CPTR75 JANUARY 2002 harcpies aval om CPIA ry. Reproduction authorized except by specific pe SOLID PROPELLANT SUBSCALE BURNING RATE ANALYSIS METHODS FOR U.S. AND SELECTED NATO FACILITIES R. S. Fry 20020221 O71 CHEMICAL PROPULSION INFORMATION AGENCY + THE JOHNS HOPKINS UNIVERSITY + + WHITING SCHOOL OF ENGINEERING + COLUMBIA, MARYLAND 21044-3204 + DISTRIBUTION STATEMENT: Approved for Bub reese;Form Approved (OMB No. 0704-0168 TS BAS HE RG OTIS SGSIST SSE TDW SUTTON, HOUSE Te Tony TEESE SOR SIS in go SE Trorisrty be cou pnt nd coitg sven fe cleo orton Gea eanmar erase brn sotyls a cose of Ps coer cera, ee ee ee arse Wb essai: Se Src tx erie Open i Rebs, 8 eo Gs gay, Sale TDN, gon VA aE ‘Pitti onts Saisdgeety an andar Pedatc salen get roses Waxnngen, BCS. REPORT DOCUMENTATION PAGE Fe RGENGY USE ONLY fan Ba oo Sepont OnTe IREPORT TYPE DAD DATES COVERED January 2002 Technical Report, Oct 1997-Mar 2001 FE MROSOS 3 FROG HOES Solid Propellant Subscale Buming Rate Analysis Methods for U.S. and Selected NATO Fates a fame ue (C:SP0700-97-D-4004 RORORS Fry, Ronald S. FR IZATION NAMES AND ROOPESSTEST TE FERFORIING ORGIEATON Report NuMBER The Johns Hopkins University Chemical Propulsion Information Agency on 10620 Lile Patuxent Parkway, Suite 202 Columbia, MD 21044-3204 | 5 ron RORTTORING ROERCY RATE ARD ADORESSTEST 10-" SRORSCRNERONTORNS [ROENEY REPORT NUMBER DTic-al Naval Ai Warfare Center Weapons Division £8725 John J. Kingman Road Code 4770000 Suite 0944 China Lake, CA 93555-6100 FL Belvoir, VA 22060-6218 71 SUPPLERENTARYROTES Hardeopies available from CPIA only. BS TRISUTOR ATARI STATEENT TB STRISTTTON CODE ‘Approved For public release; distribution is united “ABSTRACT (Maximum 200 words) Current methods used within the NATO community for analyzing small motor buming rate test data are reviewed and recommendations ‘are made to support improved prediction of internal balistics ofa full-scale sold propellant motor. ‘The NATO Research and Technology Organization (RTO), Advanced Vehicle Technology (AVT), Working Group (WG) 016 (formerly AGARDIPEP Working Group #27) undertook to evaluate methods used within the NATO propulsion community to measure buming rate in solid propellant racket systems, with the purpose of identifying similarities and differences between the member nations. This WG ‘was formed in 1996, consisting of representatives from 6 of the 15 member nations of NATO, with inputs accepted from 4 other ‘member nations and a couple non-member nations. The NATO RTOJAVT WG 016 sought to contribute to improvements in the burning rate tools to address issues that have plagued the solid propulsion industry for over 40 years: (1) Better understanding of buming rate data from various faciliies to ease the comparison of propellants from various ‘manufacturers and to improve intemational exchanges and cooperation. (2) Better accuracy and reliablity of measurements allowing a decrease in the number of tests (and associated time and cost) and ‘an improved control of manufacturing and aging. ‘Simulated and real subscale rocket motor data were used to evaluate the two fundamentally diferent families of burning rate analysis ‘methods. While organizational preferences generally dictate method usage, surveys indicate a trend toward methods that more effectively account for non-idea! taloff, favoring improved accuracy. Consistency n these definitions would promote ease in correlating data internationally. Further development of the Hessler-Glick method shows promise. The NATO propulsion community is urged to review these findings as a means of advancing their own buming rate measurement and analysis methods. "13. SUBJECT TERMS 7 NORIBER OF PAGES Buming rate Firing tests (motor) Qualitative analysis Soli rocket propellants Combustion Instrumentation Quantitative analysis Test methods 266 Combustion chambers Measurement Rocket motor PIA collection TPES Composite propellants Parametric at Solid propellants 17, SECURITY TTY GEASSIFICATION TE SECURTY CASSIFIGATION | 28. TRATATION OF ABSTRACT OF REPORT (OF THiS PAGE OF assrRACT UNCLASSIFIED UNCLASSIFIED UNCLASSIFIED UA TEN TSIOOT ISOS Sd Fam (Ree 2 Preerbot by ate 8 26-8 Bose‘The Chemical Propulsion Information Agency (CPIA) is a