INTERIOR BALLISTICS OF GUNS Edited by Herman Krier University of Illinois at Urbana-Champaign Urbana, Illinois Martin Summerfield New York University New York, New York Volume 66 PROGRESS IN ASTRONAUTICS AND AERONAUTICS. Martin Summerfield, Series Editor-in-Chief New York University, New York, New York Published by the American Institute of Aeronautics and AstronauticsTable of Contents AUthOrs... cee eee cece renee ence neeee ee, Preface List of Series Volumes 1-66 .....---++eeeeeeeees oe XViii Part I.... An Introduction to Gun Interior Ballistics and a Simplified Ballistic Code.........0eeeeeeereeeeeer ered Including: Review of the Ballistics Cycle..... 0.00.0 eee eee ener e eect e eee eee Pressure-Travel Curve Efficiencies of a Gun System and Charge . Review of Gun Propellants. Burning of Propellant Grains . . Effects of Grain Characteristics on Gun Performance Energy of Gun Propellants: General Comments . . . A Ballistic Model Equation of State . Form Function Analysis. Burning Rate Equation Variable Volume Considerations/Projectile Motion . Heat-Loss Effect . Summary of Assumptions ......... Encapsulation of Governing Equations . . Example Calculation: Standard Gun-Propellant System . Practical Interior Ballistic Analysis of Guns...... Including: Types of Gun Models Available.............. Classification of Available Models . . Optimum Gun Models . Modeling of Propellant Acceptance Tests . . Design of New Guns ..., Improvement of Existing Gun Performance . Matching, Interior Ballistic Code Predictions to Experimental Data ....55 Input Data for Interior Ballistic Models eet eee ne taInterior Ballistic Modeling Applied to Small-Arms Systems........67 Including: Review of Empirical Models. ... . Analytical Interior Ballistic Models . Propellant Deterrent Technology. . Analysis .... Specialized Interior Ballistic Models. Special Topics in Small-Arms Interior Ballistics ....... Small-Arms Gas Transmission Systems . Pressure Waves in Small-Arms Systems. . Ballistics Applied to Rapid-Fire Guns. Including: Interior Ballistic Model [Pressure, Density, and Velocity Gradients Muzzle Flash. Heat Conduction to the Gun Barrel . Heat Transfer through the Gun Barrel -i3 ics. Mathematical Modeling of Recoilless Rifle Interior Bal Includiny Physical Basis of a Mathematical Model ........-0.+00e0eee0ee00 115 Purpose and Limitations of a Model . WS Basic Assumptions . . . 16 Governing Equations . 118 Barrel Flow ..... . 118 Combustion Chamber .. . no Nozzle Flow . Boundary Conditions . Open Muzzle Boundary Conditions Boundary between Chamber and Barrel . Thrust Computation .........+.+20005 Noise Estimation. Numerical Models. Example: Typical Results . . . Outlook for Improvements in the Model . Theory of High-Muzzle-Velocity Guns ........2++0e000+ Including: Basic Requirements for a High-Speed Gun ... Preburned Propellant (PP) Gun: General. . . Description of the Preburned Propellant Gun. Equations for Disturbances Traveling in the Gas Constant-Diameter Preburned Propellant Gun. Restatement of the Equations Applicable to an Isentropic Gas Expansion ‘in a Constant-Cross-Sectional-Area Tube. wees eee 1DCharacteristic Equations for the Effectively Infinite-Chamber-Length Gun . . 140 Role of Acoustic Inertia. Equations for the Motion of the Projectile Propelled in a PPIG Gun « Finite-Chamber-Length Gun . . Numerical Results for the Projectile Behavior in a PPIG Chambered Gun. Influence of Gas in the Barrel in Front of the Projectile Applicability of the Isentropic Theory to Guns ..... Experimental Results for Guns with Heated Propellants. Analytical Considerations of the Effects of Nonisentropicity . Methods of Heating the Propellant Two-Stage Gun... ..-0-ee eee Constant-Base-Pressure Gun. . . 5 Effects of Propellant Gas Nonideality on the Performance of Preburned Propellant Guns Proposed Schemes to Increase Projectile Velocities . oT | nS OL) Modeling of Two-Phase Flow in Guns ......+-+++eeeeeeeeee+ +176 Including: Nature of the Physical Problem and the Model . . Correspondence Between Theory and Experiment Balance Equations for Heterogeneous Two-Phase Flow . Closure and Constitutive Laws Structure of the Balance Equations Methods of Solution Charge Design Considerations and Their Effect on Pressure Waves in Guns.......seseeeeeeeeeerereeeee ee 197 Including: Pressure Wave Phenomenology Experimental Considerations Charge Design Factors Influencing Pressure Waves «198 «202 Ignition Stimulus . + 204 Mass Burning Rate = 206 Permeability... . 2211 Chamber and Charge Geometry. 214 Reliable Gun Ignition Ignition and Flamespreading Phenomena in Granular Propellant Gun Charges ...... Including: Igniter/Propellant Reaction . Small-Arms Systems , Large-Caliber Guns. ++ 228 «231 +233 +6235vi Part II Applicability of Relative Quickness as a Precision Measure of Muzzle Velocity. sence cece cena tenn 246 Including: Analysis.........e0e0ee ee Sensitivity Study ....... Datum Case Influence Coef! Muzzle Velocity Data A Suggested Procedure for Prediction of MV from Closed Chamber Firings .......6.0.00006 nts and Comparison with Sensitivity of Ballistic Performance to Propellant Combustion Properties ........0+++eeeeeeeeee eens Including: Experimental Program . . Closed-Bomb Analysis ... General Experimental Program for M6MP £/175-mm. ‘Two-by-Two Experimental Plans. Processing and Testing. «0.2.2... Experimental Program for MIMP £/155-mm Length Variation Study. Web Variation Study . Predictive Approaches . Closed-Bomb Studies... Mathematical Model Studies . Description of Models . Comparison of Predicted and Experimental Results ...---+-. +++» +278 Research Test Techniques Applied to Gun Interior Ballistics ...... 281 Including: Instrumentation. . .. Pressure Measurement . Heat Sensors . Optical Techniques. Projectile Velocity and Travel Engraving and Bore Friction Forces Research Facilities and Techniques. .. . Component Tests . Ignition Train Test Fixtures Vented Chamber .......2+. Instrumented Guns and Ballistic Simulators . posable Breech Facility... eee Diagnostic Liquid Propellant Gun (LPG) Test Setup ...... fete e eect e eee e eee 282 = 282 2284 +286 Beapooade!)vii eae ccccececn ene n ss OF Part IV... . seen ee eeeeeerees Gun Propellants ........00sceeeceeeeeee ree seee cece cess ee 307 Including: Background. . 307 -307 = 309 309 +309 ait 312 313 2314 us 31S Manufacturing Processes Common Terms Used in the Propellant Field Chemical Composition . Conventional Propellants Developmental Propellants. Thermochemistry Burning Rates . Measurement Effect of Conditioning Temperature Correlation with Flame Temperature. Burning Rates of Deterred Propellants 2315 Erosive and Dynamic Burning 316 Physical Properties ........ 316 317 318 Charge Design Considerations . Ignitability and Safety .... Equations of State and Thermodynamic Data for Interior Ballistics Calculations Including: Equations of State Used in Ballistics Calculations Theoretical Approach . . Derivation of the Physical Parameters . Pure HO and “Pseudowater”” . CO}; Ny; CO; Hy... H,0/CO, Interaction Second Comparison with the Truncated Virial Equation: Pure Water. Appendix A: Outline for the Calculation of Mixture Properties of H,0, CO,, N,, CO, Hy Appendix B: Outline for the Calculation of the Helmholtz Energy of the Haar-Shenker Equation of State +325 328 332 336 Part V wo. .ee eee ee ee An Historical Perspective on Gun Interior Ballistics at the U.S. Army Ballistic Research Laboratory ...........00.000..349 Including: Highlights of Research and Development Activities from 1919 to Present ... Summary of Experimental Techniques and Facil Gun Ballistics Perspective on Current Technology Gaps and Ballistic Problems Yet to Be Solved .Authors An Introduction to Gun Interior Ballistics and a Simplified Ballistic Code Herman Krier and Michael J. Adams University of Illinois at Urbana-Champaign, Urbana, Ill. Practical Interior Ballistic Analysis of Guns Paul G. Baer Ballistic Research Laboratory, Aberdeen Proving Ground, Md. Interior Ballistic Modeling Applied to Small-Arms Systems Sidney Goldstein U. S. Army Armament Research and Development Command, Dover, N.J. Ballistics Applied to Rapid-Fire Guns Otto K. Heiney Air Force Armament Laboratory, Eglin Air Force Base, Fla. Mathematical Modeling of Recoilless Rifle Interior Ballistics Aivars K. Celmin3 U.S. Army Ballistic Research Laboratory, ‘Aberdeen Proving Ground, Md. Theory of High-Muzzle-Velocity Guns Arnold E. Seigel University of Maryland, College Park, Md. Modeling of Two-Phase Flow in Guns P. S. Gough Paul Gough Associates, Inc., Portsmouth, N.H. Charge Design Considerations and Their Effect on Pressure Waves in Guns Ingo W. May and Albert W. Horst Ballistic Research Laboratory, Aberdeen Proving Ground, Md. Ignition and Flamespreading Phenomena in Granular Propellant Gun Charges J.L. East Jr. Naval Surface Weapons Center, Dahlgren, Va. viiiApplicability of Relative Quickness as a Precision Measure of Muzzle Velocity : Moshe BenReuven and Martin Summerfield Princeton University, Princeton, N. J. Sensitivity of Ballistic Performance to Propellant Combustion Properties P. Serao U, S. Army Armament Research and Development Command, Dover, N. J. J. Pierce Hercules, Inc., Radford, Va. Research Test Techniques Applied to Gun Interior Ballistics E. B, Fisher Calspan Corporation, Buffalo, N. Y. Gun Propellants Ludwig Stiefel Armament Research and Development Command, Dover, N. J. Equations of State and Thermodynamic Data for Interior Ballistics Calculations E, G. Powell and G. Wilmot U. S. Naval Surface Weapons Center, Dahlgren, Va. L. Haar and M. Klein National Bureau of Standards, Washington, D. C. An Historical Perspective on Gun Interior Ballistics at the U. S. Army Ballistic Research Laboratory Leland A. Watermeier and John M. Hurban Ballistic Research Laboratory, U. S. Army Armament Research and Development Command, Aberdeen Proving Ground, Md.Preface In planning this new volume in the Progress in Astronautics and Aeronautics series, the volume editors were motivated by the realization that, although the science of the interior ballistics of guns has advanced markedly in the past three decades, and especially in the decade since 1970, there exists no systematic textbook or monograph in which recent advances are summarized for the benefit of either the beginner who might want to enter the field or the specialist who would like to study some particular details of it. The most widely used books covering the state of theoretical knowledge are those by the physical chemist, Professor J. Corner, The Theory of the Interior Ballistics of Guns, published by John Wiley and Sons, New York, in 1950, and by F. R. W. Hunt, Internal Ballistics, published by His Majesty’s Printing Office, London, in 1951. Both books were outgrowths of the research performed by Hunt, Corner, and their colleagues during World War II. Later, in 1965, there appeared a limited-circulation monograph by A. E. Seigel, The Theory of High Speed Guns, published by the NATO Advisory Group for Aeronautical Research and Development (AGARD) in Paris, but it was confined mainly to the gasdynamics of interior ballistics, taking advantage of the progress of compressible flow theory in the previous two decades. In the Soviet Union, a book by M. E. Serebryakov, Interior Ballistics of Tube Weapons and Solid Propellant Rockets, published first in 1942, revised in 1949, and revised again in 1962 as the third edition, printed by the State Scientific and Technical Publishing House, Moscow (GOST), fulfills the role of a textbook, but its scientific depth is rather limited. (A translated version in English is available in the U.S. from the National Technical Information Service.) In the light of current developments, however, all four volumes fall far short of the state of knowledge today. The pace of scientific progress has been rapid. Indeed, many of the articles in the present volume deal with per- formance problems and pathologies of guns that could not have been discussed quantitatively at all before 1970. The Corner and Hunt books were written before the advent of the modern high-speed, high-capacity computer, and so the theoretical methods in those books necessarily involve severe approximations in the treatment of the fluid flow and combustion processes. The Serebryakov book covers essentially the same ground. The Seigel book was written with an eye mainly on the pure gasdynamic aspects xixii of interior ballistics, with no attention to the presence of the solid propellant. It will be seen in the present volume that three new factors have entered ballistic theory in the past decade, each having flowed, it so happened, from a stream of science unrelated to in- terior ballistics. First and foremost was the detailed treatment of combustion of the solid propellant, including particularly the ignition and flame-spreading phases at the start of the combustion cycle and the associated nonsteady gas flow, all of which had reached an advanced state of development in the field of rocketry. The second was the formulation of the dynamical fluid flow equations in two-phase flow form with appropriate relations for the interactions of the two phases. The third is what made it possible to incorporate the first two factors, namely, the use of advanced computers to solve the partial differential equations describing the nonsteady two-phase flow system with simultaneous burning of the solid propellant granules. Ballistic theory augmented in this manner with two-phase flow theory and combustion theory has brought new insight and new powers of prediction and diagnosis to the ballistician. It might seem appropriate, perhaps, to revise the name of the new science of interior ballistics to reflect this augmentation, to call it perhaps interior combustion ballistics. Historically, the pioneering work of this type was the paper of Kuo, Vichnevetsky, and Sum- merfield of Princeton University entitled ‘Theory of Flame Front Propagation in Porous Propellant Charges under Confinement,” published first as the Ph.D. thesis of K. K. Kuo and then in the April 1973 issue of AIAA Journal, which showed it was possible to deduce accurately the profiles of the pressure waves in a packed propellant bed as measured by W. Squire and M. P. Devine at Frankford Arsenal. The striking success of this theory, including the prediction of a ‘‘continental divide’ in the combustion wave, launched the new science, interior combustion ballistics. It seems to us that the greatest change that may yet occur in the future in interior ballistics theory may be in the improvement of the formulation of the combustion process. Detailed as it now is, the combustion description presently employed in interior ballistics is limited by the lack of knowledge of erosive burning of propellant at gun flow conditions and by the lack of knowledge of dynamic burning rates. We do not even know whether the product gases are indeed all burned and hence substantially inert in the fluid flow field, as assumed in all the current theories, or still reacting andxiii therefore not inert. There is some evidence that the gaseous flame of a nitrocellulose-base solid propellant represents a relatively slow chemical reaction (whose characteristics have not yet been defined) and that ‘‘after-burning’’ may be significant in the combustion chamber and down the barrel. Therefore, the science of interior combustion ballistics is far from finished. To this, we may add, there are ahead of us some very great challenges involving the complex fluid flow process; for example, the multidimensionality of the flame-spreading and flow processes, the incorporation of tur- bulence, and the reduction of all of this to workable, efficient computer codes are goals still to be accomplished. Nevertheless, this volume brings together so much of the new field that the editors felt justified in bringing it out now, even while active research on interior ballistics is going on. In laying out the volume, we decided, after reflection, that the new approaches would not be comprehended adequately unless they were accompanied by articles describing in up-to-date terms the best of the analytical methods previously available and also by auxiliary articles on propellant characteristics, thermochemistry of combustion, fluid dynamics of gun flows, experimental methods in interior ballistic research, and the application of interior ballistic methods to typical modern guns. In this way, the book became more comprehensive and has acquired the character of a textbook on modern interior ballistics suitable for advanced educational purposes. With literally over a thousand references cited, this volume makes a unique contribution also to the interior ballistics field for the benefit of practicing development engineers and research scientists. With the decision to organize the volume in the manner of a textbook, it became logical to place first in the volume those papers that develop the subject from the elements of physics and fluid mechanics and then apply the resulting simplified analyses to practical gun engineering. Thus, the first paper in this category, by Krier and Adams, serves as an introduction to gun interior ballistics and also presents the rudiments of how a simple ballistic code can be developed. The predicted pressure and projectile velocity history for a large-caliber gun are compared with other models and with the experimental observations. The paper by Baer presents the state of the art of interior ballistics models with attention to how well these models do in the accurate prediction of ballistic performance. Baer includes references to seventy separate studies in ballistics research. Goldstein also reviews the application of ballistic codes, but withxiv attention to small-caliber systems. In the fourth paper, Heiney develops in detail an engineering interior ballistic code, including factors for high-rate-of-fire guns. Novel methods to account for gas velocity gradients behind the projectile and the heat-transfer losses are presented. The last paper in this category, by Ce/mins, develops an interior ballistic code for a recoilless gun. It provides a detailed analysis of the gasdynamics of the unsteady flow through the barrel and the supersonic nozzle at the rear end of the rifle. In the next group of four papers, the reader is taken into the more advanced field of detailed interior ballistics. The first of these, by Seigel, reviews the theory of one-dimensional compressible flow in a gun and derives unique limits of projectile muzzle velocity. It offers novel methods to obtain ultrahigh muzzle velocities and suggests areas of research to obtain such high gun performance. Over one hundred references are cited in the paper. The next paper, by Gough, is one that represents the new field mentioned above, which we have described with the name, interior combustion ballistics. The equations of two-phase nonsteady compressible flow with solid propellant ignition and combustion are applied to the problem of modeling the start-up phase of a gun—sometimes called the ignition phase by gun engineers, in spite of the fact that ignition in the context of combustion science refers to a particular event, the initiation of flame, and not more. The consequences of Gough’s model include the influence of the primer-propellant interface, as it is sometimes called, and the effect of charge configuration. The structure of the fluid mechanics conservation equations is em- phasized, together with required constitutive relations, and numerical methods of solution are discussed. References are given to other advanced two-phase flow formulations for the treatment of this problem. May and Horst, in the next paper of this group, review the ex- perimental evidence of pressure waves in gun charges during the ignition and flame-spreading phase, in the light of theoretical models of interior ballistics. Their work has shown that such pressure waves not only modify the final projectile muzzle velocity and thus throw off the theoretical performance predictions, but may even lead to breech blows, warhead prematures, or other destruction of the gun chamber. The final paper in this group, by East, reviews the experimental research efforts for obtaining information on ignition of gun charges, with special attention to the flame-spreading and initial flow process. He concludes that an igniter system thatxv functions satisfactorily in one gun propulsion system may fail in an apparently similar system as a result of upsetting the critical balance of parameters that governs the relationship between charge reaction and projectile motion. Two papers are offered next that deal with the problem of quality assurance and quality control of the propellant manufactured for use in guns. Although it may seem straightforward to control the propellant process by means of rigid physical and chemical analyses of the product, it turns out to be better to test the product by a functional combustion test that measures more nearly what a propellant is supposed to do. Here, interior ballistics theory is a guide. In the first paper, BenReuven and Summerfield show that the muzzle velocity produced by a particular propellant depends theoretically on a set of factors that can be subjected to measurement without test-firing the propellant in a gun. The methodology is described in terms of a group of sensitivity coef- ficients that can be evaluated, the most important test factors being the relative force and the relative quickness, conveniently measured in a closed combustor. Serao and Pierce summarize in their paper a systematic experimental program, employing closed combustor firing tests, with specially prepared deviant propellant samples, to assess the sensitivity of gun muzzle velocity to variations in propellant properties. Their work was motivated in an attempt to find reliable alternative techniques for ballistic performance prediction in lieu of actual proving ground firings for newly manufactured propellant. The two papers combine to offer a quantitative, theoretically justified, framework for quality assurance. Theoretical work in interior ballistics has to be supported by diagnostic interior ballistic measurements. The determination of muzzle velocity and pressures in the chamber and the barrel is relatively straightforward today, but more detailed information is required in order to evaluate the various interior ballistic models. The task is made exceptionally difficult by the severe environment of high pressures, high temperatures, rapid rates of change, and intense shock and vibration. The paper by Fisher describes not only some standard techniques but also some important research techniques employed in modern interior ballistics studies. A significant part of the paper deals with instrumentation for experimental study of Propellant ignition, combustion, and projectile travel ina gun.xvi The topic of propellant characteristics is an essential element of any treatment of interior ballistics. Utilizing more than one hundred references, Stiefel reviews gun propellants, including their history, terminology, manufacturing processes, thermochemistry, burning rates, and physical