of, 73. Ammonia fumes, 25. Ammunition, types of, 20, 21. Artillery, 5.• Austempering, XX. Austenite, 75, 77, XX. Autofrettage, 6. Bag gun, 18, 21, 60, 118, 125(c). Bainite, XX. Ballistic force, Ц0. Ballistics, interior, 29-58, Ballistite, 26, 9h, XII, Band, see Rotating band. Band slope, 10. Barium nitrate, 25, VI(A), Barrels, machine-gun of alloy steel, 92; of nonferrous metal, 93. Battelle Memorial Institute erosion tests, 88. Bell muzzle, 6. Beryllium-copper, gun barrels of, 93. Black powder, 2Ц, 28, 33, 106, VIII. Boat-tail, 17. Bore contraction of, 56; diameter of, 2, 55, 60, 125(c), I; dilation of, 55. Bore surface chemical reaction of, with powder gases, 135; cracking of, 70, 78, 79, 107, 113, Uh, 13Ц, 135; effect of, on erosion, 97; heating of, Ц7, 51, 53, 5U, 56, 132, XIII; lubrication of, 111-113- Bore-sear'cher, 1I4. 112/ Arabic numerals refer to sections; Roman numerals, to tables.
- 2.60 - Boroscope, 1Ц, Bourrelet definition of, 17; dimensions of, I, Built-up guns, 6, 96, I. Bullets, 17, 111, 112, 120. Burning constant, 30. Burning, rate of, 22, 23, 29-31, UU. Burst, length of, effect on erosion of, 103. Caliber definition of, 5; variation of erosion with, 62, 6Ц, XVII. Cannelure, 17, 18, 111, 122, Cannon, 3. Carbamite, 27, VI(B). Carburization, 73-77, 133. Cartridge case, 20. Case gun, 18, 21, 60, 118, 123(d), 123(e). \ Cementation, as ‘cause of altered layer, 73-77, 133. Cementite, 73, XX, Centering cylinder, 10, 60. Centralite, 27. Chalk, VI(B). Chamber, 2, 12, 13, 60, 73, I. Chamber rear slope, 12. Chamber slope, 10. Chambrage, 12, 13. ' Charge, in various guns, I, Chrome plating of guns, 98-100. Chrome-plated erosion vent plug, XXV(B). Chronograph, Ц9. Chronoscope, Ц8, Closed-chamber experiments, 36-39, 30, 80, Cold-worked guns, 6, 96, I, Cold-working as cause of altered layer, 73. Combustion heat of, 32, VII; products of, 33-33, 39, IX, X, XII. Compression slope, 10. Contraction of bore, 36, Cooling of machine-gun barrel, 31; of closed chamber, 37, 38. Copper gun barrel, 93. Copper pressure-gage cylinder 89(a). Copper rotating band, 18, Copper vent plug, XXIV, Coppering, 18, 60, 67-70, 113 "Coppers," 89(a). Cordites, various, 26, 106-108, VI(B), VII, X, XI. Cost of firing tests, XXI.