DoD Information Analysis Center operated by ‘The Johns Hopkins University, Whiting Schoo! of Engineering, under Defense Supply Center Columbus (DSCC) contract SPO700-97-D-4004. The applicable DoD’ Instruction is 3200.12-R-2, "Centers for Analysis of Scientific and Technical Information." The CPIA's mailing address is The Johns Hopkins University, Chemical Propulsion Information Agency, Atin: Security Office, 10630 Lite Patuxent Parkway, Suite 202, Columbia, Maryland, 21044-3204. The CPIA also provides technical and administrative ‘Support to the Joint Army-Navy-NASA-Alr Force (JANNAF) Interagency Propulsion Committee. ‘The Government Administrative Manager for CPIA is the Defense Technical information Center (DTIC), Code DTIC-Al, 8725 John J. Kingman Road, Suite 0944, Ft. Belvoir, VA 22060-6218. The Government Technical Manager (Contracting Officer's Technical Representative) is Mr. Stuart Blashill, Naval Air Warfare Center Weapons Division, Code 477000D, China Lake, CA 93555-6100. ‘All data and information herein are believed to be reliable; however, no warrant, expressed or implied, is to be construed as to the accuracy or the completeness of the information presented. This document was Prepared under the sponsorship of the Defense Technical Information Center and is available only to qualified users. Neither the U.S. goverment nor any person acting on behalf of the U.S. Gavernment assumes any liability resulting from the use or publication of the information contained in this document or warrants that such use or publication will be free from privately owned rights. All rights reserved, This Publication or any part thereof, may not be reproduced in any form without written permission of the , Chemical Propulsion Information Agency.PREFACE This issue of the Chemical Propulsion Technology reviews (CPTR 78) continues CPIA's recurrent series of technical summaries and status reports on topics pertaining to missile, space, and gun propulsion technology. The general aim is to collect, analyze, and discuss technology advancements in a language understood by a broad range of propulsion technologists. ‘The results reported upon here are a part of the overall technical approach of (1) Surveying the NATO solid propulsion community for subscale and non-intrusive test methods, analysis, and scaling methods, and (2) Analyzing "simulated" and "real" motor pressure-time data using multiple thickness/time and mass conservation burning rate analysis methods taken from the survey. This CPTR reviews recommendations on current buming rate measurement methods used for analyzing ‘small motor test data to allow accurate prediction of internal ballistics of a full-scale solid propellant motor. Findings and recommendations on performance scaling, and test techniques and subscale test hardware are reported in companion Chemical Propulsion Technology Reviews. JHUICPIA CPTR 73, “Solid Propeliant Test Motor Scaling,” September 2001, reviews recommendations on current burning rate measurement test techniques and subscale test hardware for accurate scaling and prediction of internal ballistics of a full-scale solid propellant motor. Detailed recommendations on current burning rate measurement test techniques and subscale test hardware for accurate prediction of internal ballistics of a full-scale solid propellant motor are reported in JHU/CPIA CPTR 74, “Solid Propellant Subscale Burning Rate «Test Techniques and Hardware for U.S. and Selected NATO Facilites,” July 2001. Recommendations on non-intrusive burning rate measurement methods are to be reported in a ‘companion Chemical Propulsion Technology Review to be released at a future date. A complete report of the NATO RTO AVT WG 016 activities is available. “The metric system of units is employed in this report ‘except where industry convention dictates otherwise. ‘The author wishes to express his appreciation to Professor L.T. De Luca, Dipartimento di Energetica, Politecnico di Milan, Italy and Dr. Guy M.H.J.L. Gadiot, TNO Prins Marurits Laboratory, The Netherlands ‘or their contributions to portions of this document. CPIA solicits comments on the CPTR effort, including suggestions on topics for future issues. For technical comments