properties of concern to the interior ballistician. The reader is to be cautioned, however, that the field of propellant characterization is not a static one. New propellants are frequently invented for special purposes, but, more than that, the perception as to what kinds of properties are significant for modern interior ballistics is being sharpened in the light of continuing theoretical development. The review by Stiefel is presented as both a source of directly useful information and a starting point for consideration of properties to be evaluated. Along with the propellant, it is necessary for exact interior ballistic theoretical work to characterize the combustion gases. The paper by Powell, Wilmot, Haar, and Klein treats nonideal equations of state for high-pressure, high-temperature gases, relating them to their molecular properties. This is an enormous subject, one that has importance not only for the gun ballistician but more widely for the pure physicist and the physical chemist and for applied scientists in many fields. Powell et al. bring the subject to the level of application to interior ballistics, with all its complications. At this point, a reader might feel satisfied that the set of articles described above constitutes an adequate unified picture. Of course, there are many other problem areas of concern to the interior ballistician—barrel heating, barrel erosion, muzzle flash sup- pression, and many more—but the main questions might seem to have been covered. Nevertheless, it was decided by the editors that an important aspect of the subject was still missing, namely, its historical development. It is really not possible to appreciate the sweep of interior ballistics as an applied science without examining in some detail its evolution, not only in its theoretical aspects but in the practical and empirical aspects, in which many of the most significant advances may be found. For this purpose, the editors invited Watermeier and Hurban of the U.S. Army Ballistic Research Laboratory, Aberdeen Proving Ground, Md., to write this kind of review. Their paper, the final one in the book, constitutes an historical commentary on the highlights of interior ballistics research and development at BRL, selected with the benefit of their long professional experience in the field. The article reaches back to the time of World War I and takes the surveyxvii all the way up to the present. Although one might have wished for a review that also covered all the significant works at other U.S. in- terior ballistics laboratories, those in the Navy, the Air Force, and private industry, and in foreign laboratories as well, it was soon recognized that this would not be feasible. It would take a whole separate book to do that job. In point of fact, however, the scientific history of BRL parallels closely the general scientific history of interior ballistics in the twentieth century. All through its existence, BRL has addressed the most important problems of interior ballistics and continues to do so. Thus, a chapter devoted to that narrowed scope was deemed to fulfill our purpose. However, in order to provide the reader with broader access to the field, Watermeier and Hurban have appended to the article a list of im- portant readings representing the work of other research and development groups. Taken together with the more than 150 cited references in the article, the scope of the field and its historical development are portrayed in exceptional detail. The editors offer this book on progress in interior ballistics in the belief that it will serve as a useful text for introducing students and others to the modern aspects of an interesting science that has an- cient roots. We believe also that it can serve as a useful reference book for those working in the field who may wish to bring them- selves up to date in some particular topic. The editors gratefully acknowledge the skillful professional work of Miss Ruth F. Bryans, Adminstrator, Scientific Publications, of the AIAA—and her patience with us as well—and the work of her associate, Mrs. Norma J. Brennan, Managing Editor of the Series. Finally, we owe a large debt of gratitude to the contributors on whose expertise we drew so extensively, who gave so much time and effort to the project, and who cooperated with us so willingly throughout the work. It is our hope that they will eventually receive the full returns due them in the form of the gratitude of the many users of this volume in years to come. Herman Krier Martin Summerfield April 1979An Introduction to Gun Interior Ballistics and a Simplified Ballistic Code Herman Krier* and Michael J. Adamst University of Ilinois at Urbana-Champaign, Urbana, Ill. Knowledge of gun interior ballistics has made considerable ad- vances in this century, as the other papers in this volume would indicate. This paper describes the basic physical principles of gun interior ballistics, resorting to as much simplification as possible, so that the key factors that determine the speed of a projectile propelled from a gun may be obtained simply by applying Newton’s force equation to the projectile. But the accurate description of these relatively large forces in time periods of the order of milliseconds is not so straightforward, requiring many untested assumptions. Predictions of the pressure history and projectile velocity for a 175- mm (M113) cannon are compared with other models and with actual gun firings. Nomenclature a axial acceleration of projectile, ft/s? A surface area, in.? A, cross-sectional area of bore, in.? B burning rate coefficient, in./s(psi)” 6c charge weight, lbm ce radiation coefficient ic sliding friction coefficient Cp specific heat at constant pressure, Btu/Ibm-°R. Cc, specific heat at constant volume, Btu/Ibm-°R E emissivity E, total energy from propellant combustion, ft-lbf E, translational energy of projectile, ft-lbf E, heat loss to gun walls, ft-lbf E, kinetic energy of unburned propellant and gases, ft-Ibf ey = ballistic efficiency Invited paper received June 27, 1978. Copyright © American Institute of Aeronautics and Astronautics, Inc., 1979, All rights reserved. Professor, Department of Aeronautical and Astronautical Engineering. agicurently Physical Scientist, Air Force Rocket Propulsion Laboratory, Edwards Air Force » Calif,a0 st z = 1D ODS preva >P : a8 SASS BRE aS NNNX nu won uuu wound H. KRIER AND M. J. ADAMS. piezometric efficiency total resistive force exhibited by projectile, Ibf maximum engraving force, Ibf resultant frictional force normal to land, Ibf mass velocity of propellant gas at solid surface, Ibm/s-ft? unit conversion factor, Ibm-ft/lbf-s? axial moment of inertia, Ibf-s?-ft conductivity coefficient, Ibm-in.?-°R rotational constant, rad/ft effective mass, Ibm number of moles per unit weight, Ibm-mole/Ibm mass of gas, lbm burning rate index average pressure of propellant gas, Ibf/in.? base pressure acting upon projectile, Ibf/in.? breech (or chamber) pressure, Ibf/in.? nondimensional breech pressure maximum breech pressure, Ibf/in.? fraction of rotating band engraved, ft maximum pressure attained in closed bomb, Ibf/in.? heat by radiation, Btu heat by conduction, Btu specific gas constant, Ibf-ft/Ibm-°R_ universal gas constant, Ibf-ft/mole-” effective radius of projectile (including lands), ft relative force relative quickness projectile travel at time f, in. instantaneous exposed grain surface area, in.? absolute temperatures, °R. isocharic flame temperature, °R chamber volume of closed bomb, in.* chamber volume of gun system at any time ¢, in. total volume of gun system at any time ¢, in.? volume of gas, in.? volume of grain time at any ¢, in.? initial chamber volume (no propellant), in.? initial grain volume, in. projectile muzzle velocity, ft/s nondimensional projectile muzzle velocity angular acceleration of projectile, rpm width of rotating band, ft projectile weight, Ibm linear distance of grain recession, in. fraction of propellant consumed at time ¢ constant of engraving, Ibf/ft? initial ramming force on projectile, IbfINTRODUCTION TO GUN INTERIOR BALLISTICS 3 constant of engraving, Ibf/ft? solid-phase thermal diffusivity, ft?/s heat-loss adjustment factor depth of engraving, ft specific heat ratio gas covolume, in.3/Ibm propellant gas density, Ibm/in.? solid propellant density, Ibm/in.? = specific force constant, Ibf-in./Ibm twist, deg dynamic burning rate coefficient STH PSVIOWRN I. Introduction J N any gun system there exist both fixed and variable factors that ultimately specify gun performance. Fixed factors are derived essentially from the gun itself (¢.g., bore length and diameter, chamber configuration and volume, etc.). Variable factors generally are associated with the ammunition, projectile weight, and propellant. However, there exists a marriage to some extent between the fixed and variable factors. For example, the bagged propellant and the projectile both must conform to a specific chamber configuration. Thus, the propellant bags and the projectile are restricted in size and in a sense fixed. As the projectile itself must conform to a specific outer diameter, changes to projectile weight must come from either a length or a mass variation. Obviously, a lighter projectile will exhibit a greater potential for a higher muzzle velocity. However, with the selection of specific projectile dimensions, the propellant volume behind the projectile becomes limited, leaving only propellant combustion as a variable. Operation of any gun-propellant system is limited further by an arbitrarily assigned maximum pressure that insures the gun system against malfunction or damage. Selection of the proper propellant will assure the attainment of a desired muzzle velocity without the violation of the maximum pressure limitation. For an established gun-propellant system, the propellant charge presents the most convenient means of modifying the performance of the gun system without reworking the design of the entire gun system. The problem of at- taining a required muzzle velocity through modification of the propelling charge is quite involved because of the interdependence of various ballistic phenomena. Vast amounts of time and effort are being expanded toward finding a solution to the problem of maintaining a uniform gun performance. The basic propellant used in most modern gun systems commonly is referred to as smokeless powder. (Composite solid propellants, which have nitrate and perchlorate salts as oxidizers, are not used in guns because of the corrosive effects of the combustion gases produced, as well as high temperatures.) Various ‘Smokeless powders display different burning rates, physical characteristics, and shapes. The overall efficiency of a particular smokeless powder as a propellant for a particular weapon can be monitored using the Pressure-time curve corresponding to the gun-propellant system. Assuming a4 H. KRIER AND M. J. ADAMS nonvariance of bore resistance and heat loss, the area under the pressure-time curve is, in general, proportional to the projectile velocity achieved by the combustion of the propellant. Ideally, a smokeless powder would produce a pressure-time curve that would rise quickly to its peak pressure, remain at this value until total propellant consumption, and then fall to zero through adiabatic expansion of the combustion gases, hence maximizing the area under the curve and thus maximizing muzzle velocity. In actuality, a smokeless powder, when burned, produces a pressure-time curve that rises gradually to peak pressure, remains at this value for only a short interval, and gradually drops to zero through adiabatic expansion of the generated propellant gases. Thus, any given propelling charge in a predetermined gun-propellant system Possesses a maximum potential velocity that can be imparted to the projectile, assuming, of course, that the propellant charge functions in an ideal manner. Since ideal behavior in the propellant in unattainable, a gun-propellant system. can be said to be performing quite well with the achievement of 85-90% of its maximum potential velocity. The maximum potential velocity will increase as pressure and charge weight increase. The work reported here attempts to investigate the internal ballistics problem through the development of a computer model of a gun-propellant system. Prior to discussion of the development of such a model, this work presents some general review of such areas as gun system and projectile design, propellant composition and behavior, gun system performance, etc. Following this discussion, a group of governing internal ballistic equations is developed and subsequently applied through various approximations to produce a computer model of a gun-propellant system. Comparison of this particular model to other models is followed by some comments on the overall sensitivity of predicted gun performance to variations in propellant parameters employed by the model. II. A Review of the Ballistics Problem A. Ballistic Cycle One essentially can define a gun system as a projectile-throwing device consisting of a projectile guiding tube to which is connected a reaction chamber. The burning of a solid propellant in the reaction chamber trans- forms chemical energy within the propellant to heat energy. Hot gases produced from this heat energy subsequently expand and act to expel the loaded projectile at a high velocity. Consider now the ballistic cycle occurring in a gun system as illustrated by Fig. 1. Upon ignition of the propellant charge, hot gases are evolved from the burning surface of each grain of propellant. (This statement assumes that all propellant grains have been ignited simultaneously and uniformly. In ac- tuality, this condition is rarely, if ever, realized. Analysis of the ignition and flamespreading problem is quite complex. '* Therefore, this work will proceed always under the assumption that simultaneous and uniform ignition of all propellant grains always occurs. A general discussion of propellant ignition will be given in Sec. III.) The pressure rise in the reaction chamber proceeds rapidly. Because initial resistance to projectile motion usually is quite high,INTRODUCTION TO GUN INTERIOR BALLISTICS 5 firing pin © initiates pro~ 3) Gun chanber pressure pellant combustion / a evolve and heat build rapidly. from burning propellant 4 Q 4 SEO SESS a 6) Rotating band is \ engraved: initial re s) Projectile Steeance co motion causes motion begins. preseure behind the projectile to Thee. 4)“At a pressure from 2-10 Kpsi . - - 1) Due to higher pressure, burning %) Projectile motion 9) Rifled bore causes rate of propellant increases; chanber projectile to spin; increases volume increases uaximim pressure occurs soon after rifling begins Fig. 1 Ballistic cycle. relatively large chamber pressures are attained before significant projectile motion has occurred. The degree of resistance to motion which the projectile exhibits is dependent, for the most part, upon the method by which the projectile is to be stabilized for free flight. Stabilization techniques will be discussed in greater detail later. Movement of the projectile down the gun bore causes the reaction chamber volume effectively to increase. Increasing the volume of the reaction chamber acts to lower the pressure in the gun system. However, since the rate of burn- ing of the propellant charge is highly dependent upon pressure, the high pressure generated before projectile motion begins causes the propellant charge to burn more rapidly. The net effect of the two competing processes is that there is observed a rapid increase in the chamber pressure until a point of maximum pressure is attained. This point of peak pressure usually is reached after the projectile has traveled a short distance down the gun bore. For example, in the 5.56-mm gun, peak pressure is attained at a point where the Projectile has traveled approximately 11% of the length of the bore.6 H. KRIER AND M. J. ADAMS Beyond the point of peak pressure, the chamber pressure begins to drop, At the muzzle, chamber pressure usually is found to be approximately 10- 30% of the attained maximum pressure. Expansion of propellant gases to a point where muzzle pressure would correspond to atmospheric pressure would require a very long bore length. Thus, expansion of propellant gases to atmospheric pressure is not feasible. Pressure at the muzzle is dependent upon the design of the system being used, as well as the type and quantity of propellant charge being used in the system. The magnitude of the muzzle pressure is of importance in gun system design in that this pressure con- tinues to work on the projectile for a short distance beyond the muzzle. Thus, the projectile continues to accelerate beyond its exit from the gun system. B. Gun Systems: General Comments Because there exists such a great variety in gun systems, a complete discussion of all types of gun systems would be quite lengthy. Therefore, this study shall consider only those gun systems that are greater than 30 mm in caliber and utilize bag propellant rather than metal-encased ammunition. Details on gun systems utilizing metal-encased ammunition can be found in Refs. 4and 5. In a gun system, impulse provides the total propulsion of a projectile. This impulse also acts to impart a large amount of momentum to the gun system. Often the momentum imparted by the launching of a projectile is used to do useful work, such as the extraction and ejection of cartridge cases in those systems using metal-encased ammunition. Systems that suffer an imparted momentum are designated as recoil systems. Generally, recoil mechanisms are incorporated to dampen the recoil effects. Another type of gun system uses discharging gases through a nozzle at the breech to counter the momentum imparted to the system upon firing. The net effect of the nozzle is to reduce the external forces on the system to zero magnitude. Gun systems of this type are referred to as recoilless guns. It should become obvious immediately that, for a recoilless gun system, the interior ballistics problem becomes more complicated. One must consider not only the combustion process but also the balancing of the nozzle orifice diameter to the generated recoil thrust in order to maintain no net external force on the gun system. 6? C. Gun Barrel A gun barrel is a simple tube that is closed at one end upon propellant ignition. The barrel provides the support and orientation of the projectile during firing, as well as the proper propulsion on the projectile through the confinement and contouring of generated high-pressure propellant gases. A typical gun barrel is illustrated in Fig. 2. The breech is referred to as that end of the barrel assembly which is opened for projectile and propellant loading but closed during the firing cycle. The opening and sealing of the breech is accomplished by a movable block or plug called the breechblock. Just forward of the breechblock is an enlarged area known as the reaction chamber which initially holds the propellant charge. At the forward end ofINTRODUCTION TO GUN INTERIOR BALLISTICS 7 Key! A= breech E = beginning of lands B = breechblock F = rifled bore C= reaction (or combustion) chamber C= muzzle D = shoulder HW = chase I = slide cylinder Fig. 2 Typical gun barrel nomenclature. 7 the reaction chamber is an area whose walls taper down to the gun bore. This tapered area usually is referred to as a shoulder. The gun bore maintains a constant diameter to the front, or muzzle end, of the gun. The cylindrical afterpart of the gun barrel sometimes is referred to as the slide cylinder. In large systems, this area moves during recoil, on bearings, and into a structure called the slide. The tapered outer barrel just forward of the slide cylinder is referred to as the chase. For gun systems whose projectiles are to be spin stabilized, the bore is rifled. Rifling involves the engraving of a set of spiral grooves into the bore surface. Rifling acts as a guide in inducing the projectile to spin at a specified rate. Rotational velocity of thé projectile will vary directly with the tangent of the developed rifling curve.’ Rifling usually is specified by a characteristic parameter called twist. Twist is specified in terms of caliber per turn, that is, the bore length measured in terms of calibers in which the rifling makes one complete turn. Twist is specified as being either uniform or increasing. In uniform twist, the rifling has a constant angle from the bore origin to the muzzle, whereas for increasing twist this angle varies in accordance with a predetermined exponential curve.” Gun barrels must be constructed so as to withstand the large pressures generated during the ballistic cycle. From the previous discussion, it becomes obvious that the gun barrel must be thickest at the reaction chamber, as this is where the greatest pressure will be realized. As shown in Fig. 3, the barrel thickness strength will diminish toward the muzzle at a fate sufficient to guarantee the structural integrity of the system throughout the entire ballistic cycle. D. Projectile Design and Stabilization As there exists a great degree of variability in gun designs, so too does there exist the same type of variability in projectile designs. Projectile design becomes dependent upon the type of target to which the projectile is to be directed. Obviously one parameter of considerable importance is the Mass of the projectile, as this parameter affects not only the velocity and acceleration of the projectile but also the rate of pressure rise during the early part of the ballistic cycle.8 H. KRIER AND M. J. ADAMS. Curve of Elastic Strength of Gun Pressure-Travel Curve of Gun 20 Maximum Pressure on Projectile Base Pressure x 10°? [Kpst] Projectile Travel [19] Chanber Rey: Fig.4 Typical projectile nomenclature. * 7 Calibers Radius Typically a projectile is designed for stability and minimum drag. For larger guns, the shape of the forward end of the projectile found to give the smallest amount of air resistance was the ogive. An ogive is defined as an arc whose center is on a line perpendicular to the axis and whose radius is expressed in calibers. As illustrated in Fig. 4, in a projectile, the chord of the projectile is the axis in question, with the radius used being around seven to nine times the diameter of the projectile. For smaller projectiles, the ogive shape often is replaced by a cone. Also, the head of a particularINTRODUCTION TO GUN INTERIOR BALLISTICS 9 projectile may be made blunter to give more effective penetration. The body of the projectile is taken to be the cylindrical area between the bourrelet and the rotating band. The bourrelet and the rotating band are slightly raised surfaces above the projectile body. These surfaces act to provide support and bearing, which steadies the projectile during its traverse of the gun bore. These surfaces also act as seals to prevent any leakage of high-pressure propellant gas around the projectile. Such leakage, if allowed to occur, would inhibit the development of the desired projectile muzzle velocity. If the cylindrical shape of the projectile body is continued to the projectile base, the projectile is said to possess a square base. If, however, the afterportion of the projectile body is tapered slightly, the projectile often is described as being boat-tailed. At velocities under Mach 1, boat- tailing reduces the drag due to air resistance and thus increases the projectile’s range. At velocities above Mach 1, boat-tailing causes increased wake dispersion, with a subsequent decrease in performance. Generally there are two techniques used to provide primary stabilization to projectiles. Projectiles launched from a gun system that does not possess a rifled bore employ external fins placed upon the projectile’s base for stabilization in free flight (e.g., a mortar round). The second technique used for projectile stabilization is the one most commonly employed, namely, spin stabilization. Spin stabilization involves inducing the projectile to spin rapidly before it exits from the bore. For spin-stabilized projectiles, the energy of rotation of the projectile involves only about 0.3% of the translational energy of the projectile.