- 261 Govoluma, 37» 38. Cracking of bore surface, 70, 78, '79, 107, 113, 13U, 135; of chamber surface, \73« Cryolite, VI(B). Decoppering, 25, 69. Density of loading, 32, 37, I. Dibutylphthalate, 25, Vl(A), VII, Dilation of tube, 55. Dinitrotoluene, 2.5, VI(A), VI’I. Diphenylamine, 27, VI(A), VII, Dispersion, 18, 82, 125(b). Driving band, see Rotating , . band. Ductility of gun steel, 15. Economic life of a gun,. 105(b), 125(f); Seo also Life of guns. Efficiency of a gun, Ц9. Elastic limit of gun steel, 16. Electric arc, altered layer produced by, 73. Elongation of gun steel, 15, . XXII(C). Energy, muzzle, Ц9, 52, XV; distribution of, 52, XV; loss of, 125(a)» Enlargement' of bore, 60-62, 125(b), 12.7, XVI; of cracks by copper, 68, 70. Equilibrium, during burping of powder, Ц2, Equivalent service round, 12Ц, Erosion curves, 61, Erosion definition of, 2., 137; dissymmetry of, 60; of drilled shot, 117; effects of, see Forcing cone, advance of; Life of guns; Muzzle velocity, loss of; Rotating band, stripping of. factors influencing, 81, 102, 130; chamber design, 13; chemical composition of gun tube, 90-9U; chemical reaction, 73, 75-77, 7.9, 135; chrome plating, 98-100; coppering, 68; cold-working of tube, 96; ’ fire, continuous, 103; gas leakage, 18, 11Ц; heat, 132; lubrication of bore, 111-113; melting point of metal, 93, 9h(a), XXIV; muzzle energy, 130; muzzle velocity, 105; physical properties of gun tube, 90, 92; pressure, 110; propellant, 2Ц, 106-109; rate of fire,103; rifling design, 9, 119, 126, 136; rotating band, 19, 131, 132; stress concentrations, 9; surface of bore, 97; temperature, 10Ц, 109, 110(c); turbulence, 13, 86; velocity, 105; weight of projectile, 105. muzzle, 2, U7, 60, 72, 98, 136;
- 262 - Er о s i on<—continued tests of, see Erosion, factors influencing; Firing tests; Laboratory- erosion gun; Vent plug tests; theories of, 75-79, 1Ш, 131-135. Expans ion of bore, 70, 13h; of powder gases, 50; radial, 96; thermal, of gun steel, XXII(B). Explosives, high and low, 22» Explos ion heat of, 32, hl, VII; products of, see Combustion, products of. Explosion pressure, 37» Explosion, temperature, 36, 38, hO, Ц1, Ю9, 110(c), VII, XII. Fterrite, 75, 77, XX. Fire, continuous, effect of, on rate of erosion, 103. Firing tests, 81-83, 92, 99, 105, XXIII. Flashless propellants, 25, 108, VI(A), VI(B), VII, IX. Forcing cone, 10, 126; advance of, 63, 119, 125(d), XVI. Forcing resistance, Ц6. Fouling, see