or suggestions contact Mr. Tom Moore, CPIA Technical Services Supervisor, at 410- 1992-9951, ext 207, or the author Mr. Ronald Fry, CPIA Senior Research Engineer, at 410-992-9951, ext. 206. Individuals employed by organizations that subscribe to CPIA services may request personal copies of this document by contacting CPIA at 410-992-7300, cpia@ihu.edu, or http:hwww.cpia.jnu.edu.ABSTRACT Current methods used within the NATO community for analyzing small motor burning rate test data are reviewed and recommendations are made to support improved prediction of internal ballistics of a full- scale solid propellant motor. ‘The NATO Research and Technology Organization (RTO), Advanced Vehicle Technology (AVT), Working Group (WG) 016 (formerly AGARD/PEP Working Group #27) undertook to evaluate methods used within the NATO propulsion community to measure burning rate in solid propellant rocket systems, with the purpose of identifying similarities and differences between the member nations. This WG was formed in 11996, consisting of representatives from 6 of the 15 member nations of NATO, with inputs accepted from 4 other member nations and a couple non-member nations. The NATO RTOIAVT WG 016 sought to contribute to improvements in the burning rate tools to address issues that have plagued the solid Propulsion industry for over 40 years: (1) Better understanding of burning rate data from various faciiies to ease the comparison of propellants from various manufacturers and to improve international exchanges and cooperation. (2) Better accuracy and reliability of measurements allowing a decrease in the number of tests (and ‘associated time and cost) and an improved control of manufacturing and aging. ‘Simulated and real subscale rocket motor data were used to evaluate the two fundamentally different families of burning rate analysis methods. While organizational preferences generally dictate method usage, surveys indicate a trend toward methods that more effectively account for non-idealtailoff, favoring improved accuracy. Consistency in these definitions would promote ease in correlating data internationally. Further development of the Hessler-Glick method shows promise. The NATO propulsion community is urged to review these findings as a means of advancing their own buming rate measurement and analysis methods.PREFACE ABSTRACT... NOMENCLATURE. LIST OF FIGURES.... LIST OF TABLES. 10 20 3.0 40 50 60 70 CONTENTS INTRODUCTION... 14 Objectives. 112 Scope of the Working Group 13. Technical Approach... 414 NATO RTO AVT Working Group 016 Membership and Paricipants BURNING RATE FUNDAMENTALS... 24 Burning Rate Physics and Features . 2.2 Burning Rate LAWS nen 23 — Combustion Stability 24 — Burming Rate Measurement Methods... 2.5 Burning Rate Measurements in Subscale Motors BURNING RATE MEASUREMENT ANALYSIS METHODS. 3.1 Thickness/Time (TOT) Rate 3.2 Mass Balance (MB) Rate. 3.3 Grain Web Thickness Definitions... 34 Burning Time Definitions. 35 Average Pressure Definitions. 3.6 Comparison of Analysis Methods .. INTERNATIONAL SURVEY OF ANALYSIS METHODS AND DEFINITIONS. 4.1. Review of Historical Surveys .. 42 Current NATO WG016 Survey ASSESSMENT OF METHODS USING SIMULATED MOTOR DATA. 5.1 International Round Robin #1 Results... 5.2 International Round Robin #2 Results . 5.3 _ International Round Robin #3 Result . 5.4 International Round Robin #3X Results. 5.5 Lessons Learned From Round Robin Simulated Motor Data ASSESSMENT OF METHODS USING REAL MOTOR DATA... 61 FIATAVIO BPD Motor Data. 62 — TNO-PML Motor Data. 63 Lessons Leamed From Real Motor Data Analysis ‘SMALL MOTOR DATA QUALITY. 7.1 General 7.2 Propellant Effect 7.3 Testing Effects. 74 Uncertainty and Error Analysis.8.0 ANALYSIS METHOD SUMMARY ... 84 82 83 84 9.0 CONCLUSIONS ~ RECOMMENDATIONS. 