* Thus, the effect of the axial moment of inertia upon the overall ballistics of a gun system using spin-stabilized projectiles is small. In spin stabilization, both the rifling and the rotating band placed about the projectile are associated intimately because of their respective functions of imparting torque to the projectile and transmitting this torque. The rotating band is composed of a soft metal that is sealed securely to the projectile body (see Fig. 4). The forward edge of the band is tapered slightly to facilitate the engagement of the band with the origin of the rifling. Before the projectile travels down the bore, the rotating band must Table 1 Energy distribution for a medium-caliber gun® Energy absorbed % of total 1) Translation of projectile 32.0 Rotation of projectile O14 Frictional losses 217 Total work done on projectile 3430 (area under pressure-travel curve) 2) Translation of recoiling parts 0.12 Translation of propelling gases 3.14 Heat loss to gun and projectile 20.17 ‘Sensible and latent heat losses in the propelling gases 42.26 Total propellant potential 100.0010 H. KRIER AND M. J. ADAMS Curve of the Maximum Permissible Pressure Based Upon Tube Strength Pressure [psi] Velocity [ft/sec] Projectile Travel [1n] Comment: work on projectile Area under pressure-travel = —;#°rk on projectste — Work Tait Area Fig. 5 Pressure-travel (solid lines) and velocity-travel (dashed lines) curves. ® or [ras be swagged to the bore diameter and engraved by the rifling. The result of the initial engraving process is to require the gun system to build up a relatively large starting pressure. Ideally, the engraving process would delay the projectile travel until the burning rate and gas pressure are at optimum values. Once the rotating band has been fully engraved, the rifled bore acts to impart spin to the projectile. The primary functions of the rotating band thus can be summarized as 1) to seal the bore to prevent leakage of the generated propellant gas around the projectile; 2) to position and center the rear end of the projectile; and 3) to help transmit a rotation to the projectile. The rotating band therefore greatly affects muzzle velocity, range, accuracy, and gun life. E, Energy Distribution Table 1 gives a typical energy distribution for a medium-caliber gun. The energy distribution for larger gun systems would be approximately the same, with the heat loss to the gun and projectile being slightly lower. It is interesting to note that the energy used in gun system recoil represents only a small part of the total energy of the system. F._Pressure-Travel Curve Every gun system is designed in correspondence to a pressure-travel curve that allows the attainment of a desired projectile muzzle velocity and yet does not allow any violation of maximum pressure constraints on the gunINTRODUCTION TO GUN INTERIOR BALLISTICS nN system. The peak pressure of this characteristic pressure-travel curve ultimately affects total gun performance. Consider the pressure-travel curves as shown in Fig. 5. Equal areas under curves P, and P, indicate that the work performed on the projectile in each case will be equal. This fact, in turn, implies that the corresponding muzzle velocities for the two curves will be equal. To justify this reasoning, con- sider the expression describing the work done on the projectile by the propelling gases: work = |P,A,ds a) Since work done on the projectile is realized as kinetic energy, one can write work= 4 (Mg V3/8.) @) Hence, one can see from examination of Eqs. (1) and (2) that equal areas under two different pressure-travel curves will provide an equal amount of work on a projectile, with this work being realized as equal muzzle velocities in the gun systems corresponding to the two respective pressure- travel histories (assuming M, = const). Examination of Fig. 5 shows that curve P, exceeds the maximum pressure allowed by the strength of the material of the gun barrel. Thus, curve P, becomes unacceptable from a pressure standpoint, although theoretically it should give the same ballistic performance as curve P,. To increase a gun system's muzzle velocity, one must increase the area under some new pressure-travel curve (P,) over that area under a curve corresponding to a gun system with a lower muzzle velocity (P,). Note in Fig. 5 that curve P, does not violate the pressure constraints on the system. One could speculate from Fig. 5 that the optimum pressure-travel curve would be that path which corresponds to the curve of maximum tolerable pressure of the gun barrel. However, if it were possible to design a propellant giving such performance, one would find problems in excessive erosion of the gun bore which would act to decrease the system’s durability. Furthermore, because a higher muzzle pressure would exist for such a system, excessive muzzle flash and nonuniform muzzle velocities would be created. Such a system also would demand a greater reaction chamber volume, which would increase the system’s overall weight while decreasing its mobility. Thus, suitable designs of gun systems can be thought of as those designs that give uniform muzzle velocities from round to round oe violating maximum pressure constraints at any point along the gun arrel. The characteristic pressure-travel curve of a gun system is dependent upon many factors, the most important of which are® 1) variation in chemical composition of a gun propellant, 2) variations in the burning rate er a gun Propellant, | 3) ignition characteristics, 4) propellant grain haracteristics, 5) loading conditions (charge weight variations), 6) en- vironmental factors, 7) variations in gun-projectile system, and 8) variations in engraving resistance profiles. As an illustrative example, Fig. 6 shows the influence of several of the factors just listed on the pressure-travel curve.12 H. KRIER AND M. J. ADAMS In region 1-2 of the pressure-travel curves shown in Fig. 6, the slope of the curves is affected primarily by the ignition characteristics of the gun system, as well as the total area of the propellant grains initially exposed to burning. The shape of region 2-3 is governed primarily by propellant grain configuration. Region 3-4 is governed primarily by the rate of reaction chamber expansion and hence is influenced strongly by such factors as engraving and projectile weight. Assuming that proper propellant ignition has been attained, one can say that the interior ballistics of a particular gun system is affected essentially by variations in the physical and chemical character of the gun propellant. G. Efficiencies of a Gun System and Charge Two parameters commonly are employed to gage the overall performance in a, particular gun-charge-projectile system. These parameters express gun system performance by use of a piezometric efficiency and a ballistic ef- ficiency. Piezometric efficiency e, is found by dividing the mean pressure by the maximum breech pressure.? The mean pressure is taken as that pressure which, if uniformly exerted upon the projectile over the entire length of the bore, will produce the observed muzzle velocity. One can express e, as =4W,V;/8 ASP am @ In general, the higher the piezometric efficiency, the flatter the pressure- travel curve becomes. Pressure [pei] Travel [1] Key: A= Strong ignition (high initial surface area) B= Weak {gnition (low initial surface area) C= Rapid change in surface area (many holes or small grains) D = Less rapid change in surface area (fev holes of larger grains) E = Rapid chamber expansion (light projectile, less rotational restatance) ¥ = Less rapid chauber expansion(hesvy projectile, large rotational resistance) Fig.6 Parameter influence upon pressure-travel curve. ®INTRODUCTION TO GUN INTERIOR BALLISTICS 13 Ballistic efficiency e, can be defined as the ratio of the total work done upon the projectile to the total work potential of the propelling charge.® Thus, one can write an expression for ballistic efficiency as y= AME V3 (y—1)/80C “ A high ballistic efficiency can be obtained by burning the propellant charge as early as possible in the projectile’s traverse of the gun bore, thereby reducing residual muzzle pressure. Ill. A Review of Gun Propellants A. Gunpowders Smokeless powders often are referred to as gunpowders, although they are neither completely smokeless nor powders. The basis for the manufacture of modern gun propellants is generally nitrocellulose, a substance produced by the action of nitric acid upon cotton. Early attempts to use nitrocullulose as a propelling agent often resulted in dramatic failures, as nitrocellulose alone is a highly explosive substance. The solution to the stability problem was to colloid nitrocellulose with ether and alcohol. The result of this process was the formation of a suitably stable gun propellant of a low burning nature whose rate of combustion could be controlled. Furthermore, the new propelling agent was found to be ef- ficient, safe to use, easier to handle, and fairly stable in various storage environments. Table2 Comparison of single- and double-base propellants * Single-base Double-base 1) Color “Amber, brown, or black _ Gray-green to black 2) Controlled Can be controlled to Can be controlled as with single-base burning maximum efficiency 3) Ignition Ignites around 315°C Ignites at around 150°-160°C temperature 4) Sensitivity Ignition is difficult, Detonates more readily than single-base, yields may detonate if burned higher potential and more heat (high explosive) in large quantities (low explosives) 5) Stability Can be made with Can be made stable as with single-base addition of stabilizing ingredients 6) Residue Some residue and smoke Little residue since there is less inert material, some smoke 7) Manufacture — Complicated but safe, Complicated and more hazardous, plenty of plenty of raw materials raw materials 8) Erosiveaction Erosion of bore, More erosive than single-base because of adiabatic flame tem- higher temperature and heat of explosion erature at 2400-3000K 9) Flash ‘Caused by hot gases Increase in flame temperature increases ten- igniting with oxygen at dency to flash over single-base muzzle, controllable14 H. KRIER AND M. J. ADAMS Smokeless powders are considered to lie within one of two mayor classes, namely, single-base and multibase propellants. In a single-base gun propellant, the only explosive present 1s nitrocellulose. All other ingredients present in a single-base propellant are used primarily for stability and burning rate control. Multibase gun propellants are distinguished further as being either double-base or triple-base propellants. Both of these multibase propellant types contain nitroglycerin to facilitate the dissolving of the nitrocellulose, thereby enhancing the propellant’s explosive qualities. Double-base gun propellants produce less gas than single-base gun propellants but retain a higher thermodynamic efficiency because of the production of more heat at higher flame temperatures. However, since double-base propellants burn at higher temperatures, the rate of gun bore erosion is increased substantially. In an effort to achieve the effects of single-base gun propellants while maintaining the manufacturing ease of double-base gun propellants without incurring the high erosion quality of double-base propellants, a third ex- Plosive, nitroguanidine, is added to the propellant mixture. The addition of this third explosive forms the class of multibase gun propellants designated as triple-base propellants. Nitroguanidine is a ‘‘cool-burning”’ explosive that keeps the maximum temperatures of the triple-base propellants comparable to those of the single-base type, and yet the triple-base propellants maintain a greater propulsive potential. B. Characteristics of Gun Propellants Gun propellant grains possess a hard, smooth finish. As the grain ages, its color changes from translucent amber to dark brown, to black, and finally to opaque. It has been observed, however, that stability of the grains is not lost through these color changes. Chemical decomposition of gun propellants is not too rapid but, if allowed to occur, may result in spontaneous combustion. Gun propellants are in a state of unstable chemical equilibrium and thus are affected readily by any impurities within them. If any part of the propellant decomposes, the products of decomposition, being of an acidic nature, act to increase the rate of decomposition. To lessen this problem of decomposition, diphenylamine is added to the propellant to counter the acidic products of propellant decomposition. However, if partial decomposition of a gun propellant has occurred, the propellant is considered unfit for gun usage, since part of the decomposition, which normally is caused by the sudden evolution of hot gases within the gun, already has taken place. Thus, the propellant has lost a corresponding number of heat units. Stability of gun propellants is relatively unaffected at temperatures below 60°F. Above temperatures of 70°F, the decomposition rate of a gun propellant increases sharply, with decomposition rate becoming quite high at temperatures around 90°F. Thus the storage of a gun propellant requires an area of uniform, low temperature. Decomposition of a gun propellant also is enhanced by the presence of moisture. Therefore, gun propellants usually are stored in air-tight con- tainers. A leaky container not only will admit moisture but also will allowINTRODUCTION TO GUN INTERIOR BALLISTICS 15 volatiles to escape through evaporation, especially if the surrounding air is subjected to changes in temperature. One desires to maintain the volatile integrity of a gun propellant, as the loss of volatiles may act to increase the burning rate of the propellant to such a degree that pressures within the gun system may become excessive. Triple-base gun propellant is by far the most stable propellant because of its relatively low nitrocellulose content 9%), extremely small content of volatile components, and low hydroscopicity. Gases evolved from a triple- base gun propellant are less erosive because of the presence of a large percentage of nitroguanidine (=55%). Triple-base propellants also burn more completely, leaving less residue material in the gun system after firing. Finally, although usually requiring more ingredients to manufacture, triple- base gun propellants cost less to produce. Table 2 gives a summary of the comparison of single-base and multibase gun propellants. A discussion of the manufacturing processes of a single- base gun propellant can be found in Refs. 10 and 11. Granulation and shaping are important parameters in controlling the burning rate of a specific gun propellant. Therefore, gun propellants are manufactured in many different shapes, including flakes, strips, sheets, pellets, and perforated grains. Perforations in propellant grains are used for controlling the rate of gas liberation as well as the burning time of the propellant grains. Figure 7 illustrates, for several types of propellant grains, the exposed burning area as a function of the percent of the grain con. Table 3 Physical properties (dimensions) of propellants (Dried grain dimensions) 080) 950 web Single perforated (SP) Multiperforated (MP) (web = 0.25 (D-3d)) Use Type Lyin. D, in. Web, in, d= perforated diam 175-mm M6 MP 1.024 0.439 0.077 E . . 0.046 105mm Me 0.369 er 0.165 0.0288 0.016 105-mm MI MP 0.32 0.143 0.024 . . . 0.016 155mm MI MP 0.22 0.049 0.015 0.019 $S-mm MI MP 0.409 0.183 0.034 0.016 MI SP 0.214 0.051 0.017 0.016 MI MP 0.525 0.229 0.042 0.02 105-mm, M30AL 0.622 0.299 0.041 0.03216 H. KRIER AND M. J. ADAMS s+ 1.0 A sy, 0.2 04 0.6 0.8 1.0 z Key: A= chord shaped grain D ~ Single-perforated grain B = square shaped grain E = Multi-perforated grain C = Strip shaped grain Fig. 7 Change in the ratio of instantaneous surface area t fraction of propellant burned (Z). I grain volume (5,/Vq,) with sumed. For the ballistic study undertaken in this investigation, only cylindrical perforated grains were considered. Cylindrical perforated grains are made in various lengths and diameters, depending upon the gun system in which the grains are to be used. Table 3 lists the average physical dimensions of various cylindrical perforated grains in use in some of the more common gun systems. C. Ignition of Gun Propellants It has been stated that the problem of modeling propellant ignition characteristics is quite complex. However, one can make some general comments concerning the ignition of a gun propellant. Gun propellants are ignited essentially by high-temperature heat transfer by hot propellant gases as opposed to sudden shock. A shock to a gun propellant could cause a detonation rather than the desired burning of the propellant. Generally, ignition of a gun propellant proceeds with the use of an explosive train consisting of a primer, ignitor, and propellant charge. Upon ignition, heat flows from a hot primer flame to the propellant grain in the following sequence*: 1) blow from the firing pin... 2) detonates a small sensitive explosive in the primer... 3) causing the primer to ignite... 4) where the generated heat and flame trigger the ignitor; the ignitor acts to amplify the previously generated heat and flame... 5) causing the sub- sequent ignition of the relatively large and insensitive propelling charge; generation of propellant gases then can proceed. This sensible heat, plus any heat due to the adiabatic compression of hot gases in the vicinity of the primer flame, is the sole source of heat made available for ignition. Heat flow to the main charge proceeds primarily by two methods of transport, namely, conduction and radiation, Where one body surrounds the other, the net radiation between the two bodies may be written as!?INTRODUCTION TO GUN INTERIOR BALLISTICS 17 4 -ca{(35)'- (Ho) dt eal (T05 700 6) which implies that the rate of radiate heat flow varies with the area, as well as the fourth power of the temperature of the radiating body. Most solids possess an emissivity much higher than that found in gases. For luminous flames possessing large numbers of solid particles suspended in the flame itself, radiate heat transfer will be much more intense. For this reason, black powders generally are used in gun primers. The products of com- bustion for this type of primer contain a large amount of solids such as potassium carbonate. These solid materials radiate a more intense heat, as opposed to the smokeless powders found in nonluminous flames. Heat flow by conduction from a hot gas to a solid surface can be written as? 4 G8 Ge =KC AT} ae (7,-T) © where D is the average diameter of the solid. From Eq. (6) one can see that the temperature, mass, and velocity of the propellant gases being generated are important. When the hot primer gas is enhanced by the presence of incandescent solids, radiation effects not only act to augment the con- duction process but actually exceed those effects of conduction. Ideally one would like to ignite all propellant grains at the same time. This task would involve instantaneously bringing all of the propellant grains into contact with the primer flame, which, in turn, would require primer to develop such a flame. However, large primers would require so much black powder that the firing would not be smokeless. Furthermore, packing more powder into the designated area of the primer also becomes undesirable in that the flow of the hot primer flame during the ignition process becomes hindered. Thus, the ignition process itself may become irregular. Theoretically, the most satisfactory primer would consist of, say, an ex- plosive gas that would penetrate the entire explosive charge and liberate solid particles. D. Burning of Propellant Grains The burning time of a cylindrical perforated propellant grain can be controlled by several means, namely, 1) the size and shape of the propellant grains; 2) the number of perforations in each grain; 3) the web thickness, i.e., the amount of solid propellant between burning surfaces; 4) the quickness,t i.e., rate of burning of the grain; and 5) the percentage of volatile materials, inert materials, and moisture present; a 1% change in volatiles in a low-volatile-content propellant may cause as much as 10% change in quickness. ¥The quickness of a propellant is a relative term only. It expresses the rate of burning of one Propellant to other propellants. A quick propellant will burn faster and produce a higher ressure in a given gun system as opposed to a slower-burning propellant.18 H. KRIER AND M. J. ADAMS. Instantaneous Surface Tnitial Volume Grain = M6 Single Perforated Chord Type o 2 “4 6 8 1.0 Fraction of Propellant Burnt Fig. 8 Ratio of instantaneous surface area to initial volume vs Z: _single-perforated grains. The burning rate of a confined solid propellant is directly dependent upon temperature and pressure. As temperature and pressure increase, the rate at which the propellant burns increases. Controlling the burning rate of the propellant becomes essential if one is to keep a gun system from violating the pressure constraints on its material. At constant pressure, the rate of burning of propellant grains is proportional to the amount of surface area exposed to burning by each grain. Thus, a propellant charge is composed of accurately sized grains of a particular shape. The rate of evolution of gas from a burning propellant depends upon the surface area being exposed to the burning. For a given mass of propellant, the initial burning surface will depend upon the form and dimension of the individual grains. As burning continues, the rate of combustion as well as pressure variations will depend upon how the exposed surface area of the grains is changing, that is, whether the exposed surface is increasing or decreasing as burning continues. Propellant grains are defined as being either regressive, neutral, or progressive burning grains. A regressive burning grain is defined as a grain whose total exposed surface area decreases with burning. Cord, strip, and ball propellants characterize a regressive grain configuration. ‘A neutral burning grain is defined