Coppering. Free run-up, 125(d). Friction, bore, 19, h6, h7, 53, 136. Fringing, 18. Gages, pressure, 89. Gas checks, 20, 121, 126, Gas leakage experiments on, 116-119; theory of erosion by, 18, 110, 11U-122, 126, 132. Gas reactions at high tempera- tures, 38-Ц2, Gas-washing theory of erosion, 93, 132. Gases, powder, composition and temperature of, 33-36, 38-Ц1, 5o, 135, VITI-XII. Gilding metal rotating bands, 18, 68. Granulation, 23, 31, 110(a). Graphite, as bore lubricant, 111-113. Graphite glazing on powder, 25, VI(A)„ Grooves, 8, 70, I; erosion of, 60, 93, 105(a), 105(b), 107, 11Ц. Gun built-up, 6, 9б>, I; case, 21, 60; chrome-plated, 98, 99; cold-worked, 6, 96, I; definition of, 5; laboratory erosion, 88; nickel-plated, 101; radially expanded, 6, 96, I; tapered bore, 7; wire-wound, 6, I. Guns,- see also Supplementary index of guns by calibers; classification of, 5, I; construction of, 6; dimensions of, I,
- 263 Gunpowder, see Black powder. Hard layer, hardened layer, hardened skin, 71J See also Altered layer. Hardness, of altered layer, 7h, XIX; of gun barrels, XXII(C). Heat absorbed by bore surface, 51, 75, 132, XIII; of combustion, 32, VII; of explosion, 32, VII; of friction, 53. Heat capacity of powder gases, . u1, U2. • . • . Heat checking, 70, 93» Heat flow in gun, 5h» Howitzer, definition of, 5. Hypervelocity guns, 3, 105(c), 112. Hydrogen, effect on steel of, 73. Igniter charge, 28, 31. Inconel vent plug test, XXV(B), Inclusions in steel, 78'. Interior ballistics, 29-58. Invar gun barrels, 92. Ionization of powder gases, 135. Iron, alpha and gamma, ?5, 77, 79. "J" band, 122. Jacket of bullet, 17'. Keyhole, 82, 92, 125(e)* Lands, 8, 9, 70, I; erosion of, 60, 93, 105(a), 105(b), 107, 11Ц. Length of gun, 105, I. Life of guns, 123-130; criteria for, 125; effect of muzzle velocity on, 3, 105, XXVII; formulas for, 127-130; of special materials, 92., 93, 98-101; of various calibers, XXVII, XXVIII. Life of machine-gun barrels, 92, 10I1, 107, XXIII, XXVI. Liner, loose, 6. Lubrication of bore, 111-113» Martensite, 75-77, 79, XX. Materials for gun tubes, tests of by erosion vent plugs, 9I1; by firing, 92, 93, 99; by laboratory erosion gun, 88,. Melting of bore surface, 132. Melting point, relation of, to erosion, 93, 911(a), XXIV. Metal, weight of lost during erosion, 6I1, XXIV. Metals, pure, erosion resist- ance of, 9h(a). Methane, formation of, in powder gases, 39, VIII, X, XI.