94 10.0 REFERENCES APPENDIX A. APPENDIX B. APPENDIX C. APPENDIX D. APPENDIX E. APPENDIX F. Fundamental Definitions . ‘Comparison of Analysis Methods . Assessment of Analysis Methods Using Simulated Motor Data Assessment of Analysis Methods Using Real Motor Data. General Burning Rate Fundament Test Hardware and Measurement Methods Analysis Methods. s Non-Intrusive Measurement Methods. Test Motor Scaiin Future Developments . NATO PROPULSION INDUSTRY CONTRIBUTORS (2 PGS) NATO PROPULSION INDUSTRY CONTRIBUTORS (68 PGS) NATO PROPULSION INDUSTRY CONTRIBUTORS (17 PGS).. NATO PROPULSION INDUSTRY CONTRIBUTORS (22 PGS).. CHEMICAL PROPULSION TECHNOLOGY REVIEWS ISSUED BY CPIA.... soon EA INITIAL DISTRIBUTION..... viNOMENCLATURE pressure coefficient of ballistic steady burning rate, nondimemsional area of the burning surface, cm? area of the grain port cross-section, cm* area of the nozzle throat, cm? circular center perforated grain ‘end burner grain standard acceleration gravity at sea level, 9.807 mis? ratio of burning surface area A, to nozzle throat area A, nondim. ratio of the combustor cavity volume to nozzle throat area, m mass burning rate, glom’s pressure exponent of ballistic steady burning rate, nondimemsional pressure, MPa : P: combustion chamber, pressure, MPa % burning rate, cm/s Tye Mass balance burning rate, cm/s fror__thickness/time burning rate, cm/s SR universal gas constant, 1.987 cal/mole, K SCP star center perforated grain t time, s t, —__buming time, s T temperature, K Ty initial propellant temperature, K T, combustion chamber gas temperature, K Try reference temperature (298 K) V, combustion chamber free volume, cm® w, web thickness, mm. Deanne ggnare Greek Symbols k —_burmed gas specific heat ratio M average molecular mass, gimole % temperature sensitivity of steady chamber pressure, K* Pp. combustion chamber gas density, g/cm? % —_propettant density, g/cm? ©, __ temperature sensitivity of steady burning rate, K" ‘Subscripts amb ambient avg average b- burning © chamber max — maximum Nomenclature unique to each analysis method is provided within Appendix B. vilFIGUI Figure 4 Figure 2. Figure 3. Figure 4. Figure 5. Figure 6. Figure 7. Figure 8 Figure 9. Figure 10. Figure 11. Figure 12. Figure 13. Figure 14, Figure 15. Figure 16. Figure 17, Figure 18. Figure 19. Figure 20. Figure 21 Figure 22. Figure 23. LIST OF FIGURES TILE PAGE ‘Simplified Diagrams of Several Grain Configurations Burning Surface Does Not Remain Perpendicular tothe Axis in Larger End-Burning Grains Classification of Grains According to Pressure-Time Characteristics. Various Burning Rate versus Pressure Relationships Basic One-Dimensional View of Double Base Propellant Burning, ‘Schematic of Granular Diffusion Flame (GDF) Two-Stage Flame Structure for AP-based Composite Solid Propellant... ‘Schematic of Beckstead-Derr-Price (BDP) Mutiple Flame Structure of AP-based Propeliants.... Acoustic Oscillation Modes in a Rigid, Closed Cylindrical Cavity... Modem Crawford Bomb (Solid or Liquid Strands). ‘Traditional Power Law Burning Rate Behavior. Behavior of Solid and Liquid Strand Burning Rate Relative to Motor Size. Various Burning Time Definitions ‘Typical Web Shape of CP Grain Definitions of the Burning Times. “Hump” Curves for 5-inch CP Grain with 3-inch Bore, 9 inch Length Definitions of Tangent-Bisector Method Definitions of the Brooks Method... Pressure versus Time Traces for the Round Robin #1 Cases RR #1 — Relative Burning Rate Error (%) at 7 MPa and 10 MPa Reference Pressures.. RR #1 ~ Relative Exponent Error (%) at 7 MPa and 10 MPa Reference Pressure: RR #1 — Relative Coefficient Error (%) at 7 MPa and 10 MPa Reference Pressures... RR #1 ~ Burning Rate versus Pressure Behavior for all 4 Cases ‘Compared with Reference... RR #1 - Buming Rate versus Pressure Behavior for Case 1 Compared with Reference (Expanded Scale)Figure 24. Figure 25. Figure 26, Figure 27. Figure 28. Figure 29. Figure 30. Figure 31. Figure 32, Figure 33, Figure 34. Figure 36. Figure 36. Figure 37. Figure 38. Figure 39. Figure 40. Figure 41. Figure 42. Figure 43. Figure 44. Figure 45. Figure 46. Figure 47. Figure 48, Measured Reference Rate at 10 MPa and Exponent versus Pressure Level at Beginning of Burning Definition. Pressure-Time Behavior for Round Robin #2 Test Cases RR #2_- TNO-PML Round Robin Results (All Cases). RR #2 - TNO-PML Round Robin Results (Low Pressure) RR #2 - TNO-PML Round Robin Results (High Pressure). RR #2 —Comparison of Burning Rate Error (%) at 7 MPa Relative to the Reference Rate F;,. without Web Correction (Data from Table 18). RR #2 — Comparison of Power Law Exponent Error (%) Relative to Reference (Data from Table 17) RR #2— Comparisons of Power Law Coefficient Error (%) Relative to Reference (Data from Table 18)... BRR #3 — Pressure-Time Behavior for Round Robin #3, Group 1 with Concentric Grain Bore Cases : RR #3 -Pressure-Time Behave for Round Robin #3, Group 2 with Grain Bore Offset Cases. : Comparison of Burnout and Tailoff Processes for RR #3 Group 1 Baseline and Group 2 with Grain Bore Offset. 