as a grain whose total exposed surface area remains unchanged during burning. As shown in Fig. 8, a single perforated cylindrical grain with a large length/web ratio approaches a neutral burning character. As the outer exposed surface area decreases during burning, the inner exposed surface area increases. The net result is that there is no observable change in the net area of exposed burning. Note that, as the length/web ratio becomes smaller, this particular type of grain configuration becomes regressive. Progressive burning grains are defined as those grains whose total surface area increases with burning. Multiperforated grains are considered as progressive. Figure 9 illustrates the burning character of a seven-perforatedINTRODUCTION TO GUN INTERIOR BALLISTICS 19 10 Grain = M6 Seven Perforated Chord Type Instantaneous Surface Area Initial Volume o 2 4 6 8 1.0 Fraction of Propellant Burnt Fig. 9 Ratio of instantaneous surface area to initial volume vs Z: _seven-perforated grains. cylindrical grain of various length/web ratios. Notice that, as the length/web ratio becomes smaller, the burning transforms from progressive to neutral to regressive in character. The advantages of using a progressive multiperforated grain do not come without some penalties. If a multiperforated charge is not fully consumed, as may be the case for smaller changes, portions of unburned grains in the form of slivers may remain which will be ejected from the gun upon firing, thus increasing the possibility of obtaining irregular muzzle velocities, muzzle blast, and flash. Being an undesirable feature, the slivering problem in multiperforated grains may be relieved somewhat by using rosette-type grains. E. Effects of Grain Characteristics on Gun Performance As stated previously, performance in a gun system can be analyzed using the pressure-travel curve. The shaping of the characteristic pressure-travel curve for a gun system is dependent upon several grain characteristics. Although, in the final design of a gun system, all grain characteristics may act simultaneously to determine the ballistic performance of the gun, it becomes convenient to catagorize the independent effects of several grain characteristics on the pressure-travel curve. As shown in Fig. 10, changing the grain configuration to a more Progressive design (under the restrictive assumptions that all grain types Possess the same initial surface area, composition, and charge weight) results in a lower peak pressure and higher muzzle pressure. Note that, since one has assumed the initial surface areas to be equal for different grain types, the regressive grains must be the smallest of the grains con- sidered. As shown in Fig. 10, if one holds the parameters of charge weight, composition, and configuration constant but varies grain size, the larger grains will give much lower peak pressures with higher muzzle pressure than the smaller grains. This effect is to be expected, as one recalls from Figs. 8 and 9 that, as grains become smaller, their burning character becomes regressive.20 H. KRIER AND M. J. ADAMS Observed Variation Comment Small Grain charge weight equal in all Preasure Travel Degressive Grain Independent variation 4m grain configuration; charge weight equal in all cases Weutral Grain Progressive Grain Pressure Travel High A Independent variation Medium 4 in loading density (A) grain size and conf: Low & tion equal in all Pressure Travel Fig. 10 Sensitivity of pressure-travel curve with parameter variation. © One can conduct similar investigations to predict the effects of variations in composition and web thickness on the pressure-travel curve. These relationships, coupled with those of grain size and configuration, become useful design tools when applied to the task of designing a specific gun system. For example, hand-held guns would require minimal muzzle blast but at the expense of reaching high pressures. Thus, these gun systems characteristically would use small regressive propellant grains. On the other hand, large gun systems require lower peak pressures. Thus, these gun systems generally employ large neutral (or progressive) propeliant grains. F. Energy of Gun Propellants: General Comments Ignition of a gun propellant provides a complex mixture of several gases. Furthermore, each propellant formulation is known to have a characteristic explosion temperature. If one assumes that all changes in equilibrium of the generated gas mixture occur equivoluminar and independent of density, the decomposition of a unit mass of propellant always will liberate the same amount of energy, which then heats the product gases to the same tem-INTRODUCTION TO GUN INTERIOR BALLISTICS 21 perature. (The assumption used implies that the explosion temperature is not high enough to cause dissociation of the main constituents of the gas mixture. The assumption seems valid for ‘‘cool’? propellants whose un- cooled explosion temperature is not greater than 3000K.°) Defining 7, as the isochoric flame temperature, one can write the energy released by the decomposition of a propellant grain as A=NRTy @ One can determine the magnitude of the force constant ) experimentally by burning a known propellant charge in a constant volume chamber and measuring the maximum pressure produced. It is standard practice in interior ballistic theory to assume an equation of state of the simple covolume type, namely, the Noble-Abel equation of state. This equation can be expressed as (Pin) (V=n'n) =RTp (8) Assuming that the covolume 7 is known accurately (the covolume can be determined by firing a series of charges of different masses and measuring the corresponding maximum pressure) and knowing the chamber volume and maximum pressure achieved for a specified charge, one can use Eqs. (7) and (8) to find A as (assuming total propellant combustion) A= (Pmax/¢) (Veen) 0) IV. A Ballistic Model A. Approach to the Problem There have been many ballistic models introduced which theoretically predict, with varying degrees of success, the interior ballistics of a gun system. '3.!4 To obtain a complete picture of the ballistic cycle, one employs a ballistic model centered around a Lagrangian frame of reference.'° [In a Lagrangian reference frame, one follows individual fluid elements in the course of time to obtain a picture of the flow process being examined. Although the Lagrangian approach appears to be the natural way to set up problems of fluid motion, it is not as convenient or meaningful as the Eulerian description of flow processes in which fluid motion is observed by specifying and analyzing various scalar and vector fields. A Lagrangian formulation often gives more information than one really needs. However, in certain one- dimensional problems (such as found in most gun simulations), the Lagrangian point of view becomes quite useful in observing such phenomena as pressure oscillations between the breech and projectile base during the early portion of the ballistic cycle. For a more complete discussion on Lagrangian and Eulerian methods, the reader is referred to Ref. 16.] However, although a model of this type gives a more extensive view of the ballistic cycle occurring in @ gun system, the model usually involves considerably more computation time on a computer. In the ballistic model presented in this study, the primary objective was to develop, using Newtonian mechanics, a ballistic model with accuracy com-22 H. KRIER AND M. J. ADAMS parable to that obtained from much more elaborate ballistic models. A simplified ballistic model as presented here possesses the advantage of beiig able to predict gun performance using considerably less time in computation. However, in simplification of the ballistic problem, one only can hope to observe overall relative variations in various gun systems. Internal phenomena, such as ignition process and initial pressure oscillations, will not be predicted when one applies the simple ballistic model proposed in this study. Development of a simple model of gun behavior proceeds through con- sideration of the following areas, namely, 1) Noble-Abel equation of state, 2) form function analysis, 3) propellant burning rate equation, 4) variable volume considerations, 5) pressure considerations, 6) projectile motion (Fisher-Trippe engraving model), and 7) heat-loss effects. B. Equation of State . As stated in the preceding section, it is generally standard practice in the formation of interior ballistic theory to assume an equation of state of the covolume type, specifically the Noble-Abel equation of state. Recall that this relation can be expressed as (Paice) (V, en) = (10) One can rewrite Eq. (10) as P4l(1/p,)—1] =X ay When using Eq. (11) in ballistic theory, one assumes that generated propellant gases are well mixed. Use of Eq. (11) requires that one know the propellant gas density. This parameter can be expressed in terms of the solid propellant density, an easily measured quantity. The propellant gas density can be defined as __mass of gas mixture an Pe Volume of gas mixture The mass of the generated propellant gas is equal to the mass of the solid propellant burned, namely, mass of gas mixture=CZ(t) (3) The volume that the generated propellant gas occupies can be expressed as V, YO (Choy U-Z)) (14) From the preceding definitions, one can express the propellant gas density as CZ(t) a (15) V(t) — (Clos) (1-Z(t)] Ps23 INTRODUCTION TO GUN INTERIOR BALLISTICS Jou PT _7A eX(U-1)9+x[9p+ (U-1)TIZ— [P+ P]Z *S njos oF yaworddy “sures qwe|jadosd jo SuL9aHs 01 dn suo PIOY seIMUOY “(JeNb9 212 sassaUyDNI q>M I 11 pus p *s}2q ut punoy aq eo wia}qoud Bu94q15 ) Atjeotsiowuuks paoe|d ase sofou wy) aUNssy, iu :1=P pur Ju + y=2 a104m (suonesojsad w yas PT ures yestapuryéo) =Z —yeinunoy pesausH eX(U=1) 7+ 2x[96+ (U1) 7) —X[P + TT 2ie~ pM =9 pure iz + y=no104MH 0 ai = ud aT =Z paresoysad-uanag 2X9E-x(TEt+DZ]p—(G+T0]z *s eXET~ X(TE=PZ1E—X19 + TOT 2, ae (ee) AT (Ae Z)=z poresojsad asus we Nee ‘s ob cee, oy ) ze A Teo =Z px0y X9-XE(T+ NP) —(Te+ WZ *S eXEt 2X(T+ Ub) —X( TE + eH)Z auinjon uresd [en 79uit) 91ye adky, Hures8 Jo kaze aoejins snoaueIUeISUT quang a818yp Jo uonswsy saysnowoas uyes3 je>qdsy autos 405 74 /*5 pu Z 405 senuuog p a1Q8L,24 H. KRIER AND M. J. ADAMS To obtain an expression for the gas pressure generated from a burning solid propellant in a variable volume system, one begins by rearranging Eq. (11) to obtain P. MU C/eg)— 1) (16) Substitution of Eq. (15) into Eq. (16) yields ae ACZ() 4° (HO (Che) UZ 1-120) or ACZ(t) 7 po 17 VA Clie.) +2 a= H7a,)11 uy Pa Note that use of Eq. (17) requires a knowledge of both the changes in chamber volume and fraction of propellant consumed as functions of time. C. Form Function Analysis Mathematical determination of a relation that takes account of the effect on gas evolution of the changing burning surface area commonly is referred to as form function analysis. For simple grain geometries, such as sheets, cords, or single perforated grains, the desired expression is derived rather easily. For more complex geometries, such as a seven-perforated grain, these derivations become more involved and less exact. For simplicity, determination of the form function for a particular grain geometry will proceed under the restrictive assumption that, upon burning, all exposed grain surfaces recede at the same rate. With this assumption, one can define a simple relationship between the fraction of propellant consumed Z(t) and the distance that each exposed grain has recessed. An expression for Z(t) is obtained from considering the initial geometrical shape of a typical grain of the propellant charge. One can define Z(t) as zr) [Vay (1)/Voe] ) In considering how a particular grain recedes with time, one realizes that the greatest importance lies with the least linear dimension D of the original geometrical shape of the grain. If after a time f the fraction of D remaining unburnt is defined as f, then the smallest dimension after a time ¢ becomes Df. Examination of Eq. (18) shows that the volume of the propellant grain is cubic in D. Thus, after a time ¢, the fraction of propellant unburnt becomes cubic in f and is denoted by 0(f), which is called the form func- tion. Table 4 gives an expression for the fraction of propellant unburnt Z(t) and the exposed surface area/initial grain volume S,/V, for several types of propellant grains. For perforated grains, the least linear dimension is defined as the linear distance that the propellant recedes normal to the grain surface. The critical dimension for perforated grains becomes the web thickness, as thisINTRODUCTION TO GUN INTERIOR BALLISTICS 25 dimension, with the exception of multiperforated grains, dictates when the propellant grain will be consumed completely. In nonrosette multiperforated grains, there remain slivers of propellant after the web has burned through. The regressive burning of these slivers tends to reduce the progressive character of early burning of the grain. Exact mathematical determination of the regression of these slivers is difficult. D. Burning Rate Equation The rate at which the surface of a solid propellant grain regresses is given by the burning rate equation. It has been found experimentally that the burning rate of a solid propellant is dependent primarily upon the pressure of the gas under which the combustion process proceeds. Exact mathematical determination of this pressure dependence has not been made, as the combustion process itself is quite complex. Burning rate experiments most often are conducted under conditions that are far different from those encountered in a gun system. Therefore, extrapolation of formulated burning rate laws to a gun system usually yields poor results unless some adjustments are made to account for dynamic effects that burning propellants actually encounter in a gun system. Various forms of a burning rate law have been proposed !79; however, as a first approximation, this study uses the steady-state burning law to model the recession rate of a burning propellant in a gun system. The steady-state burning rate law is expressed as ax a 7 BPA as) Determination of the constants B and n can be made from examination of data obtained from either closed bomb or strand burner data. Examination of Eq. (19) shows that the burning rate is dependent upon the average pressure of the gas mixture. Recall that, for the simple ballistic model proposed in this study, average pressure is calculated under the assumption that all exposed surfaces recede at the same rate, implying that the propellant grains retain their shape during burning. (The implication here is not that the surface area remains unchanged but rather that the grains “shrink”? symmetrically.) However, in actuality the burning rate of a propellant grain is influenced by the flow of gas over and within the per- forations (if any) of the grain. This gas flow acts to destroy the symmetry of the grains during burning. Therefore, the assumption of constant shape of propellant grains during propellant burning is somewhat ideal. As suggested by Krier," a first approximation to a dynamic burning rate of a solid propellant can be written as ax an dP, a =eri{i ty sie S4) (20) where y is a function (unknown at this time) that is dependent on pressure and flame structure. This particular burning rate law has been used by26 H. KRIER AND M. J. ADAMS Osborn, !” who chose ¥=1. Since y is a function of pressure, the value of y is changing constantly as the propellant burns. Exact dependence of ¥ upon pressure is not known. However, as plots of dP,/df vs burning rate ob- tained from closed bomb firings are not exactly linear, future consideration of a dynamic burning might be warranted. E, Variable Volume Considerations Examination of Eq. (17) shows that one must define the volume of a gun system as it changes with time if one is to determine the pressure of the propellant gas mixture. However, the rate of change of the volume of a gun system is dependent upon the pressure acting upon the base of the projectile. The volume of a gun chamber at any time f can be defined as rd, Ve(=V, j £ di 1 (ot) = Vag + Cartes 1) Pressure generated from the burning of the gun propellant does work on the projectile base. This base pressure acts to change the volume of the chamber. One may write AP, _ od aL, sW, * dv, de Thus one may write A,—# (22) Force projectile otating a Band of Rifling Soments: 7, = resultant force normal to the edge of a land F, ™ resistive force composed of the engraving-press fie sliding forces and the frictional resistance forces Fig. 11 Free-body diagram of forces acting upon a projectile during projectile motion. 3INTRODUCTION TO GUN INTERIOR BALLISTICS 27 The rate of change of the volume of the chamber can be found from in- tegration of Eq. (22). Therefore, one may write Ve =VatA,)” Vide (23) F. Pressure Considerations During projectile travel along the bore of the gun, there arises a pressure gradient in the gas being generated from the burning propellant behind the projectile. The pressure gradient appears from hot gas and unburned propellant being accelerated along with the projectile. Friction at the bore surface also contributes to the formation of the pressure gradient. The result of the pressure gradient is that the calculated average pressure of the gas mixture, P,, must be adjusted to obtain a suitable value of base pressure P,. Furthermore, pressure in a gun system usually is measured at or near the breech. Hence, measured pressure usually is expressed as breech pressure Ppp. Acommonly used expression that relates base pressure to average pressure is given by? 2 Ye) -W- ea Py: Pyal: (14 2 ae 7-1 Bd The approximation shown in Eq. (24) is based upon a special solution to the Lagrange ballistics problem. Equation (24) assumes that the gases and unburnt propellant grains can be considered as a well-mixed fluid of constant density. This restriction implies that projectile velocity varies linearly from zero at the breech to a value V, at the projectile base. Use of Eq. (24) is restricted to guns firing at moderate muzzle velocities (<2000 ft/s) if computational accuracy is to remain unchanged.” Aconversion of breech pressure to base pressure can be defined as” P,: P, =[1473! my“, [i 4] 5) i 2 gd : BY Thus, with a means of theoretically converting one pressure to another, one can take into consideration that base pressure works to change the gun system volume, whereas a higher average pressure controls the gun propellant burning rate. G._ Projectile Motion Projectile motion proceeds under the influences of two types of forces, namely, propulsive and resistive forces. Consider the free body diagram shown in Fig. 11, The single force acting on the projectile to provide propulsion is the Pressure acting over the bore cross-sectional area. As suggested by Fisher and Trippe, ! resistive forces can be defined as 1) force required for engraving the rotating band, 2) drag resistance from Projectile motion through the barrel where a metal-metal pressfit situation exists, 3) propulsive forces used for angular acceleration, and 4) frictional sliding resistance from the resolved force normal to the edge of the land.28 H. KRIER AND M. J. ADAMS Following the resistive model of Fisher and Trippe," the resistive force as shown in Fig. 11 can be resolved as Fp=C)FatFo (26) where F, represents engraving and sliding forces (Ibf). The engraving and sliding force can be resolved further as Fp =0 for S, =< 0 (27a) Fo = Z)+Z,bPrS, for S, < Wop (27%) Fo = Fomax—ZaPe(Sp—Wos) for Sp > Won (27) The Fisher-Trippe model of engraving and sliding force is based upon data obtained from experiments conducted at Calspan to measure forces required to extrude brass and aluminum bar stock. Although this engraving model was formulated specifically for the 175-mm gun system, appropriate adjustment of the given constants in the model provides a reasonable first approximation of resistive forces encountered in other gun systems. Table 5 lists, for several gun systems, the various parameters used for the Fisher-Trippe resistance model. From examination of the free body diagram in Fig. 11, one can define axial acceleration of the projectile as aW, =P,A,—F,sin0 — Fycosd (28) Similarly, one can define rotation acceleration of the projectile as le= [ka= R’ [Fycosd—Fpsind) 9) Table5 Typical engraving data for several gun systems Parameter 175-mm 155-mm ‘Axial moment of inertia, Ibf-s?-ft 0.210 0.108 Rotation constant, rad ft 0.546 0.592 Sliding friction coefficient 0.600(¢)" 0.600(€) Twist, deg 9.0 9.0 Portion of rotating band 1.316 0.694 engraved, ft Depth of engraving, ft 6.010 -> 4.83x 10-3 Length of rotating band, ft 0.146 0.083, Effective radius of 0.290 0.255 projectile, ft Initial ramming force, Ibf 5.0x 104(e) 3.0 104(e) Maximum engraving force, Ibf, 7.6 x 104(e) 5.9x104(e) Constants of material,” Ibf ft? Ly 2.29107 1.25107 Za 2.29% 106 1.25% 106 Heat-loss coefficient Le ee aa *(e) denotes estimated values. > © Engraving data follow model of Fisher and Trippe.INTRODUCTION TO GUN INTERIOR BALLISTICS 29 With the elimination of F, from Eqs. (28) and (29) and subsequent sub- stitution of Eq. (26), one can express the axial acceleration as R’LPpA, (cos —C,sind) — Fy] [W,R’ (©os8 = Cysinb) +1k (sind + C,cos8) } (30) Subsequent integration of Eq. (30) will yield projectile velocity and travel readily. Thus, v, fe adr G1) where V, (0)=0, and i; v,dr (32) where S,(0) =0. H. Heat-Loss Effect Throughout the ballistic process in a gun system, a portion of the available energy generated from burning propellant is lost by heat transfer to the bore surface. Various methods have been developed to account for this heat loss?!; however, the approach taken for this investigation was to adjust the value of the specific heat ratio y upward by a constant 8 such that the total estimated available energy of the propellant gas is reduced. 62° As outlined by Heiney, the energy equation of a gun system can be written as E,=E, +E; +E, (33) Thermal and chemical energy released by a gun propellant can be expressed as E,=CZ(t)C,(T)-T) (34) One can define the specific heat ratio ‘y by the following expression: (y-1) =R/C, =¥/C,T9 35) Translational energy of the projectile can be written as E,=W,V3/2 (36) whereas the heat loss of the propellant gas to the bore surface is approximated as E,= 46M, V3 @7) where Me=W,+[CZ(t)/3] (38)30 H. KRIER AND M. J. ADAMS The kinetic energy of the accelerating propellant gas and unburned propellant can be written as E,= 2 (CZ(1) ¥3/3) G9) Summation of Eqs. (36, 37, and 39) yields E, +E, + E,=((1+8)/2)MeV, (40) One now may write Eq. (33) as C2(t)C, Tol = (T/T) 1 = [UL +8)/21Me V3 (41) or, upon substitution of Eq. (35), C2(t)AU1 = (7/T 9) = LB 1) +8) /21M V3 (42) Elimination of the temperature ratio in Eq. (42) is accomplished by using the equation of state and the definition of the specific force constant \. Recall that Z T p,{v.cn-c[ +2 1-2) ]} 0.056 Specific heat ratio 1.254 Specific force constant x 10 ~$, Ibf-in./lbm 3.804 Burning rate coefficient, in./s(psi)” 0.005 Burning rate index 0.680 9) Heat loss to the gun surface can be accounted for adequately by applying an adjustment factor that reduces the available energy released by the burning propellant. 