- 26Ц - Mineral jelly- in cordite, 26, 27, VI(B). Molybdenum liner, 93. Monel metal gun barrel of, 93; vent plug test of, XXV (B). Mortar, definition of, Munroe effect, compared with erosion, 73. "Mushroom pad" obturator, 20. Muzzle energy, Ц9, 128-130, XV, XXVIII. Muzzle erosion, 2, Ц7, 60, 72, 98, 136. Muzzle velocity, h-9, 99, 100, 105, 127-129, I, XXVII, XXVIII; see also Velocity, muzzle» Nickel-plated guns, 101. Nickel steel for guns, 13, 91, II, XXII, XXIII. Nitralloy, vent plug test of, XXV (B). Nitride theory of erosion, 79. Nitrocellulose, see Powder, nitrocellulose. Nitroglycerin, 22, 2Ц, 26, 106, 107, Vl(A), VI(B), VII, X. Nitrogen in altered layer, 77. Nitroguanidine, 23, 108, VI(B). Obturation of breech, 20. Obturation of .bore, see Gas leakage. Ogive, definition of, 17. Origin of rifling, definition of, 11. Paris gun, 103(d). Pastilles, 66. Pearlite, 73, XX, Penetration of armor plate, 3. Perforation of powder grains, 23. Picrite, 2.3, 108, VI(B). Plug gage, 63. Potassium nitrate, 23, VI(A). Potassium sulfate, 23, VI(A), VI(B). Powder American, 23, 107, 108, VI(A); black, see Black powder; British, 26, 108, VI(B); burning of, 29-Ц2; chemical composition of, 2Ц, 23, 26, VI(A), VI(B), VII, X; double-base, 2Ц, 107, 119; aee also Cordites, various; effect of type of, on rate of erosion, 8Ц, 106-108, XXVI; energy distribution of, 32, XV; flashless, 23, 108, VI(A); VI(В), IX; FNH, 23, 27, VI(A), VI(B), VII, IX; forms of, 23; ignition of, 30;
- 265 - Powder—continued NH, 25, 2?, VI(A), VI(B); nitrocellulose, 22, 2Ц, 25, 26, 106, 107-109, VI(A), VI(В), VII, X; products of combustion of, 3U, IX, X, XII; pressure of, 29, I; propellent, 2, 22-26, VI(A), VI(B); pyro, 25, 27, VI(A); quickness of, 31; single-base, 21;, 1Q7, 119; smokeless, 2Ц; stability of, 27; unburned, 66, 133. Powder chamber, see. Chamber, Powder gases, see Gases, powder, Powder grain, dimensions of, 23 . Powder potential, 32. Pressure decrease of, with erosion, 105(a), 105(b)•> effect of, on erosion, 105, 110; explosion, 37; maximum, position of pro- jectile at, Ц5; muszle, 15; starting, 18; in various guns^ I, Pressure gages, 89, Pressure-time curves, ЦЦ. Pressure waves, 31. Primers, 28, Probability of hitting an air- plane, 3, 105(b). "Probert" rifling, 122, 126, 136. Projectiles, 2, 17-21, Ц3, 105, 113, I, V; See also Bullets, Proof stress, 16, Propellant, see Powder, Proportional limit, 16. Proportional method, 16. R, D, system of rifling, 122, 126, 136. Radial expans ion, 6. Range of-gun, decreased by erosion, 2, 125(b). Rapid fire, effect of, on । erosion, 103, Ю1;, Resistance, forcing, 18, Ц6, 126, 132. Rifling depth of, I; design of, 8, 9, 122, 126, 136; length of, 10, I; twist of, 8, 9, I. Rotating band composition of, 18j design of, 10, 17, 18, 122, 126; dimensions of, 18, I; effect of, on erosion, 19> 119, 131, 132; on gas leakage, 1111. engraving of, 18, Ц6, 126, 132; erosion of, Till, 117, 118; stripping of, 105(b), 125(e); See also Friction, bore. Rounds, variation of erosion with number of, 61,
- 266 - Scoring, 6$, 113, 117* Service round, 12Ц. Shot start pressure, 18, Ц6, 126, 132. Small arras, 3, 99, 100. Solder as decoppering agent, 69. Solvents, powder, 2Ц. Sorbite, 73, 93, XX. Stabilizers, 23-27. Star gage, Th, 6O-61, 6Ц, 99, XXII, XXIII. Steel alloy, 92, 9h(b), 9h(c), XXV; gun, composition of, 13, 92, II, III, IV, XXII(A); heat treat- ment of, 1.3, 92, XXII(C); specifications for, 13; thermal trans- formation of, 92, XXII(B); tensile properties of, 13, 16, 92, 9h(b), XXII(C); tests of, 92, 9h(b), 9h(c), XXIII, XXV?A), XXVfB); microstructural con- stituents of, 73, 77, 93, XX. Strength of gun tube 6, 70. Stress on gun tube, 6, 33, 37, £8. Stress concentrations, effect of, on erosion, 9. Surface of eroded bore, 70-79. Temperature of barrel, 31, 3h, ЮЗ, 10Ц, XIII, XIV, XXVI; of bore surface, 3h, 132; of explosion, 36, 38, Ц0, Ы, 109, 110(c), VII, XII; of powder gases, 38, Ц0, Ы, vii, хи. Tensile properties of gun steels, see Steel, tensile properties. Tensile strength, 16, 103- Time of flight, 3. Tin, 23, VI(A). Tinfoil, use of, in de- . coppering, 69. Tin oxide, 23, VI(A). Tip, definition of, 82. Transformation points of steel, 92, XX, XXII(B). Travel of projectile, h3, I. Trinitrotoluene, 22. Troostite, 71, 73, 76, XVIII, XX. Tube, 2, 6, I; heating of, during firing, h7, £l-5h, 36, 132, XIII, XV; reaction of, to firing, 31, 33-37. Tumbling, 82, 123(e).