63 RR #3 — Group 1 Baseline Relative Burning Rate at Reference Pressures 7 and 10 MPA nnn eS —— and 10 MPa.. RR #3— Group 1 Baseline Burning Rate Relative Error (%) RR #3 — Group 2 Off-Axis Bore Burning Rate Relative Error (% RR #3 — Group 1 Baseline Constant (a) Relative Error (%). RR #3 - Group 1 Baseline Exponent (n) Relative Error (% RR #3 - Group 2 Off-Axis Constant (a) Relative Error (2). RR #3 — Group 2 Off-Axis Exponent (n) Relative Error (%)... Rate versus Pressure for RR #3, Group 1 Errors in Thickness/Time Rate Methods for RR #3.. Errors in Mass Balance Rate Methods for RR #3 Deviations From Fitted Rate-Pressure Line for RR #3..... Errors in Iterative Two-Point Thickness/Time Rate for RR #3 RR #3X Group 3 with Progressive Pressure Trace Shape.Figure 49, Figure 50. Figure 51. Figure 52. Figure 53. Figure 54. Figure 55. Figure 56, Figure 57. Figure 58. Figure 59. Figure 60. Figure 61. Figure 62. Figure 63. Figure 64, Figure 65. Figure 66, Figure 67, Figure 68. Figure 69. Figure 70. Figure 71. Figure 72. Figure 73. Figure 74 Figure 75. Figure 76. RR #3X Group 4 with Regressive Pressure Trace Shape .. RR #3X Group 5 with Offset. Groups 6-8 and 10 Similar, but with Variations of Ignition Pressurization Rates and Tailoff RR #3X Group 9 with Higher Rate Equation and Bore Offset. Group 1 Baseline Progressive Burn Relative Error. Group 2 Baseline Regressive Burn Relative Error... Group 3 Baseline Progressive Burn Relative Error. Group 4 Baseline Regressive Burn Relative Error.. Group 5 Baseline Progressive Burn Relative Error. Group 6 Baseline Regressive Burn Relative Error. Group 7 Baseline Progressive Burn Relative Erro.... Group 8 Baseline Regressive Burn Relative Error. Group 9 Baseline Progressive Bum Relative Err0F .... Group 10 Baseline Regressive Burn Relative Error. Relative Burning Rate Error for All RR #3 and #3X Perturbations Using BC and HG Analysis Methods.. Average Normalized Pressure-Time Behavior of Baria Motor Tests .. Comparison of Filtered with Raw P-t Data (p,: black points with line, Ps, Pie, Pao) ‘Normalized First Derivatives (p's/py, P'ss/Ps, P'avPre: P's/Pae). First Derivatives During Burnout Ignition of Nominally Identical Motors Tested at Three Pressures... Burning Rate Residuals Dependence on Estimated Values of Burning Rate. Pressure-Time Behavior for TNO-3 Motor Firing Pressure-Time Behavior for TNO-4 Motor Firing Pressure-Time Behavior for TNO-5 Motor Firing Pressure-Time Behavior for TNO-6 Motor Firing... Pressure-Time Behavior for TNO-7 Motor Firing ‘Second Derivative of a Typical Pressure-Time Tail-off Transient. 8882828 BB ‘Second Derivative of a Filtered Pressure-Time Tall-off Transient... ‘2x4-inch Grain Deformation at +20" C and -54" C.Figure 77, Figure 78. Figure 79. Figure 80. Figure 81. Figure 82. Strain Situation of the 2x4-inch Grain at -54" C (Deformation x3)... Normalized Motor Pressure and Low-Level Pressure Oscillations. Filtered Waterfall of Log-Magnitude Normalized Pressure (in % of Mean Pressure) Fitted FFT at 50% Burn, M=70.... ‘Magnitude of Forcing Function (Top as Measured; Bottom Corrected for Post-Firing Noise). 07 Multiple Small Natural Pulses Occur Throughout Motor Operation, But Appear to Maximize Near 58% of Bumn.. 108 xiTABLE Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. Table 9. Table 10, Table 11. Table 12. Table 13, Table 14. Table 15, Table 16. Table 17. Table 18. Table 19. Table 20, Table 21 Table 22. Table 23. Table 24. Table 25. LIST OF TABLES TLE PAGE NATO RTO AVT WG 016 Membership... Evaluating Effect Chamber Volume Increase. Buming Time Definition Buming Time Definitions Summarized in Miller & Barrington's 1969 Review. Historical Sources of Characteristic Time Definitions in US, ‘Summary of Data Analysis Methods (Refer to Appendix B) .. ‘Summary of Burning Rate Analysis Methods