10) The engraved rotating band forms a perfect seal with the gun bore. 11) Engraving forces can be represented by a linear increase from an initial value followed by a linear decrease to zero from the peak value. 12) Sliding friction losses between the engraved rotating band and the gun bore are taken to be negligible. 13) Boundary-layer growth in the gun bore is ignored. 14) A special solution to the Lagrange problem is used to relate average pressure to projectile base pressure. J. Encapsulation of Governing Equations The following equations and initial values are employed to construct the ballistic model used in this investigation: Propellant Burning Rate Law dx — =BP, (47 dt 4 (47) Projectile Acceleration av, R’[P5A, (cost — Csin8) — Fp} «sy dt — [W,R(cosd ~C;sin8) + IK (sind + C,cos6) ] Projectile Travel ds {" ah) Saya (a 49) dt? Jo\ drINTRODUCTION TO GUN INTERIOR BALLISTICS 33 ‘Table 7 Standard gun system performance: comparison of several theoretical models je eerie epee >= >>> GEERT OPPPPS a Parameter Calspan Heppner’ ‘Maximum breech pressure, kpsi 51.85 49.64 50.10 ‘Maximum base pressure 47.51 41.92 44.00 ‘Time to maximum pressure, ms 1.58 9.00 9.2 Breech muzzle pressure, kpsi 9.95 9.0 10.0 Muzzle velocity, ft/s 2997 2989 3000 Time to projectile exit, ms 18.3 19.1 19.4 Projectile spin, rpm 15,750 15,750 15,700 Time to propellant burnout, ms 9.50 = 9.40 Total Chamber Volume (including propellant and gas volumes) Vp(t) = Veo + AgS(t) — Cl U/p,) +Z[9— U0) 1) 60) ‘Average Gas Pressure (with heat loss) _deZ= (y-1) 0/8) (M,V3/2) Gt 4 VO y where M,=W,+ (CZ/3) Pressure Conversion (base to average) op lpg tet Byer? Py Pa=[t A] (2) Initial values are as follows: X(0)=Xp (asmall number of order 10 ~? in.) (53) v,(0)=0 (54) 5, (0) =0 (55) Vz (0) = Veg — (c/s) [approximate due to conditionon X(0)] (56) One proceeds to calculate gun performance by simultaneously integrating Eqs. (47-49) subject to the specfied initial values and utilizing Eqs. (50-52). V. Example Calculation: Standard Gun-Propellant System Any analysis using the model can proceed only after a suitable standard gun- Propellant system has been defined adequately. The standard gun-propellant system is defined as that set of system parameters which, when used in the model, yields a desired theoretical performance consistent with that per- formance observed in the actual gun system.34 H. KRIER AND M. J. ADAMS. 60. 50. Calspan [3] 40. 30. Breech Pressure x 10°? [psi] 20. 10 ° ° Fi) 20 Tine x 10? [sec] Fig. 13 A comparison of the average pressure-time predictions for three different ballistic codes (175-mm M113 cannon). The standard system parameters consist essentially of average tabulated values of gun-propellant properties (e.g., bore length, projectile weight, etc.). Definition of all empirical system parameters (e.g., heat loss, engraving, and burning rate constants) is accomplished by simultaneously adjusting these empirical parameters until a desired theoretical ballistic performance has been achieved. Obviously, since there exists an infinite combination of adjustments which would give suitable ballistic behavior within the model, such ad- justments should proceed with the help of experimental data (e.g., closed bomb burning rate parameters) whenever possible. For this study, primary emphasis was placed upon simulating the 175-mm MI13 gun system employing an M437 projectile. The corresponding propellant for this system was an M86-M6 multiperforated propellant under a zone 3 loading condition. This particular gun-propellant system was chosen because of the availability of both theoretical and experimental performance data. 32223 These data provided a means of determining the predictive ac- curacy of the model. Table 6 lists the standard gun-propellant parameters employed for this study. The results of a simulation of the standard system are represented by the theoretical performance curves shown in Fig. 12. In examining the curves in Fig. 12, one should be aware of the fact that time, as used in Fig. 12, isa relative quantity because of the assumption that the total propellant bed has been ignited at t=0. Proper consideration of the ignition process would be required if one were to consider time as an absolute quantity in the model.INTRODUCTION TO GUN INTERIOR BALLISTICS 35 Acomparison of pressure-time results obtained from the UIBC (University of Illinois Ballistic Code) simulation of the standard system and those results of similar standard gun system simulations under other models}? is shown in Fig. 13. The theoretical curve as predicted by the model has been shifted so as to be on a time scale consistent with that of the other theoretical performance curves. ‘As shown in Table 7, peak breech pressure, as predicted by the model, is consistent with those values obtained from the other models. However, the model predicts a different pressure rise behavior, as can be observed in Fig. 13. It is felt that this behavior in the model may be due in part to the exclusion of boundary-layer effects within the gun bore during ballistic simulations. In an actual gun system, there arises a growing boundary layer behind the projectile as the projectile moves down the gun bore. This boundary layer acts to remove available energy and momentum from the generated propellant gases. The boundary layer has an accelerating effect upon the projectile through con- stricting the gas flow within the gun bore. In ignoring boundary-layer effects in the model, one may “‘overadjust”” the propellant burning rate parameters when attempting to attain a desired standard system performance. (Recall that the burning rate constants in the model are semiempirically derived quan- tities.) Such an overadjustment would account somewhat for the observed redistribution of the available propellant energy over the ballistic cycle. It thus means that any simulation of absolute gun performance requires adequate consideration of any boundary-layer effects upon projectile motion. However, as the UIBC model is to be used essentially to predict relative differences between gun system simulations, it will be assumed that the exclusion of boundary-layer effects will not influence radically the results of any com- parison of relative gun performance data. Acknowledgment This work was supported by the U.S. Army Picatinny Arsenal, Dover, N.J. References 'Krier, H., Van Tassell, W.F., Rajan, S., and VerShaw, J.T., ‘Model of Flame Spreading and Combustion Through Packed Beds of Propellant Grains,” Univ. of Mlinois, TR AAE 74-1, 1974. ?Kafadar, A.D., Olson, R.L., and Paul, B.E., ‘*Propellant Ignition Studies,” OEA, Des Plaines, Ill., Air Force Armament Lab., Rept. AFATL-TR-72-83, May 1972. *Fisher, E.B. and Trippe, A.P., “Mathematical Model of Center Core Ignition in the 175mm Gun,” Calspan, Rept. VQ-5163-D-2, March 1974, ‘Celmins, A., “A Computer Program for a Cartridge Expelling Gun,” Ballistic Research Labs., Rept. 2285, April 1973. SCorner, J., Theory of Interior Ballistics of Guns, Wiley, New York, 1950. ‘Engineering Design Handbook—Gun Series—Interior Ballistics of Guns,” Army Materiel Command, Pamphlet AMCP 706-150, Feb. 1964, ’*Engineering Design Handbook—Gun Series—Research and Development of Material,” Army Materiel Command, Pamphlet AMCP 706-252, Feb. 1964. **Weapons Systems Fundamentals ~ Analysis of Weapons,” U.S. Navy Weapons Systems, NAVWEPS Operating Rept. 3000, Vol. 2, Sept. 1963. *Baer, P.G. and Grollman, B.P., “Theoretical Studies of the Use of Multipropellants in High Velocity Gun Systems,” Proceedings of the 2nd Solid36 H. KRIER AND M. J. ADAMS Propulsion Conference ICRPG/AIAA, Anaheim, Calif., AIAA, New York, June 1967, pp. 22-30. ‘Fitzsimmons, F.J., “Concept Scope of Work for PEMA Project 57734186, Acceptance of Propellant Produced Via the Continuous Process (Project AUTOCAP),” Picatinny Arsenal, Rept. ASRSD-QA-A-P-55-73, Dec. 1972. ''Krier, H. and Shimpi, S.A., “Predicting Uniform Gun Interior Ballistics: Part I. An Analysis of Closed Bomb Testing,” Univ. of Illinois, TR AE 74-5, 1974. "Bird, R.B., Stewart, W.E., and Lightfoot, E.N., Transport Phenomena, Wiley, New York, 1960. Fisher, E.B. and Trippe, A.P., ‘Mathematical Model of Center Core Ignition in the 175mm Gun,” Calspan, Rept. VQ-5163-D-2, Picatinny Arsenal Contract DAAA. 21-72-C-0577, March 1974, “Flemming, D.P., ‘The Computer Simulation of Solid Propellant Guns and Rockets,” Centre de Recherches pour La Defense, Quebec, Canada, DREVIN- 1889/70, Aug. 1970, 'SJoglekar, A.M., Phadke, M.S., and Wu, S.M., “Iterative Modeling of Interior Ballistics of Small Arms,” Journal of Spacecraft and Rockets, Vol. 10, July 1973, pp 450-456. \Karamcheti, K., Principles of Ideal-Fluid Aerodynamics, Wiley, New York, 1966. "Osborn, J.R., “Evaluation of Solid Propellant Ballistic Properties,” Combustion and Flame, Vol. 20, 1973, pp. 193-197. '*Knapton, J.D. and Stobie, I.C., “Transient Burning Rates of Propellants,” Ballistic Research Labs., Interim Memo. Rept. 179, Aberdeen Proving Ground, Md., Jan. 1974, Krier, H., ‘Solid Propellant Burning Rate During a Pressure Transient,” Combustion Science and Technology, Vol. 5, 1972, pp. 69-73. Heiney, O.K., ‘“Analytical and Experimental Interior Ballistics of Closed Breech Guns,” Air Force Armament Lab. (ATWG), Eglin Air Force Base, Fla., Rept. AFATL-TR-69-42, May 1969. Hunt, F.R.W., Internal Ballistics, His Majesty’s Stationery Office, London, 1951. *Heppner, J.D., ““Final Report on USATECOM Project No. 9-7-0023-27, Special Study of Setback and Spin for Artillery, Mortar, Recoilless Rifle, and Tank Am- munition,” U.S, Army Test and Evaluation Command, Rept. DPS-2611, Jan. 1968. 2 Fitzsimmons, F., private communications, Picatinny Arsenal, Dover, N.J., 1974,Practical Interior Ballistic Analysis of Guns Paul G. Baer* Ballistic Research Laboratory, Aberdeen Proving Ground, Md. Interior ballistic models are available for the design of new guns, improving the performance of existing guns, and solving problems in existing guns. The type of models available can be classified into empirical and semiempirical classes. The semiempirical models are subdivided further into the closed-form solution, the numerical solution to the governing ordinary differential equations, and the numerical solution to the general partial differential equations. The performance of these models for gun design, improvement, and other problems will depend upon how well the model is matched to the problem. Data used by these models are a critical limitation on the predictive accuracy of the model and will strongly influence the resullts obtained by the user of the model. I. Introduction HOSE in research and development of solid propellant guns find them- selves matching a large array of possible solutions, as embodied in a large variety of gun interior ballistic models, to a large array of different gun problems. Deciding on a model that will solve the problem with the least effort challenges those charged with this responsibility. The problem toward which interior ballistic analysis is applied can consist of 1) the design of ignition systems that will reduce or eliminate pressure waves in guns; 2) the replacement of gun firings in propellant acceptance tests with interior ballistic models; 3) the design of new guns; and 4) the improvement of existing guns. It is the purpose of this paper to discuss the types of gun models now available and to determine how well these models do in the solution of these problems. Additionally, we shall discuss the kind and accuracy of the input data required by these models and how the accuracy of the data influences each model’s prediction. Il. Types of Gun Models Available Background We can distinguish two periods in the development of interior ballistic models: before and after the digital computer, All of the models before the Invited paper received Oct. 2, 1978. This paper is declared a work of the U.S. Government and therefore is in the public domain. “Research Physicist, Propulsion Division, 3738 P.G. BAER computer consisted of either closed-form solutions to the governing dif- ferential equations or tabular numerical solutions, since the users of these models desired answers without too many computations. A number of references! discuss these precomputer models. In all of these models, the terms in the differential equations were reduced to dimensionless notation, The resulting simplified differential equations were solved either analytically, producing a number of closed-form equations, or numerically, producing a large number of tables. The work done by Bennett‘ in 1921 illustrates this latter approach. In a few cases, mechanical analog computers were used. The development of the digital computer greatly facilitated the work of the interior ballistic analyst, allowing him to consider approaches previously denied because of the mathematical complexity. The initial work using the digital computer for solution was the repeat by Hitchcock® of Bennett's original work, with the tables being extended to more cases. At the same time, Taylor? used the digital computer to solve the differential equations that then were presented in graphical form using dimensionless coordinates. These tables and graphs required that the user reduce his particular interior ballistic problem to the dimensionless coordinates of the tables or graphs, obtain the dimensionless solution by interpolating in the tables or by reading off the graph, and then invert the dimensionless solution to engineering units. If the number of cases to be solved was large, this could be a long process. This table/graphical method of solution was changed by Baer and Frankle? with the introduction of the direct numerical solution of the ordinary dif- ferential equations on a case-by-case basis, thus putting the digital computer under the direct control of the user. With this method, the user could punch the data of the interior ballistic problem onto cards in the engineering units of the problem, submit the cards to the digital computer, and get back the complete interior ballistic trajectory of gun in engineering units. At the same time, some users"! were using the electronic analog computer in a similar manner. Although the analog computer was and is capable of providing gun interior ballistic trajectories much more rapidly than the digital computer, the results obtained were not more accurate than three significant figures and in some cases worse, the accuracy of the results depending upon the stability of the operational amplifiers, integrators, and multipliers of the analog computer circuits. They did have great utility in parametric gun design problems. The analog computer was, however, limited to the solution of the ordinary dif- ferential equations of interior ballistics. Attempts to apply them to the solution of the partial differential equations of interior ballistics were un- successful because of the great demand for numerical precision in the in- tegration of the partial differential equations. Solution of the one-dimensional (1-D) partial differential equations of interior ballistics began in the late 1960's when the speed of the digital com- puter had increased to a point that such solutions could be obtained in one hour or less per case. The first 1-D interior ballistic code was developed by Baer? and applied to experimental studies of high-velocity guns. In the early 1970's, a number of authors'™"® developed 1-D models that simulated ignition and flamespreading in the packed propellant bed and the subsequent two- phase flow of gas and propellant grains down the gun tube.PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 39 Classification of Available Models Gun interior ballistic models can be divided into two main classes: empirical and semiempirical. The empirical models are derived by fitting an arbitrary mathematical equation to gun experimental data. The resulting equations are generally in the form Vin =f; (GP,GP2,GP3°**) a) Pm =F) (GP),GP;,GP;***) Q) where V,, is the muzzle velocity and P,, the peak breech pressure. GP,, GP,--+ are gun parameters such as barrel length, barrel diameter, propellant chamber volume, charge weight, propellant web, etc., or combinations of these parameters such as charge-to-mass ratio, loading density, expansion ratio, etc. One example of an empirical model is the differential coefficients first used by Sugot!’ in 1928. This method is discussed in the engineering design handbook on gun interior ballistics '* in detail, and coefficient tables are given for a number of guns. Another example is the Frankford Arsenal small- arms graphical model, which was obtained by reducing experimental data from the firings of 11 different small-arms weapon systems. '* Least-squares curves were drawn to give the best fit to the data. These curves related maximum pressure, muzzle velocity, propellant weight, projectile weight, expansion ratio, and chamber pressure at a given projectile travel. These curves were fitted by White! to an empirical equation of the form: Venue = VermF Fx 8) where Viow = muzzle velocity Vim = velocity at given charge-to-mass ratio, pressure, and travel %, = pressure factor F, = expansion ratio factor and a simple computer program was written to compute the muzzle velocity. Semiempirical models represent dynamics of the gun interior ballistic process by equations that model the physics of the process. Some of the equations in the semiempirical models represent the physics of the process; other equations (usually those representing propellant burning, projectile frictional resistance, and heat loss) are empirical approximations. _ The semiempirical models are not necessarily better than the empirical models. In some problems, the empirical models can do a better job of gun performance prediction than can the semiempirical models, provided that the empirical models are used within the range of experimental data from which they were derived and that data for the corresponding semiempirical models are either in error or lacking. On the other hand, the range of applicability of the semiempirical models is not limited to a particular size or class of guns; thus the semiempirical models can be considered general-purpose models. All of the semiempirical gun interior ballistic models contain equations Tepresenting the physics of the gun interior ballistic process. These equations40 P.G, BAER can be arranged in a hierarchy of complexity ranging from the simple closed- form solutions to the complex partial differential equations in three space dimensions and time. In the most complex formulation of these equations, the region between the breech face and the projectile base is divided up into a series of control volumes. The three space coordinates of six vertices defining a control volume are given in the cylindrical coordinate systems, that is, with axial, radial, and angular coordinates. Within each control volume is contained hot propellant gas and sections of the burning propellant grains. Equations of the conservation of mass, momentum, and energy describes the physical processes taking place within these control volumes. In general, for such a control volume element, the law of conservation of an intensive property such as mass, momentum, or energy may be written in terms of the various possible contributions: rate of accumulation net rate of transfer of net rate of transfer of of property in element | = | Property into element by | + | Property into clement by bulk flow of medium molecular transfer net rate of production of [ net rate of production of ] property at surface of element } * | property within the element These equations are called the field equations for unsteady-state multiphase flow and have been discussed in a book by Soo” and also in reports by Panton”! and Bowen.” If assumptions of angular and radial uniformity in the intensive properties are made, these field equations reduce to partial dif- ferential equations in which the independent variables are axial position and time (1-D equations). These equations, when coupled with equations of state, projectile motion, propellant burning rate, bore friction, and heat loss, describe in considerable detail the physical processes taking place in a gun. Table 1 Classification of lumped parameter rior ballistic models Type of gun Modet Max. no. of Closed Traveling Recoil reference propellants Igniter breech —_Recoilless High-low charge __ model a 10 mv mi Y Y Y ¥ 6 1 N iv N N N N 10% > Y Y Y Y N Y 288 10 Y Y N N N N 298 1 Y Y N N N N 30 4 Y Y N N N N 3r 1 N Y N N N N 32° 1 Y Y N N N N 33 2 Y Y Y Y Y N 34 2 Y Y N N N Y 38 2 Y Y¥ N N N N ® Deterred propellants. Set upon analog computer only. © Thermodynamics properties of gases computed within code,PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS. 41 ‘The derivation of the 1-D equations and their use in interior ballistic models is described by Gough” in this book. The 1-D equations, in turn, can be reduced to ordinary differential equations, in which the dependent variable is time, by making the assumption that the propellant gas and unburned propellant form a mixture where density varies with time only. Mathematically, this means that the momentum conservation equation is reduced to a closed-form equation. This reduction was done by Love and Pidduck and Kent.5 A further assumption is made that all of the propellant burns at a space- mean pressure, which is proportional to the density of the propellant gas in the region between the breech face and the base of the projectile. This results in the following equations, which are presented in word form for generality: Energy Conservation [ energy from ] - [propelian gas 4 [Projectile kinetic burnt propellant internal energy. energy propellant gas +| andunburnt | + [heatloss] + [ eo canes ] propellant oe Projectile Motion rojectile base] [proiectie resistance] {bore area) |” pressure pressure [ projectile ] = acceleration [projectile mass] [projectile velocity] = time integral of [projectile acceleration] [projectile travel] =time integral of [projectile velocity] State of Propellant Gas [ propellant space mean ] [propellant gas internal energy] gas pressure {propellant gas volume} Burning Rate of Propellant ; rate of regression {mass burning rate] = [propellant density] [ Propellant ] of propellant surface, surface area} | function of space mean pressure [Provetlant mass pared | time integral of sass burning] rate42 P.G. BAER Propellant Form Function propellant euitees| = function of [ propellant initial dimensions, area propellant volume during burning Heat Loss ‘bore surface heat loss temperature difference (heat loss rate] = [' fy | [eat loss] between propellant gas and bore surface [heat loss] = ‘ime integral of [heat loss rate] Pressure Drop [Projectile base ] function of [eee mean pressure, friction pressure, ] pressure charge-to-mass ratio rojectile base pressure, friction pressure, ] charge-to-mass ratio [ projectile i} b function of [P reech pressure For recoilless guns, we have to add to the preceding equations the mass conservation equations: ls