- 267 - Tungsten steels, XXII, XXIII, XXV(B). Turbulence, '-13, 86». Twist of-rifling-, 8, 9, I. . ' г . . Vaseline., used in cordite, 26, 27, VI(B). Velocity, muzzle, Ц9, 99, - 100, 103, 127-129, I, XXVII, XXVIII; effect'of coppering on, 67; effect of, on ero- . sion, 3, 103; increase of, in chrome-plated gun, 100; loss of, 99, 100, 101 , 103(a), 103(b), 123(a), 123(c), 123(d), 126; loss of, by erosion, 99, 103(a), 103(b), 123(a), 123(c), 123(d), 126. "Vena contracta,” theory of erosion by, 132. Vent, effect of sifce of, on erosion, 83(a), 83(b). Vent plug altered layer of', 73, 9h(b); effect of turbulence on, 13, 86. Vent plug tests, 80, 8^-87, 9Ц, Ю9, 110(c), 113, XXIV, XXV(A), XXV(Bj. Wad, wax,-112. Warming round, 12Ц. Water-gas reaction, 39rh1, 30. Wear, frictional, 2, h7, 136. Web thickness, 23, 31» Weight of metal lost by ero- sion, 6Ц, 9h(b), XXIV, XXV. White layer, see Altered layer. Yaw, 82; See also- Keyhole. Yield-point, 16. Yield strength, 13, 16. SUPPLEMENTARY INDEX OF GUNS BY CALIBER .30-caliber machine-gun altered layer of, 72; area under erosion curve of, 6Ц; chrome-plated, 99; cost of firing to de- struction, 83, XXI; dimensions, I; heating of barrel, 31, l0i|, XIII, XIV, XXVI; life of, 103, ЮЦ, 127, ..XXVI;' partially chrome-plated, 100; purpose of, I(notes). ,.30-caliber machine-gun dimensions, I; distribution of energy in, XV; heating of barrel, 31, 103,. XIII, XIV; life, 123(a), XXVIII; purpose, I(notes). 1.1-inch Navy gun cracking of surface of 70.
IF1ED - 268 - 37 gun altered layer in, 72, 7h; bore of, photograph of, 70; chrome-plated barb of, 99; cost of firing to de- struction, 83? XXI; cracking of surface of, 70; dimensions, J; erosion of, 6l, 62, 107, 119, XVII; forcing cone in, advance of, 119; heat of friction in, £3; life, XXVIII; purpose, I(notes); scoring of, 65. 90-mm gun dimensions, I; purpose, I(notes). 3.7-in. gun loose liners for, 6; nickel-plated, 101; R. D. system of rifling in, 122, 126, 136. 105-mm gun altered layer of, 7h; ammunition, semifixed, 21; dimensions, I; erosion of, 103; heating of tube, 51, 5U, 55, XIII, XIV; purpose, I(notes). 75-mm gun altered layer in, 7h, 79; dimensions, I; heating of tube of, 5h, XIII, XIV; life, XXVII, XXVIII; pastilles in, 66; pressure-time curve, Uh; purpose, I(notes). 3-in. gun bore surface of, photo- graphs of, 71; chrome-plated liners of, 99; dimensions, I; distribution of energy in, XV; erosion of, 60, 61, 105(a), 105(c), XVII; heating of tube of, 51, XIII, XIV; life, XXVII; loose liner for, 6; pressure-time curve, ЦЦ; purpose, I(notes). U. 5-in. gun eroded surface of, 107; loose liner for, 6; nickel-plated, 101. U.7-in. gun ammunition, semifixed, for, 21; chambrage of, 12; dimensions, I; erosion of, 105(b); life, 125(f), 130, XXVIII; purpose, I~(notes). 5-in. gun ammunition, semifixed, for, 21; life, XXVIII; 6-in. gun ammunition, semifixed, for, 21. 6-in, howitzer eroded surface of, 107.
UNCLASS 155-пип gun cost of firing to destruction, 83, XXI; dimensions, I; erosion of, per round, 60, 61, 6Ц, XVII; unsymmetrical, 60; forcing cone of, advance of, 63, 125(d), XVI; life, 125(f), 130, XXVIII; pressure-time curve, hU; purpose, I(notes); weight of metal lost in firing, 6Ц. l55~ram howitzer pastilles in, 66. 8-in. gun dimensions, I; erpsion of, XVII; life, 125(b), XXVIII; purpose, ifnotes). 12-in. gun altered layer in, hardness of, 73, XIX; cracking of surface of, 70; effect of design of rifling on erosion, 102; weight of metal lost per round, 6Ц. 1h-in. gun chrome-plated, 98; cracking of surface of, 70; dimensions, I; erosion of, 61, XVII; life, 125(c), 130, XXVIII; purpose, l(notes). 16-in. gun altered layer in, 7Ц; cost of firing to destruction, 83, XXI; dimensions, I; erosion of, 62, XVII; . life, 130, XXVIII; ' ' purpose, I(notes).