and Definitions. ‘Summary of Frequency of Use of Analysis Methods and Burn Time Definitions. Buming Rate Analysis Round Robin Summary... RR #1 Small Motor Ballistic Simulations Examined .. ‘Typical Burning Rate Analysis Results NATO RTO/AVT WG016 RR #1 RR #1 — Power Law Relations Obtained from Correlating the Recovered Data, RR #1 ~ Variation of Exponent and Reference Rate with Mean Pressure Definition and Burn Time Definition... RR #2 Small Motor Ballstic Simulations Examined ... RR #2~ Typical Burning Rate Analysis Results NATO RTOIAVT WG016...... RR #2~ Burning Rate r, Determined by Several Participants. RRR #2~ Power Law Exponent (n) Determined by Several Participants .. RR #2~ Power Law Coefficient (a) Determined by Several Participants RR #2 ~ Relative Burning Rates at 7 MPa (Relative to Baseline tycasec 1.2 and to Reference fn) RR #3 Small Motor Ballistic Simulations Examine RR #3— Group 1 Combined Baseline Results and Relative Errors....... RR #3— Group 2 Combined Off-Axis Bore Results and Relative Errors RR #3X Small Motor Ballistic Simulations Examined.. RR #3 Reference Burning Rate Value: Percent Relative Error (1-Fyfyyq) of Burning Rates for BC and HG Methods . xiTable 26. Table 27. Table 28. Table 29. Table 30. Table 31. Table 32, Table 33, Mean Values of Ballistic Parameters of B Series at Reference Pressure 4.5 MPa for Modified HG and Standard BPD-1 Methods. Overall Values of Ballistic Parameters and Correlation Factor of All Baria Data Groups at Reference Pressure 4.5 MPa Mean Values of Ballistic Parameters of All Baria Data Groups At Reference Pressure 4.5 MPa (standard deviation for 36 groups of fire tests)... ‘Averaged Mix Variability of all Baria Series At Reference Pressure 4.5 MPa. Mean Values of Ballistic Parameters of A4 Series at Reference Pressure 4.5 MPa (standard deviation for 4 groups of motor tests). Mix Variability of A4 Series at Reference Pressure 4.5 MPa ‘Average Pressure and Burning Rate Analysis Using TNO, BC and HG Methods... Overall Values of Ballistic Parameters and Correlation Factor of TNO Tests at Reference Pressure 7 MPa for TNO, BC, and Modified HG Methods...4.0 INTRODUCTION ‘The rocket motor designer must have a good understanding of the variation of propellant burning rate with both pressure and temperature in order to produce an efficient design and minimize design iterations during development. Our understanding of burning rate analysis methods can be improved by examining practices employed by various facilities and countries engaged in the measurement and analysis of burning rate in solid propellant systems. 1.1 Objectives This report reviews solid propellant burning rate analysis methods used by over 20 faciliies from 7 NATO member countries. This represents a complete survey of all the international facilities involved in burning rate measurement, and is a thorough representation of the fundamental methods used in the solid propulsion community today. The historical basis and fundamental factors influencing these methods are reviewed, including the advantages and disadvantages encountered in their use. Survey results from the participants of a NATO Working Group are presented and discussed. Conclusions are drawn and recommendations are made for future applications. 4.1.4. NATOIRTO AVT Working Group Formation ‘The NATO Research and Technology Organization (RTO), Advanced Vehicle Technology (AVT), Working Group (WG) 016 (formerly AGARDIPEP Working Group #27) undertook to evaluate methods used within the NATO propulsion community to measure burning rate in solid propellant rocket systems. This report summarizes the objectives, approach, findings, and recommendations relative to test techniques and hardware used by the surveyed countries and facilities. A complete report of the NATO RTO AVT WG 016 activities is available’, The WG was formed in 1996, consisting of representatives from 6 of the 15 member nations of NATO, with inputs accepted from 4 other member nations and a couple non-member rations. The WG conducted its activity from October 1997 to March 2001. The justification and relevance ofthis task to the Solid Rocket Industry includes the importance of solid propulsion to tactical and strategic rockets, missiles and space launch systems; the influence solid propellant burning rate has on performance; and the influence burning rate testing has on program costs.