rate of increase of } = [mass burning rate ] _ [, gas mass flow rate ] ropellant gas in gun of propellant through recoil nozzle eee , weight of solid [nnburar etl = [initia solid] _[ weight of propellant] _ |. propellant remaining in gun weight burned ejected through nozzle Recoil Force and Impulse thrust by propellant recoil force force from [recoil force] = | gases flowing through] — | from projectile] + | counter recoil nozzle motion cylinder [ aa =time integral of [recoil force] Most of the interior ballistic models that rely on the numerical integration of ordinary differential equations (ODE) will have almost all of the preceding equations, particularly those of energy conservation, projectile motion, state of propellant gas, propellant burning rate, propellant form function, and pressure drop. These equations may not necessarily appear separately but may be combined into one or two differential equations. At the present time, however, the trend is to keep the equations separate to facilitate understanding of the model and also to allow for modifications.PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 43 There is considerable diversity among the various interior ballistic models of the lumped parameter type. The user of any of these models (18 were discussed in Ref. 26) must relate the model to the interior ballistic problem. For instance, a model that considers only one propellant burning in a gun would be useless for howitzer problems that contain two or more propellants. Table 1 categorizes the lumped parameter models according to gun type, maximum number of propellants, presence of an igniter, and presence of a recoil mode. _ With the use of appropriate assumptions, the ODE interior ballistic equations can be integrated, resulting in the closed-form models. Reference 1 outlines the various assumptions used and the integration methods used to obtain seven closed-form models. The major assumption is in the use of a pressure exponent of one in the burning rate law; other assumptions involve the elimination or combining of terms in the energy equation and sim- plification of the form function. These models range in complexity from a few algebraic equations to a number of tables in which the user must interpolate. ‘The latter, although more accurate than the former, are little used today since digital computers that can integrate the basic ODE’s rapidly are available to all potential users. Three closed-form models that are in use today are LeDuc, Mayer-Hart, and RD38. The LeDuc* system was based on the observation that a hyperbolic curve whose general equation takes the form y=ax/(b+x) would represent the velocity-travel relation of a projectile in a gun, where y represents velocity and x represents travel. Alger>’ and Patterson®* refined the method and ap- plied it to the U.S. Navy charge assessment calculations. Until recently, the method, because of its simplicity, was used widely both in the U.S. Army and Navy. Grollman® has re-examined the LeDuc method and has shown it to bea powerful tool in generating the pressure-time curves needed in recoil impulse calculations, provided that the maximum pressure and muzzle velocity are known from either other theoretical methods or experimental results. The Mayer-Hart method was developed assuming 1) zero shot-start pressure, 2) that propellant gas covolume equals reciprocal of propellant density, 3) pressure exponent of one in burning rate equation, 4) constant burning surface of propellant, and 5) no heat loss or frictional energy loss. With these assumptions, a series of closed-form equations was obtained from the integration of the basic ODE’s. This model was re-examined by Kravitz, ! and closed-form equations were developed in which the covolume and shot- start assumptions were removed. Subsequently Kravitz‘? adapted the method to guns in which two propellants of the same composition but different webs were burning. Strittmater‘? made the standard Mayer-Hart assumptions but made the assumption that bore friction is proportional to chamber pressure. This was equivalent to giving the projectile an effective mass somewhat higher than the actual mass. With the data from 165 firing records, a factor was obtained which brought the theoretical prediction of muzzle velocity closer to experimental muzzle velocity. An unpublished numerical experiment by Baer using data for 96 guns ranging in size from 0.30-cal rifle to 280-mm cannon indicated that, given the experimental maximum breech pressure, the Mayer- oes was able to predict the muzzle velocity with an average error of 77%.44 P.G. BAER The RD38 method“ is a closed-form solution used by the interior bal- listicians in the United Kingdom. The method originally was developed by Crow, 5 who assumed 1) that all of the energy losses in the energy equation can be neglected; 2) that propellant gas covolume equals reciprocal of propellant density; and 3) pressure exponent of one in the burning rate equation. Since the method was developed in 1921, the interior ballisticians in the United Kingdom have made various empirical corrections to the closed-form equations based on the large amount of experimental results. With proper attention to the various empirical corrections, the method, in experienced hands, can give good predictions for conventional guns. Optimum Gun Models Corner? stated that there were two classes of optimum gun problems: 1) the choice of the best charge for a given gun, and 2) the choice of a “best” gun for a given projectile, with maximum velocity as a criterion. With an assigned peak pressure, it was decided that maximum muzzle velocity occurred when all of the propellant was burned at or near the muzzle ejection of the projectile. Recently the optimum gun problem has been re-examined by Reis and Spahr“ and by Horst and Burnett.*” The procedure used by these authors is to modify the ODE equations of interior ballistics such that, when the pressure profile of the gun is specified as a function of projectile travel, the mass burning rate of the propellant will be such that this pressure profile will be maintained behind the projectile base as it moves down the bore. Thus, the maximum velocity would be obtained under these circumstances for a given gun and projectile. Both programs computed the surface area of the propellant necessary to maintain this pressure. Reference 46 complicated their solution by computing heat loss at various stations down the bore; thus, at each time point, the solutions of double integrals were required, increasing computer time and storage over the simplified heat loss procedure used in Ref. 47. As would be expected, the surface area of the propellant increases from small values near start of motion to large values at propellant burnout. A closed-form solution to this problem was developed by Riefler,** who developed the equations on the basis of a new efficiency term on which to compare the performance of guns. The gun that would be 100% efficient on this new basis would have a flat pressure-space curve to burnout of the propellant, followed by adiabatic expansion of the propellant gas to the gun sm GUN 3g mROVECTLE MUZHE VELOCITY 2hm/s 30 roretian 40 (Groine oewsrty ov evee ; : i MAXUM PRESSURE “s5tiaho Fig. 1 Design solutions for # 60-mm gun using the Mayer- Hart conventional gun theory and the constant-pressure optimum gun theory. MAYER HART THEORY ry 3 700) 120 us ‘CALIBERS "OF TRAVELPRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 45 muzzle. Considerations of a constant-pressure gun led Baer“ to rederive these equations on the basis that a portion of the propellant is used to pressurize the chamber to maximum pressure prior to shot-start. Also, because of the pressure gradient, the pressure at the breech face would be higher than at projectile base, even though both pressures were constant during the burning of the propellant, a consideration neglected by Riefler. Solutions to these equations allow one to design an optimum gun with “deal” propellant progressivity. Design solution for an ideal 60-mm gun capable of launching a 1160-g projectile at 2.0 km/s is illustrated in Fig. lasa plot of charge-to-mass ratio (C/M) vs calibers of projectile travel. It is assumed that M30 propellant is used at a loading density of 0.9 g/cm?, and the maximum pressure is 80 kpsi (551 MPa). One notes that, as the C/M is in- creased, the calibers of travel necessary to obtain the 2.0-km/s muzzle velocity decrease to a minimum of 72 calibers at a C/M=2.5. Subsequent increase in C/M causes an increase in calibers of travel. III. Role of Models in Solving Interior Ballistics Problems The types of interior ballistic problems which the preceding models can be used to solve are 1) reduction of pressure waves in guns, 2) modeling of propellant acceptance tests, 3) design of new guns, and 4) improvement of existing guns. Reduction of Pressure Waves in Guns Pressure waves in guns have been connected to the role of the igniter. Understanding this phenomenon requires the modeling of the radial flow of gases from the central tube igniter, subsequent ignition of propellant charge by the igniter gas, and then subsequent flow of hot propellant gases through the packed charge up to the base of the projectile. The problem is complicated, in some cases, by the insertion of fins on long booms into the propellant charges. In other cases, the projectile is surrounded by the propellant on the rear and sides, requiring the igniter charge to push it forward out of the chamber into —--» = aa] Process CONTROLLER x Fig. 2 Flow diagram for continuous N 1 propellant manufacturing plant. I INTERIOR BAUISTIC F CloseD chawacR FRING TESTS46 P.G. BAER the bore. Such problems require the use of two- or three-dimensional models, which have not yet been developed. The modeling of the flow from an igniter at the base of the charge or of that of a central tube igniter assuming no radial flow can be accomplished with the one-dimensional codes. This is discussed by East.°° The role of the one-dimensional code in the reduction of gun pressure waves is discussed by Horst.*! Modeling of Propellant Acceptance Tests Interior ballistics modeling of propellant acceptance tests proved to be a severe test of the predictive capabilities of the models. In 1972, the Product Assurance Directorate decided that the current batch process for making cannon propellant would be replaced with a continuous automated propellant production line. One of the critical problems for the operation of this propellant production line was the development of a rapid and accurate method of charge weight estimation to allow the on-line loading of gun charge packages without the delays encountered in using proving-ground gun firings to establish propellant charge. An on-line gun interior ballistic computer model was proposed to replace the gun in the charge establishment role. The placement of such a model in a continuous propellant manufacturing plant is illustrated in Fig. 2, which is a flow diagram showing the relationship between the propellant line, the propellant line process controller, and the interior ballistic model. Raw materials go in the beginning of the line, and the fabricated propellant grains come out at the end of the line and are packed into charges and sent to storage. Samples of the propellant grains are taken off the line and subjected to chemical, physical, and closed chamber firing tests. The results of these tests go to the interior ballistic model, which, in turn, provides the process controller with information needed to determine the charge weight in the propellant package. Functional details of the gun model are illustrated in Fig. 3. Propellant chemical composition data are processed by gun ther- modynamic models such as the BLAKE code? to produce propellant ther- modynamic data for use by the gun simulation model. Propellant grain dimensions and density for the gun model are provided by physical [CLOSED CHAMEER FRNGS [AND OATA REDUCTION. CCHEMKAL ANALYSIS [ome ow fa (rrsecay, pacrenes THERMODYNAMIC MODEL y we. Tate cove) (we) ran BUENSONS) No DAA) {oman eases) once Tor Tp £ _ GON PRORCTIE | AT Ga SowXaTion mone [marae vem (eat ance Cx ECE OF necer on REJECT? Fig.3 Flow diagram of interior ballistic model for continuous propellant testing.PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 47 Table 2 Ballistic acceptance plan for 175-mm gun propellant charge, M86A2" ee 70°F (21°C) Velocity limits: 2940-3060 ft/s (896-933 m/s) Maximum standard deviation: 10 ft/s (3 m/s), 0.33% 39: 30 ft/s (9 m/s), 0.99% 140°F (60°C) Velocity limits (info.)®: 3015-3130 ft/s (919-954 m/s) ‘Maximum standard deviation (info.): 16 ft/s (5 m/s) Maximum zone 3 pressure: 48,400 psi (334 MPa) Maximum excess charge pressure: 58,000 psi (400 MPa) Sample: 5 rounds at 70°F (21°C), 5 rounds at 140°F (60°C). Data taken for information only. Table3 Ballistic acceptance plan for M4A2 155-mm howitzer propellant charge* Velocity limits: 1825-1875 ft/s (556-572 m/s) Maximum standard deviation: 9.2 ft/s (2.8 m/s), 0.50% 30: 27.6 ft/s (8.4 m/s), 1.50% ), 5 rounds at 140 10,500 psi (279 'C): 44,100 rounds at 70°F (21 ual pressure at 70°F (21* le individual pressure at 140' sample: Permissible MPa); permis: psi (304 MPa). examination, and propellant burning rates are provided by closed chamber tests. Gun performance data from the model are compared with propellant acceptance criteria, and a decision is made to accept or reject the lot. If ac- cepted, the propellant charge weight then is computed. Four different interior ballistic models were tested during this investigation. They were the CALSPAN code’s (a complex 1-D code), the multipurpose- multipropellant (MPR) code?” and the Hitchcock code* (both lumped parameter codes), and the empirical differential coefficient model.'® The procedure used for the validation of these models was to manufacture a number of lots of propellants with properties that spanned the extreme range of impetus, relative quickness, chemical composition, density, and geometry variability to be expected during production. These lots, 8 lots of M6 propellant for the 175-mm M113 gun and 15 lots of M1 propellant for the 155- mm M185 howitzer, then were fired at the proving ground where interior ballistic data were taken. Using the chemical composition, physical properties, and the closed chamber combustion data for each of these lots of propellants, as measured at the propellant manufacturing arsenal, it was the objective of the project to see how well the interior ballistic models would predict the measured gun performance. The maximum standard deviation between Prediction and experimental muzzle velocity had to be equal to or less than the maximum standard deviation in muzzle velocity used in the ballistic ac- ceptance of these propellants (Tables 2 and 3).*2 It will be noted, however, that the standard deviations given in the tables are round-to-round standard deviations produced by measurement errors, by variations in projectile and gun dimensions, and by variations in the weights of the propellant charges.48 P.G. BAER 350 X EXPERIMENTAL DATA, + riast Taal > 9est FIT Fig. 4 Muzzle velocity vs maximum breech pressure, 175-mm M113 gun, zone 3, propellant tots A-H, MPR code. 2 100! 7230300350400 MAXIMUM BREECH PRESSURE’ (iRo} The simulation results for the CALSPAN code are given in Ref. 54 for the 175-mm firings only. The muzzle velocity errors ranged from 0 to 4.06%, this range of errors being attributed to use of closed-bomb burning rates in the gun calculations and uncertainties in computing the propellant thermodynamic data needed for the code. The results of using the MPR code to predict the performance of the 175- mm gun with the eight lots of propellant are illustrated in Fig. 4.55 In this figure, muzzle velocity for each of the eight lots is plotted vs maximum breech pressure. Experimental results and model predictions are illustrated in the figure. The best results of several simulation trials were a percentage error of 2.71% in muzzle velocity and 18.71% in maximum pressure. [The percentage error was defined to be the 30 (standard deviation value) of the absolute difference between predicted and measured values divided by the standard muzzle velocity or standard maximum pressure.) This was clearly too large a value for charge assessment calculations. Simulation results for both the Hitchcock code and the differential coef- ficient method are given in Refs. 55 and 56, The Hitchcock code, in order to give acceptable results, had to be modified. This modification consisted of the addition of terms to the quickness function so as to account for charge weight correction and a web independent relative quickness. The result of this charge gave the results illustrated in Fig. 5% for the eight lots of the 175-mm propellant. The percentage error for this was 0.49% in muzzle velocity and 7.21% in maximum pressure. On the basis of these promising results, simulation work with the Hitchcock code was extended to 209 175-mm production lots, 32 155-mm production lots, and 15 155-mm experimental propellant lots, where ballistic properties spanned the range of properties expected during propellant production.‘ The percentage error in muzzle velocity for these runs ranged from 1.56% to 2.10%. These larger errors were attributed, in part, to errors in the determination of relative quickness and relative force from closed chamber measurements. As indicated in Ref. 58, thePRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 49 X EXPERIMENTAL DATA UNMODIFIED CODE = MODIFIED CODE Fig.5 Muzzle velocity vs maximum breech pressure, 175-mm M113 gun, zone 3, propellant lots AGH, Hiteheoek code. MUZZLE VELOCITY (ws) 00 00 300 2s0 300380 (MAXIMUM. BREECH PRESSURE” (Po) measurement errors in closed chamber firings could cause a 1.5% to 2.7% error in muzzle velocity. In the differential coefficient method, the muzzle velocity and maximum pressure variations are predicted by two equations: AR RQ\? ARF)? . ACW AacW)? a ipa wo tae) +a ae eae) row +4r( ow) ® AP Ri ARQ)? ARF RF\2 ACW Wy? Boat ae +e (FE +4, op +e ae) Cw +2() © where AV = predicted muzzle velocity minus standard muzzle velocity v = standard muzzle velocity ARQ = experimental relative quickness minus standard relative quickness RQ = standard relative quickness ARF = experimental relative force minus standard relative force RF = standard relative force AP = predicted maximum breech pressure minus standard maximum breech pressure P = standard maximum breech pressure Aacw = predicted charge weight minus standard charge weight cw = standard charge weight 4),4, = constant terms b,,b;,etc. = differential coefficients These equations are different from the usual differential coefficient equation shown in Ref. 18 in that a constant and second-order terms have been added to account for large differences in RQ, RF, and CW. In Ref. 56, the50 P.G. BAER EXPER MENTAL OATA THEORY AND #8 08 EXPER MENTAL WORK LEAST SQUARES FT Fig.6 Muzzle velocity vs maximum breech pressure, 175-mm M113 gun, zone 3, propellant lots AGH, differential coefficient method. MUZZLE VELOCITY Imad 700 255 a0) __350 200 "MAX MUM BREECH PRESSURE Pal linear form of Eqs. (4) and (5) without the constant term gave an error of 0.34% in muzzle velocity and 6.15% in maximum pressure. The simulation results for the eight lots of the 175-mm propellant are illustrated in Fig. 6.55 In Ref. 56, this work was extended to the 209 175-mm propellant production lots, the 32 155-mm propellant production lots, and 15 155-mm_ experimental propellant lots, In this approach, the coefficients for the differential coef- ficients were obtained from the experimental data for a set of propellant lots using least-squares regression techniques. These lots were picked to span a wide range in ballistic properties. Once the coefficients were determined, the equations were used to predict muzzle velocities and maximum pressures for either the lots used in the coefficient determination or in other lots not used in the coefficient determination. The effect of leaving off constant or higher- order terms was checked. The errors in muzzle velocity ranged from 0.89% to 3.34%. Like the Hitchcock code results, the errors in the closed chamber determination of RQ and RF could influence the results markedly. It was concluded that the differential coefficient method gave somewhat better predictions than the Hitchcock method and thus should be used as the preferred interior ballistic model on the continuous propellant line. The disadvantage of the model was that it required experimental firings of a set of propellant lots whose ballistic properties would span the complete range of properties expected during production. The Hitchcock code could be used for start-up of the line with a new propellant and/or a new gun, since the method would require the firing of only one lot in order to calibrate the model. The more complex models such as the MPR code or the CALSPAN code gave poor performance in these tests because of their requirement for good estimates of bore friction profiles, dynamic burning rates, heat loss, etc., lack of which gave the results noted. Closed chamber measurement errors seriously affect the ability of the models to predict the performance of the guns. Improvements in both precision and accuracy of closed-bomb burning rates and impetus parametersPRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 51 are essential. Further discussion on this problem is given in the section on the effect of data accuracy on model prediction. Design of New Guns Another use for interior ballistic models is in the design of new guns. When designing a new gun, the model is not used alone but is incorporated as part of an overall weapons system model. Such a model is shown in Fig. 7.7 Another weapon system model incorporating interior ballistic models is discussed in Ref. 58. The user first starts with a target definition and a vulnerability model, in which the target is defined and its vulnerable areas outlined. He then progresses to a terminal ballistic impact model, which defines the kind of projectile necessary to defeat the target and its impact velocity at the target. This information then is used by an exterior ballistic model to describe the ballistic trajectory from the gun to the target, thus defining the velocity at which the gun has to launch the projectile. Shape and dimensions also are defined by a special projectile design model coupled to the exterior ballistic model, so that the projectile may hit the target accurately with minimum loss in its kinetic energy due to atmospheric drag. With the projectile shape, dimensions, and muzzle velocity defined by input from the other models, the user proceeds to a gun design model that will have a gun interior ballistic code and possibly a sabot design code incorporated in it. With the gun design model, the user proceeds to design a gun and compute an interior ballistic trajectory necessary to launch the projectile at the desired muzzle velocity. The gun design is subject to numerous external constraints such as minimum overall weight, maximum noise, maximum projectile ac- celeration, minimum recoil, etc. Numerous iterations may have to be made through the entire weapon system model loop before an acceptable design may be found. Another gun design model is the TOGA code, *’ which incorporates a ter- minal impact model, an exterior ballistic model, a sabot design model, and two interior ballistic codes. This code is used for the design of antitank guns. As mentioned previously, the terminal impact model and the exterior ballistic model define the projectile flight weight and launch velocity. Since the high- velocity projectiles are fin-stabilized, a sabot is needed to hold the projectile in TARGET DEFINITION MODEL VuINERABTLITY MODEL TERMINAL BALLISTICS Fig.7 NAPA ig. 7 Weapons system model. + EXTERIOR SALLETICS OF PROJECTILE’ MODEL I [oN INTeROR pau ioTCs SiN BE Rove uSoee52 P.G. BAER the bore and provide the low sectional density necessary to launch the projectile at high velocities. The sabot design code provides the weight of sabot necessary to accomplish this. This weight, when added to the projectile flight weight, gives the in-bore weight of the projectile. The Mayer-Hart code is used to design the gun. The code has been modified such that, if the maximum breech pressure or maximum projectile base pressure, muzzle velocity, projectile weight, projectile diameter, and propellant thermodynamic characteristics are defined, then the user can use the code to make a plot of propellant charge weight vs calibers of projectile travel for several loading densities. Such a plot for a 60-mm gun launching a projectile at 2.0 km/s is illustrated in Fig. 1, where it is compared with results from the optimum gun code. The user then can trade off gun length with charge weight, and, if the loading density is defined, the user then will determine the chamber volume of the gun. It is to be noted from these curves that there is a projectile travel below which it is impossible, given other constraints, to attain the desired muzzle velocity. The other interior ballistic model used in the TOGA code is the LeDuc model, which is used, once the other parameters have been determined by the Mayer-Hart code, to plot the, pressure-time and pressure-space curves. This can be used both to design the thickness of the gun wall and to aid in the design of the recoil mechanism. Once the interior geometry of the gun has been determined, the interior ballistician has to determine an acceptable propellant charge. Although the Mayer-Hart code also will predict a propellant web, this web is based on two assumptions: 1) the pressure exponent in the burning rate equation is unity; and 2) the propellant has a constant burning surface. For low-muzzle-velocity weapons this may be sufficient, but for high muzzle velocity (velocities greater than I km/s) this is not sufficient, and propellants with progressive geometry dom SMOOTH BORE GUN 12g. PROJECTILE sesmm WEB MS PROPELLANT GhamBeR. VOLUME = 747 ce Provectie Travel - 20 00] 2500 Sao e ag, messune ‘Sirens 3 —,/ 200-3 = f £ Fig.8 Experimental and predicted °°) wre vaociny/ Emule velocity s propellant weight fora B | tekinenia 7 EU tavmmsnesiicoe gent is Sy hoo 2 ; fume voc | a 3m] atti # = 2 120 Max pRessunt Preoicreo. 208 300 20 ae PROPELLANT WEIGHT (a)PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS. 53 must be chosen. The procedure then is to use one of the lumped parameter interior ballistic models to predict the charge weight and web necessary to attain the desired muzzle velocity, holding the maximum pressure and projectile acceleration values within the desired limits. This procedure is discussed in the section on improvement of existing guns. Predicting how a new design will perform after it is built is a difficult problem for the interior ballistician. Based on his previous experience, he must make estimates of the values for propellant burning rate, which is higher than the closed chamber burning rate, the frictional resistance profile, and the heat loss. He then can make predictions of muzzle velocities and maximum pressure for a number of charge weights using these estimates and then vary the estimates and redo the predictions. This will produce a series of curves within which data from the experimental firings may lie. The general and safe procedure is to start firings in the new gun with low charge weights and in- crease the charges gradually until the desired muzzle velocity is reached. The interior ballistician can, while this is going on, improve his predictions based on the initial experimental data and thus guide subsequent experimental work. ‘As an example of the problems that the interior ballistician can encounter in attempting to predict the performance of a new and novel gun, consider the following case. We were asked to set up a firing program in a smooth-bore 40- mm ballistic tube.°? The gun had no chamber, being of uniform internal diameter from the breech face to the muzzle. Thus the propellant chambers would be determined by the distance between the projectile base and the breech face. There also would be no shot-start to retard projectile motion while the propellant ignited. It was decided, in order to provide rapid and uniform ignition, to pack the MS propellant in a cylindrical tube around an igniter tube consisting of black powder and a low-energy detonating cord. This charge assembly extended the length of the chamber. Predictions of the probable firing results were made using the MPR code?’ assuming no shot- start pressure and a frictionless bore. Burning rates for the M5 propellant were derived from closed chamber firings. The agreement between these pre- experimental predictions and the experimental results is illustrated in Fig. 8, in which muzzle velocity and maximum breech pressure are plotted vs propellant weight. It will be noted that, at the low charge weight of 200 g, the differences (MAK PRESSURE = 227 MPa 900 MSOAT PROPELLANT wren ig. 9 Optimum loading density for 155-mm MI85 howitzer using 0-, 1-, 7-, and 19-perforated M30A1 Propellant. MUZZLE VELOCITY Inn) 8 6 * 10 LOADING DENSITY (g/cc),54 P.G. BAER between predicted and experimental results is fair (6 MPa for breech pressure and 70 m/s for muzzle velocity). At the high charge weight of 500 g, the agreement is poor (170 MPa for breech pressure and 230 m/s for muzzle velocity). These results also emphasize the wisdom of starting experimental firings at low propellant weight levels, where the possibility of prediction errors will have litte consequence, and then gradually increasing the propellant weight, carefully monitoring the firing results. Improvement of Existing Gun Performance Another problem to which interior ballistic models are applied is the im- provement of the performance of existing guns. This problem may take two forms. In the one, the interior ballistician is asked to see if it is possible to get a higher muzzle velocity using the existing projectile. In the other, the projectile weight may be either increased or decreased, and the interior ballistician is asked to determine the maximum muzzle velocity that can be obtained while staying within the pressure constraints of the gun. One has several options. A propellant with a higher impetus will increase the muzzle velocity but also may increase the gun erosion rate, since, in general, high-impetus propellants have high flame temperatures. Work is underway to develop high-impetus propellants without the higher flame temperature. Another option is to vary the geometry of the propellant, keeping the charge weight and propellant composition fixed. This option is explored in detail in Ref. 60, where the gun was the high-velocity 105-mm M68 tank gun. The procedures used to increase the muzzle velocity was to increase the piezometric efficiency by 1) varying propellant web, 2) using mixes of varying web of the same propellant, 3) using mixes of propellant which vary in composition and web, and 4) varying propellant grain shape. It was concluded, as a result of this study, that only varying propellant grain shape offered the greatest possibility of increasing the muzzle velocity of the 105-mm M68 gun, It was found that the 19-perforated propellant, the propellant with the most progressivity, increased the muzzle velocity by 120 ft/s (36.6 m/s). avn 8.8 EXPONENT cchaae wr. Force BR COEF, Fig. 10 175-mm M113 gun, zone 3, sensitivity study: percentage change in ax muzzle velocity vs percentage change in input parameter, MPR gun model. CovOWUNE wee PROM. WEIGHT SH. RATIOPRACTICAL INTERIOR BALLISTIG ANALYSIS OF GUNS 55 Another system is to vary both the charge weight and web, keeping the maximum pressure constant until the optimum muzzle velocity is reached. This is illustrated in Fig. 9 for the 155-mm howitzer.“! The procedure used to develop these curves was to use the MPRGUN code to simulate the 155-mm M185 howitzer using the zone 8, XM 203EI charge. A series of charge weights then was selected for a cord propellant. At a charge weight, a series of parametric runs was made varying the web until the desired maximum pressure was attained. The muzzle velocity at this peak pressure then was plotted vs loading density. The procedure then was repeated for the next charge weight, etc. This was continued until the maximum muzzle velocity was passed. From this curve, one could determine the optimum loading density and the optimum web. This procedure was repeated for the single-perforated grain case, the 7- perforated grain case, and the 19-perforated grain case. IV. Matching Interior Ballistic Code Predictions to Experimental Data All of the semiempirical interior ballistic models require that, prior to use as valid predictors, their predictions be matched to experimental gun firing data. This is because of the use of empirical equations in the model; constants for these equations generally have to be determined from gun firings. Depending on the complexity of the model, this matching procedure can be simple or complex. The unknown matching parameters generally are associated with 1) the burning rate equation, 2) the bore resistance function, and 3) the heat loss equation. In the complex 1-D interior ballistic models, there will be additional matching parameters associated with propellant ignition and propellant grain drag equation, relating the velocity of the propellant grains to that of the propellant gas. The closed-form Mayer-Hart model* uses two matching parameters: the specific heat ratio to account for heat loss, and a bore resistance parameter that is used to modify the projectile mass. These two parameters are all that is needed, as long as the model is not used to predict maximum breech pressure. If maximum breech pressure is to be predicted, then an adjusted burning rate coefficient is required to account for unity pressure exponent. Additionally, the code can be used only to predict maximum breech pressure for constant burning surface propellants, e.g., single-perforated grains. Degressive or progressive grain geometries will give erroneous breech pressure results. The Hitchcock®°* model contains seven adjustable factors. These are a quickness empirical factor, a quickness adjustment factor to improve charge weight correction, a velocity empirical factor, a velocity adjustable factor to improve charge weight corrections, a relative force exponent, a relative quickness exponent, and a specific heat ratio to account for heat loss. Of the seven adjustable parameters, only the quickness empirical factor and the velocity empirical factor are varied to match to a particular experimental firing. The other factors have constant values, which have been determined from matchings to numerous experimental firings. During the development of @ propellant acceptance test model, the values of these parameters, but not the Specific heat ratio, were varied in an attempt to improve the predictive capability of the model. In Ref. 60, the MPSG lumped parameter code was56 P.G. BAER matched to the experimental maximum breech pressure and muzzle velocity by systematically varying a propellant erosion constant, the shot-start pressure, and a constant bore resistance pressure until the muzzle velocity and maximum breech pressure within the lo experimental variations were obtained. In all subsequent parametric variations, these three values were held constant. In the MPRGUN code,’ a function minimization procedure is used to automate the matching process. This function takes the form OVi-STDi STDi+ STDAi function= 3 STDC, © fo] where OVi = value of ith output variable from simulation STDi = value of standard (or matching) ith output variable STDAi = value of additive variable STDCi = value of coefficient (or weighting) variable 1 = number of output variables being matched When this procedure is used, all of the data for a master case are read in, plus data for the function minimization subroutine. A master case trajectory is computed and printed out, followed by a parametric line output that displays input and output parameter values, together with the value of the function as defined previously. Also displayed on the line is the smallest value of the function. When the program obtains the smallest value of the function within the error limits put on the input data, the function minimization routine will stop searching, after it determines that the function is the smallest possible value. The program then prints out the values of the input data which will best match the output data and then runs a trajectory with those before stopping. The function minimization method used is that of Powell and is described in Ref. 62. Table 4 shows a typical parametric line listing using the function minimization routine. The case illustrated here is the 155-mm howitzer, zone 8, with the standard muzzle velocity and maximum breech pressure being matched by varying the maximum engraving pressure, PR2, and the constant resistance pressure after engraving, PR3. The minimum function underlined in the table gives a predicted pressure of 47,493 psi (experimental maximum pressure 47,500 psi, 0.015% error) and predicted muzzle velocity of 2710.60 ft/s (experimental muzzle velocity of 2710 ft/s, 0.022% error). One or more input parameters can be matched against one or more output parameters, but there should be a one-to-one correspondence between the input and output parameters; otherwise, a nonunique solution will result. For instance, one could vary burning rate coefficient, constant resistance pressure, and heat loss coefficient to match against maximum pressure and muzzle velocity, and a combination of these three input parameters would be found. However, if one started with a new set of initial values for these three parameters, the function FThis is a projectile velocity correction to the propellant burning rate in the form of 1r=BP" +KoV, where r is the rate of regression of propellant surface, B the burning rate coef ficient, P the space mean pressure, V the projectile velocity, » the burn rate exponent, and Ku the propellant erosion constant.57 PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS nrL202 bud 6°L586 zad 3SV2 1X4N 0109 31374WO9 NOTivaa1I L0~-31Za9* =X4 20-31289° G0-3998T* TO+3000T* TO9*ILT €4°2966 TOTETLZ 9O°LESLY O9°2O61 €6°TbR6 20-31289" GO-3SSST* TO+3000T* G99°LLT SBTTLO6 HT"BOLZ OG *OG4LH 11°2902 S6*E0B6 20-31289" G0-39GGT" 10+3000T* ye6"ZLT 0079906 24r2tz2 = 2E*OS4LH 095286T OL HBG LO-31209* 9O-3dHT9* TO+3000T* DZH*TLT ZEra966 _EL*BOLZ _—9L*9ESLY TT*ZL0Z bo" L 996 ZO-31209* L0-31289* TO+a000T” L2S*IZ1 20°L966 O9*OILe eS*E6¥ly GE*LZ0e vO°L¥O6 §0-3269S* 0-39002* TO+40001" 2u9°9LT 6E'9OO6 00 HO9Z aL *LATLY LZ*EIZZ TETOHSS §0-32695* G0-32695* [O+30001* »Z2B°ILT 02°8566 96*HTLZ 09*2LGL4 ¥S*H96T 9992266 §0-30%¥G* 0-308" 10+3000T* OOBtILT ZAr9SE6 HOTSTLZ —-2570KSL¥ BHTOSET 9z"S¥66 40-3h52T* YO-abHZT* TO+3000T* GBO°ZZT 8075566 © H9TLILZ —H9009LH EL*STOI $276866 40-3S9ET* EO-30¥ZT* TO+9000T* Z2TH*6L1 SESH © 9OTGELZ «= HAZOZLY LE*GEYT © —S2*6966 40-3S9ET* ¥0-36618* TO+3000T* TeE*IZT €9°E966 20%HOL2 2T*2TOLy LO*SHz2 S2"6866 40-3S9ET* 40-3S9ET* TO+3000T* 22G*@2T 6074566 SHtOIL2 9OTLSGL4 LO*OLBT S2*6966 40-32802" 4O-3Z224E" TO+9000T* GZE*ZLT G2rAy66 —ZETOZLZ ~—«ES*OTAL L6°0OBT 92*LL001 0-32802* 40-32802* TO#30001" O9Z"2L1 ET"IS66 LetOTLZ LatOEILy LO°0DRET SZ°EEDOT 0-3€G62* ¥O-3HEED* TO+000T* 9L9*TLT 15"2566 62*60LZ += HA*LLBLY GHTHETZ G2*EEOOT 40-3662" EO-aHLL1* 10430001" BSE*BLT LETTLO6 460222 —OT*2689 94°6A9T B2*E0L6 40-3662" €0-39602* 10+3000T* E9G*ZLT LZ"9Z66 -GOTLZLZ «= THTTTT® Ger6ERT Tz*E9E0T %O-3E562" 40-3€S62* TO+I00OT* HLGtaLT 9T*HHOS Ad*HZLZ —HOTOZSLY FH*H9LT S2*EEOOT %0-3L266% ¥0-3L256" 10430001" GO¥*ELI 2BTHO6 —GOTHELZ —OT*LOELY 9¥°6EST G2*EE0OT EO-3ETET* EO-JETET® TO+3000T* BBS*ELT IL*o£66 LI*LELZ BETEZLY Oet¥9HT S2°EEOOT E0-3EHLT* EO-3EHLT® 10430001" 292°6LT 09°LEOH = BZ*OHLZ-—«-HL*GOTL4 9¥*OBET SZrEEOOT €0-3€00€* €0-3£00E* 10+3000T* 99%*OLT £E76HO6 IS*RELZ 26°S4RD GHTLYET 65°4996 £0-39826* £0-39826* 10+30001* 16z2°SH1 206206 © ONTEELZ «—LaTTTT9® 29°S92T Tht 88%6 T0-369€T* T0-369ET* 00+35266" 000°0 E9*L2TOT 19°2692 gordyoTy 00°0Ga —26°6S92 TO-3¥62T* TO-3462T* 004+30L66* 000° yO°RGTOT 2676892 BOTRHTTY O00°SGL2 96°62EL T0-31Z@T* T0-31ZGT* 00+36966" 000°0 GETLGTOT 21°1692 428601 OOT0EL 966ZEL T0-30€9T* 10-30€8T* 00+36966* 000°0 OB°IGTOT 99°TO92 SE*ZBOTe OOTSTL gorezEL 10-36€8T* T0-36£81* 00438966* 000°0 GETISTOT 6192692 46"S90TY 00°00L 96r6ZEL T0-38202* 10-3802" 00+39966* 000°0 O9*HLTOT G0°9892 —20°S990% 007002 00 OSTL 10-369T2" TO-38912* 00+36966* 000° OE'TATOT 2a°eB92 Ey0Z60% 007002 onrszoe 10-30622" T0-306z2° *0000°0 00°0 o0*o 00° o0°o oro THNN ONAT wz NI-082 “Saad “ZW TWA XWWd ead ead GQ*= VHdIv HATA HWW 39aVHD 2a€02WK 82 WHSST 98sey> TALOT WX WIM *g BUOz “49zZ}1MOY UNUE-SET 405 S)INsOd W eZIUUIUE UOHIUNI p1858 P.G. BAER TableS Input data for interior ballistic models Type of Data type model required* Reliability Projectile weight A Good, based on direct measurements Propellant weight A e Bore area or diameter a’ - Chamber volume A’ Maximum projectile travel AY " Propellant grain dimensions AY Somewhat uncertain since random variations due to shrinkage can occur Propellant thermodynamic AY ‘Somewhat uncertain since theory data: force, flame assumes constant gas composition temperature, specific heat during interior ballistic cycle ratio, etc. Propellant density A Somewhat uncertain since theory assumes that propellant grains are incompressible Propellant burning rate A’ Fair for homogeneous propellants; under strand burner or very uncertain for deterred closed chamber conditions propellants Mathematical form of the LP, PDE Valid theory lacking, and data for burning law under gun existing theories very uncertain conditions and data required for the laws in present use Ignition characteristics of LP, PDE a propellants under gun Engraving and frictional LP, PDE a forces between projectile and bore Air shock pressure ahead of LP, PDE Good theory, data somewhat projectile uncertain Heat loss to bore LP, PDE Valid theory lacking, and data for existing theories very uncertain Friction between propellant LP, PDE q ‘gas and bore surface Recoil motion and energy LP, PDE Valid theory, data somewhat uncertain because of lack of detailed measurements *Aeall models, A’ PDE = partial differe! all models with some empirical model exceptions, LP =lumped parameter models, and ‘equation models, minimization routine would find a new set of input values which matched the experimental maximum pressure and muzzle velocity. At the present time, the user needs considerable experience in the use of such methods. Users of other interior ballistic models, when faced with matching the model’s predictions to experimental data, must locate the adjustable parameters in the model and parametrically carry out a matching process similar to that just outlined. V. Input Data for Interior Ballistic Models Interior ballistic models have differing requirements for data. Some of the simple empirical models will require little data, whereas the more complexPRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 59 lumped parameter and partial differential equation models will require a considerable amount of data. These data vary greatly in reliability. On the one hand, the data that can be measured directly, such as projectile and propellant weights and the gun dimensions, are considered to be highly reliable, whereas the data required for the unproven theories used in some of the models are to be considered highly uncertain. Input data required by the interior ballistic models are listed in Table 5. These data are categorized according to the type of model they are used in and their reliability. It can be seen from this table that the first five items have good reliability; the next four items are somewhat uncertain because of the reasons noted; and the remaining items, with the exception of recoil and air shock pressure, are highly uncertain because of the lack of valid theories and good data to support existing theories. Of these uncertain data and theories, those that are of the greatest importance to the interior ballistics code user are those of gun burning rates, engraving and bore friction, and heat loss from propellant gas to bore surface. Many different gun burning rate equations have been evolved for use in the interior ballistic models. Some of these are = bp" M r= DP*+KV, ® r= bP*(1+K,V,) @) r= S(PP,--*) (10) Equations (7-9) normally are used in the lumped parameter codes, with the coefficients b and K, and the exponent n acting as adjustable constants; that is, their values are determined by matching the code predictions to ex- perimental gun firing data. In such matching, initial estimates of the values of 5 and n are provided from closed chamber or strand burning tests of the propellant. Experience in such studies has shown that burning rate exponents ranging from 0.85 to 0.95 appear to be suitable estimates for the gun burning rate equation. Experience also has shown that the propellant erosion term K, has a significant effect only on predicted muzzle velocity. Variations in K, will affect the muzzle velocity significantly only for cases where propellant burnout occurs close to muzzle ejection of the projectile. Equation (10) states that burning rate is some function of the pressure derivative P. A number of theories for burning rate with the pressure derivative included have been developed. These theories and the results of using these theories to compute burning rates under gun conditions have been examined by Nelson. Some of these theories are in a form suitable for use in lumped Parameter models; others require the numerical solution to partial differential equations and thus would be costly to use in lumped parameter models. At the present time, no conclusions have been reached on the usefulness of these theories in the practical interior ballistic models, although research on these theories is being pursued actively. Engraving and bore friction equations in the interior ballistic models take many forms. In the closed-form models and the earliest lumped parameter60 P.G. BAER models, the equations used were some additive correction to the projectile weight; thus the energy lost in engraving and bore friction was proportional to the kinetic energy of the projectile. In such models, the resistance to initial motion of the projectile produced by the rotating band engraving in the rifling was provided by a shot-start pressure, which disappeared once the projectile began to move. In the later lumped parameter models, this method was replaced by a table of projectile resistance pressures or forces as a function of projectile travel. In such a table, the engraving and frictional forces en- countered by the projectile could be modeled correctly. In practice, such resistance profile data rarely are available, and so the resistance profile tables serve as another series of adjustable constants used to match code predictions to experimental data. One type of table is the three-point table, which has values of resistance pressures at start of motion, position of maximum engraving (typically the length of the rotating band), and position of post- engraving falloff (usually at a position 1’ time the rotating band length). The value of resistance for this last position stays constant for the remaining travel of the projectile. In matching studies, the resistance pressure at maximum engraving will modify the maximum breech pressure, whereas the resistance at the post-engraving falloff position will modify the muzzle velocity. Initial estimates of these two values are 10% and 1% of the measured maximum pressure. The disadvantage of this method is that for subsequent predictions the resistance profile has to be held constant such that, when one uses values of parameters which are increasingly different from those used in the matching, the predictive error will be greater. A comprehensive engraving and bore friction model was developed by Stahls® which used the dynamic motion of the projectile and the characteristics of the rotating band and gun bore to predict the engraving and frictional resistance. Incorporation of this model into a lumped parameter code by the author led to inconclusive results, since the data required by the model were not available, and there were too many data required to be used as adjustable constants. Pilcher® has developed a model for the post-engraving bore friction based on Franklin Institute fric- tional data.” In this model, the post-engraving bore friction is based on the characteristics of the rotating band, the gun bore, and the projectile velocity. With the use of this model, the engraving region still has to be modeled by a fixed resitance profile. Initial use of this model in the lumped parameter codes has been unsatisfactory; more work has to be done in developing a com- prehensive engraving and bore friction model. In the closed-form and earlier lumped parameter models, heat loss was accounted for by either increasing the specific heat ratio, thus reducing the chemical energy available to accelerate the projectile, or introducing a correction term into the projectile weight analogous to the correction term used for work in overcoming engraving and bore friction. The later lumped parameter codes used an empirical heat loss term developed by Hunt? in the energy equations: = 0:38d'7 [xt (Vela) \ToV5 ay ** TE (0.602175 (C0875) Jy2,PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 61 where a bore diameter, in. x projectile travel, in. Vp = projectile velocity, in./s c = total propellant weight, Ib v, = estimated muzzle velocity, in./s V. = chamber volume, in.? A = borearea, in.? T) = adiabatic flame temperature of gas, K This equation requires an estimate of the muzzle velocity, and so the user has to provide one or make an estimate from an empirical equation in the code, compute the trajectory, and then use the predicted muzzle velocity to provide a better estimate. Generally one or two iterations provide acceptable convergence. The disadvantage of such a model is the requirement for an estimate of muzzle velocity which lengthens the computer running time because of one or more iterations to provide such an estimate. Another heat loss model used in the lumped parameter and PDE models is the Nordheim model. This model is the following: £,= |) A,h(T-T,)4t (12) where the heat-transfer coefficient A is UES 5k )Mly/M (y= 1) 1Re¥, (13) and the friction factor is A= [14.244 log j9(2.54D,)] ~2 and where A, = total bore surface behind moving projectile, in.? space-mean gas temperature, K bore surface temperature, K specific heat ratio molecular weight of propellant gas, Ib/mole gas constant, in.-Ib/mole-K gas density, Ib/in.? projectile velocity, in./s heat-transfer factor AN? PER NY. The parameter k, is used to match the heat loss predictions to experimental results, Since heat loss has a lesser effect on predicted maximum pressure and muzzle velocity than do bore friction or burning rates, the tendency of users of the code is to leave the heat loss correction alone unless adjustments of the burning rate or bore friction adjustable parameters do not give the results desired. The user of a code with the Nordheim heat loss model in it will find that increasing the value of the adjustable parameter (normally k, =1) will62 P.G. BAER decrease the predicted muzzle velocity, and decreasing the parameter will increase it. Other, more complex heat loss models exist based on various assumptions as. to the type of thermal boundary layer in the gun. One is the Thornhill model,? based on the development of a steady-state boundary layer behind the projectile. This requires values for the thermal conductivity of the propellant gases and the viscosity of the propellant gases. Since these values are uncertain under the conditions existing in the gun and since the model assumes steady- state boundary-layer development for the highly unsteady-state boundary layer in a gun, the model is little used. More complex models have been developed for heat loss from a gun. These models are discussed by Anderson® and Shelton.” The effect of data accuracy on model predictions can be shown best by a sensitivity plot, as illustrated in Fig. 10.5° This figure shows the percentage change in muzzle velocity for a 175-mm gun zone 3 charge as a function of percentage change in various input parameters. These parameters are burning rate exponent, charge weight, force (impetus), burning rate coefficient, covolume, web, projectile weight, and specific heat ratio. It will be noted that the first five parameters give a positive increase in muzzle velocity when they increase in value; the remaining three parameters give a decrease in muzzle velocity when they increase in value. The burning rate exponent gives the greatest increase in muzzle velocity, and the covolume term gives the least. With the aid of such plots, the effect of an error in the measurement of a particular parameter on a predicted muzzle velocity can be determined readily. For instance, a 1% error in the measurement of propellant force would produce a 0.55% prediction error in muzzle velocity for the 175-mm gun. This figure is a graphical representation of the differential coefficient equations; thus the differential coefficients can be used for this purpose. Coefficients for these equations are given in Ref. 18 and also can be used for data accuracy estimates. VI. Discussion and Conclusions Selecting an interior ballistic model out of the collection of models described here depends primarily on the problem. The simplest use is the prediction of gun muzzle velocity, with the maximum breech pressure being determined by the maximum tube stress of an existing gun or by a maximum projectile ac- celeration constraint. For this the Mayer-Hart code is the best. The Mayer- Hart model, used in conjunction with the LeDuc code, is very useful in the gun design problem where maximum gun pressures and muzzle velocities are specified by external considerations. The potential user will find that the various lumped parameter models, in particular the MPRGUN code and the Hitchcock code, are most useful for propellant charge design and op- timization, particularly with existing guns for which there is some body of experimental firings. For problems involving igniter design, reduction of pressure waves, and the prediction of heat stress distribution in gun tubes, the user would be advised to use the complex 1-D interior ballistic models, such as the NOVA code.PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS. 63 In the use of any of these models, the user must not be misled into thinking that a more complex model will give better results than a simpler model. The results from the gun acceptance tests modeling indicate that the converse was true; namely, the simpler model will give the more accurate results. This is because of the detailed use of input parameters involving propellant com- bustion, engraving, bore friction, and heat loss by the more complex models. Since the values of these parameters are the most uncertain and have a significant effect on predicted gun performance values, the result is poorer predictions for the more complex models. The relatively poor performance of the MPRGUN code and the CALSPAN code in the gun acceptance test modeling project pointed out clearly the gaps in good input data for these and other models. Good theories and data are lacking for propellant combustion under the dynamic conditions in guns for resistive forces between the projectile and bore during engraving, for passage down the bore, and for heat loss from propellant gases to the bore surface. References 'Laidler, K.J., “A Comparison of Interior Ballistic Systems,” Catholic Univ. of America, Contract NORD 9692, Rept. NAVORD 750, Sept. 1947. 2Corner, J. Theory of the Interior Ballistics of Guns, Wiley, New York, 1950. 3Hunt, F.R.W., Internal Ballistics, Philosophical Library, New York, 1951. ‘Bennett, A.A., ‘Tables for Interior Ballistics,"” Office of Chief of Ordnance, War Dept., Doc. 2039, 1921. SDirector of Cambridge Mathematical Lab., “The Application of the Bush Dif- ferential Analyser to the Problem of Internal Ballistics,” Advisory Council on Scientific Research and Technical Development, Ministry of Supply, Rept. AC 1375, i941. ‘Hitchcock, H.P., “Tables for Interior Ballistics," Ballistic Research Lab., Rept. 993, 1956. TTaylor, W.C., Kelso, J.W., Mayers, M.R., and Blakely, W.W., ‘A Graphical System for Interior Ballistic Computations,” Ballistic Research Lab., Rept. 825, 1952. ®Taylor, W.C. and Yagi, F., “A Method for Computing Interior Ballistic Trajectories in Guns for Charges of Arbitrarily Varying Burning Surface,” Ballistic Research Lab., Rept. 1125, 1961. °Baer, P.G. and Frankle, J.M., “The Simulation of Interior Ballistic Performance of Guns by Digital Computer Programs,” Ballistic Research Lab., Rept. 1183, 1962. Edwards, A.G., “Interior Ballistic Analysis of Various Gun and Launcher Systems,” Picatinny Arsenal, TR 3193, 1965. ‘Fundamentals of Ballistics,” U.S. Army Ordnance Center and School, Rept. ST9-153, 1964. "Baer, P.G., ‘‘A Mass Point Computer Program for the Gas Dynamic Problems of High Velocity Gun Interior Ballistics," Proceedings of 2nd ICRPG/AIAA Solid Propellant Conference, Anaheim, Calif., 1967. Gough, P.S., “Fundamental Investigation of the Interior Ballistics of Guns,” Space Research Corp., Final Rept., Contract 0174-73-C-0501, SRC-R-75, 1974. “East, J.L. and McClure, D.R., “Projectile Motion Predicted by Solid/Gas Flow of Interior Ballistic Model,” Proceedings of the 10th JANNAF Combustion Meeting, Chemical Propulsion Information Agency, Publ, 243, 1973, 'SFisher, E.B. and Trippe, A.P., “Development of a Basis for Acceptance of Continuously Produced Propellant,”” Calspan Corp., Rept. V2-5163-D-1, 1973.64 P.G. BAER '6Krier, H., Van Tassell, W.F., Rajan, S., and Verhsaw, J.T., “Model for Flame Spreading and Combustion Through Packed Beds of Propellant Grains,” Univ. of Ilinois at Urbana-Champaign, TR AAE74-1, 1974. '"Sugot, G., Ballistique Interieure Theorique et Tables Numeriques, Gouthier- Villars, Paris, 1928. '8“Interior Ballistics of Guns,” Army Materiel Command, Engineering Design Handbook, Ballistic Ser., Pamphlet AMCP 706-150, 1965. Whyte, B., “Interior Ballistics,” General Electric Armament Dept., Burlington, Utah, Letter Rept., Feb. 20, 1969. 200, S.L., Fluid Dynamics of Multiphase Systems, Blaisdell Publishing Co., Waltham, Mass., 1967. 21Panton, R., “Flow Properties for the Continuum Viewpoint of a Nonequilibrium Gas-Particle Mixture,” Journal of Fluid Mechanics, Vol. 31, Pt. 2, Jan. 1968, pp. 273- 303. Bowen, R.M., “Continuum Theory of Mixtures,’” Ballistic Research Lab., Contract Rept. 45, 1971. 2Gough, P., ‘Modeling of Two-Phase Flow in Guns,” published elsewhere in this volume. Love, A.E.H. and Pidduck, F.B., “‘Lagrange’s Ballistic Problem,” Philosophical Transactions, Vol. A222, March 1922, p. 167. 2Kent, R.H., ‘The Motion of Powder Gas, I and II,’ Ballistic Research Lab., Repts. 36 and 37, 1936. 2Baer, P.G. and Stals, J., “A Comparison of Gun Interior Ballistic Computer Programs,” Ballistic Research Lab., Rept. in review, 1978. 27Baer, P.G., “MPRGUN, A Multipurpose-Multipropellant Weapon Interior Ballistic Simulator,” Ballistic Research Lab., Rept. in review, 1978. 2Trafton, T.R., “An Improved Interior Ballistic Model for Small Arms Using Deterred Propellant,” Ballistic Research Lab., Rept. 1624, 1972. Smith, C.M., “An Interior Ballistic Model for Rolled Ball Propellant, Part II, A Computer Implementation (IMBRB),” Frankford Arsenal, MR M71-7-1, 1971. Heppner, L.., ““Final Report of Special Study of Electronic Computer Program for Interior Ballistics,”” U.S. Army, Test and Evaluation Command, DPS-1711, 1965. 3'Heiney, O.K., “Analytical and Experimental Interior Ballistics of Closed Breech Guns,” Air Force Armament Lab., AFATL TR-69-42, 1969. Peterson, E.G., “Interior Ballistics of Guns,’ Hercules Inc., Bacchus Works, Magna, Utah, 1967. *3Nusbaum, M.S., Barbarek, L.A.C., and Zimmerman, F.J., “Hypervelocity Weapon Feasibility Study,” Air Force Systems Command, Eglin Air Force Base, Fla., Rept. ATL-TDR-64-35, 1964. Schlenker, G., “Update of Cannon Interior Ballistics,” U.S. Army Weapons Command, Rock Island Arsenal, TN RDF 73-1, 1973. ¥Voltis, P.M., “Digital Computer Simulation of the Interior Ballistic Process in Guns,” Watervliet Arsenal, Rept. TR-6615, 1966. %Challeat, J., “Theorie de Affuts a Deformation,” Revue d’Artillerie, Vol. 65, 1904-1905, pp. 184-186. +7 Alger, P.R., “The LeDuc Velocity Formula,” U.S. Naval Institute Proceedings, Vol. 37, No. 138, June 1911, pp. 535-540, 3Patterson, G.W., “The LeDuc Ballistic Formulae,” U.S, Naval Institute Proceedings, Vol. 38, No. 143, Sept. 1912, pp. 885-892. 3Grollman, R., ‘LeDuc Revisited,” Ballistic Research Lab., Rept. in preparation, 1978. 40Mayer, J.E. and Hart, B.L., ‘Simplified Equations of Interior Ballistics,”” Journal of the Franklin Institute, Vol. 240, No. 5, Nov. 1945, pp. 401-411.PRACTICAL INTERIOR BALLISTIC ANALYSIS OF GUNS 65 ‘“iKravitz, S., “Interior Ballistic Integrations for Powder of Constant Burning Surface,” Ballistic Research Lab., Memo. Rept. 552, 1951. Kravitz, S., “Interior Ballistic Calculations with Dual Granulation Propellant,”” Picatinny Arsenal, TR 3135, 1963. 4Stritmater, R.C., ‘A Single Chart System of Interior Ballistics,” Ballistic Research Lab., Rept. 1126, 1961. “éCoppock, S.W., ‘The Method of Internal Ballistic Calculation in Use in the Research Department,”” Army Research Dept./Ballistics, ARD/BAL/82/42, 1942. “Crow, A.D., “A Short Method of Calculation of Internal Ballistics and Its Ap- plication to Gun Design,” Woolich Arsenal, Research Dept., RD Rept. 38, 1921. “Reis, G.E. and Spahr, H.R., “The Theoretical Maximum Muzzle Velocity of Large Caliber Guns,”” Sandia Labs., Rept. SC-DR-72 0460, 1972. 47Horst, A.W. and Burnett, W.M., ‘An Approach to Design Optimization of the Interior Ballistic Cycle for Closed Breech Guns,” Naval Ordnance Station, Rept. ITHTR 359, 1972. “Riefler, D.W., “MVEL: An Improved Measure of Efficiency for Interior Ballistics,”” Journal of Ballistics, Vol. 1, March 1977, p. 201. “Baer, P.G., ‘A Constant Pressure Closed Form Solution to the Optimum Gun Problem," Ballistic Research Lab., Rept. in preparation, 1979. SEast, J.L., Jr., ‘Ignition and Flamespreading Phenomena in Granular Propellant Gun Charges,”” published elsewhere in this volume. $!May, 1.W. and Horst, A.W., “Charge Design Configurations and Their Effect on Pressure Waves in Guns,” published elsewhere in this volume. Fitzsimmons, F.J., “Concept of Scope of Work for PEMA Project 5774186 Acceptance of Propellant Via the Continuous Process (PROJECT AUTOCAP),” Product Assurance Directorate, Picatinny Arsenal, Rept. SMUPA-QA-A-P-55-73, 1972. Freedman, E., “A Brief Users Guide for the BLAKE Program,” Ballistic Research Lab., IMR 249, 1974, “Searo, P., ‘‘A Mathematical Prediction of Ballistic Performance for M6MP Propellant in the 175-mm Gun System,”” Product Assurance Directorate, Picatinny Arsenal, Rept. ARSD-QA-A-P-58-74, 1974. ‘Baer, P.G., May, I.W., and Frankle, J.M., ‘‘A Comparison of Several Predictive Approaches in Charge Establishment for Large Caliber Artillery Systems,” Proceedings of 11th JANNAF Combustion Meeting, Vol. 1, Pasadena, Calif., Sept. 1974, pp. 55-66. SeBaer, P.G., Bushell, M.S., May, I.W., and Frankle, J.M., ‘The Development of an Interior Ballistic Model for Automated Continuous Production Control,” Ballistic Research Lab. (to be published). *'Shear, R. and Grollman, B., ‘Target Oriented Gun Analysis for Feasibility Studies,” Proceedings of the XVI Army Operational Research Symposium, Vol. Il, 1977, pp. 990-1003. “Nichols, L.F., ‘Integrated Projectile Systems Synthesis Model (IPSSM),”” Frankford Arsenal, Status Review of AMC Integrated Weapon System Synthesis Modeling Program, Jan, 23-24, 1974. Baer, P., ‘‘40 mm Experimental Firings,” 1977 (unpublished). ©Grollman, B.B. and Baer, P.G., “Theoretical Studies of the Use of Multipropellants in High Velocity Guns,” Ballistic Research Lab., Rept. 1411, 1968. ‘Combustion Working Group, ‘The Role of Ignition and Combustion in Gun Propulsion: A Survey of Developmental Efforts," 13th JANNAF Combustion Meeting, Vol. 1, Sept. 1976. pp. 315-339.66 P.G. BAER Powell, M.J.D., “‘An Efficient Method for Finding the Minimum of a Function of Several Variables Without Calculating Derivatives,” The Computer Journal, Vol. 7, July 1964, pp. 155-162. «Proving Ground Acceptance Test Procedure for Artillery Propellants, Charge, Propellant, 155-mm M4A2,” Aberdeen Proving Ground, Suppl. P-15SH-75, Cl, Jan. 13, 1975 and Aug. 5, 1974. “Nelson, C.W., “Response of Three Types of Transient Combustion Models to Gun Pressurization,’* Ballistic Research Lab., BRLM 2752, May 1977. “Stals, J., “Development of a Dynamic Bore Friction Model for Guns and Mortars and Its Conceptual Application to Early-Round and Gun Erosion Inaccuracies,” Australian Defense Standard Labs., Rept. 573, 1974. “Pilcher, J.O. and Wineholt, E.,“Analysis of the Friction Behavior at High Sliding Velocities and Pressures for Gilding Metal, Annealed Iron, Copper, and Projectile Steel,”? Ballistic Research Lab., Rept. 1955, Jan. 197. Montgomery, R.S., “Friction and Wear at High Sliding Speeds,”” Benet Lab., Rept. WVT-TR-75028, June 1975. Adams, L.H. (ed.), ““Hypervelocity Guns and Control of Gun Erosion,” National Defense Research Committee, Washington, D.C., Summary Tech. Rept. of Div. 1, 1946. ® Anderson, L.W., Bartlett, E.P., Dahm, T.J., and Kendall, R.M., “Numerical Solution of the Nonsteady Boundary Layer Equations with Application to Convective Heat Transfer in Guns,” Aerotherm Corp., Mountain View, Calif., Final Rept. 70-22, 1970. 79Shelton, S., Bergles, A., and Saha, P., ‘Study of Heat Transfer and Erosion in Gun Barrels,”” Air Force Armament Lab., AFATL-TR-73-69, March 1971.Interior Ballistic Modeling Applied to Small-Arms Systems Sidney Goldstein* U.S. Army Armament Research and Development Command, Dover, N.J. Empirical models have, in the past, been successfully programmed and used to predict the performance of different gun system designs. In recent years analytical models have been developed which can serve as a means for determining where a specification may be changed or whether enough energy is available to cycle the weapon properly so that it can function reliably. The incorporation into these models of such particulars as propellant deterrent effect on the burning rate, propellant grain size distribution, propellant temperature sensitivity, and system energy losses requires further detailed work. It is concluded that each of the models has a useful, although limited, function in the proper design of a small-arms weapon system. Nomenclature acceleration of projectile, in./s? Pidduck-Kent constant, dimensionless = area of base of projectile including appropriate portion of rotating band, in.? = covolume of ith propellant, in.?/Ib diameter of bore, in. specific heat at constant volume of ith propellant (cy, is a function of 7), in.-Ib/Ib-K = mean value of specific heat at constant volume of ith propellant (over temperature range T to T,,), in.- Ib/Ib-K &, = mean value of specific heat at constant pressure of ith propellant (over temperature range T to T,,), in.- Ib/lb-K charge weight, Ib initial weight of ith propellant, Ib initial weight of igniter, Ib mer os q G Invited Paper received July 24, 1978, This paper is declared a work of the U.S. Government and therefore is in the public domain. *Physicist, Fire Control and Small Caliber Weapons Systems Laboratory. 6768 u ay W wD ZS RD RRS Wl S.GOLDSTEIN specific heat of gas at constant pressure, cal/g-K specific heat of solid propellant per unit mass, cal/g- K particle diameter of narrow fraction of screened ball propellant, in. representative mean diameter, mean grain diameter, in. incremental time, s incremental temperature, K incremental distance traveled by projectile, in. mass fraction burning rate for ith propellant, s~! major diameter of ith propellant grain, in. activation energy for gas phase reaction, cal/mole energy lost due to heat loss, in.-Ib force of ith propellant, in.-Ib/Ib force of igniter propellant, in.-Ib/Ib fraction of charge weight or volume with radius R,, dimensionless constant for conversion of weight units to mass units, in./s? burning rate velocity coefficient, in./(s-in./s) burning rate displacement coefficient, in./(s-in.) number of propellants, dimensionless moles of gas per unit mass, moles/g pressure, psi time rate of change of pressure, psi/s space-mean pressure resulting from burning i propellants, psi pressure on base of projectile pressure of gas or air ahead of projectile, psi peak chamber pressure breech pressure, psi resistance pressure, psi propellant linear burning rate, in./s linear burning rate of ith propellant, in./s adjusted linear burning rate of ith propellant, in./s propellant burning rate at some reference tem- perature, in./s one-half web size, in. equivalent mean grain radius, in. average radius of propellant grain for a given sieve cut, in. universal gas constant, cal/mole-K surface area of partially burned ith propellant grain, in.? time, s mean temperature of propellant gases, K flame temperature of propellant, K