*$ 1.1.2 Justification for Studying Solid Propellant Burning Rate Solid Rocket propulsion remains the major propulsion concept for the tactical and strategic missiles, and for many first stage launch systems. Among the parameters controlling the solid rocket motor operation, burning rate plays a very important role. The burning rate determines, with the buming area, the ‘combustion processes, the mass flow rate, and therefore directly controls the pressure and thrust of the motor. Burning rate is a characteristic of the propellant that can be measured independently, at least for the more usual combustion regimes. Accuracy of solid rocket thrust-time prediction has become increasingly more important in solid rocket design. One of the most significant variables in this prediction is the propellant burning rate. Accuracy of this value depends on empirical methods for calculating burning rate from subscale motor tests and for correlating this rate with predictions derived from full-scale motor tests. Thrust is very sensitive to the reference propellant burning rate. A variation in propellant burning rate of +1%, for example, wil result in 2a thrust variation of 1.5 — 2%. The methods of determining burning rate must be reliable and correlations dependable to predict thrust to an accuracy of +3% using the Solid Performance Program (SPP). Burning rate measurement is an important and significant activity in the solid propeltant industry is devoted to, first during the development of a new propellant, then during the manufacturing (quality control), or for the service life (aging). All the countries with a tradition in the development and manufacturing of solid propellants are equipped with facilities for the burning rate measurement. These facilities are being ‘continuously improved to increase the accuracy and reliability of the burning rate data.‘The NATO RTO AVT WG 016 sought to contribute to improvements in the burning rate tools to provide: ‘+ Better understanding of burning rate, r,(p,T,), data from various facilties to ease the comparison of propeliants from various manufacturers and to improve international exchanges and cooperation. ‘Improved measurement accuracy and relabilty to allow a decrease in the number of tests (and associated time and cost) and improved control of manufacturing and quality assurance and the assessment of aging. 1.2 Scope of the Working Group Activity ‘The working group has reviewed and compared methods for measuring steady-state burning rate of solid rocket propellant through current subscale motor practices with an emphasis on data analysis methods and non-intrusive techniques. The overall focus of the working group was approximately 70% small ‘motors, 25% non-intrusive diagnostics and 5% other methods including strand burners. After three years of technical interchange meetings, the AVT WG 016 (formerly AGARDIPEP Working Group #27), whose charter was “Evaluation of Methods for Solid Propellant Burning Rate Measurements” completed its last ‘meeting in Ottawa, Canada on 18-23 October 1999. Six NATO countries were actively participating, with a few others providing technical support. This report is the product of the six AVT WG 016 Meetings ‘conducted between October 1996 and October 1999, supplemented with collaboration by WG members between the meetings and throughout CY2000-2001 during preparation of the final report. Specific objectives of WG 016 relevant to this report are to: 2) Review the small-scale motors used by the various NATO countries and the problems encountered. b) Compare measurement methods and evaluate the differences ©) Produce an Advisory Report to the NATO community with the following features: 41) Provide information suitable as training for entry-level person and reference for the expert. 2) Collect, analyze and condense information into a language understandable to @ wide range of technologists and managers. 4.3 Technical Approach 134 General Summary WG activity included analyses, presentations, and discussions in support of completing a final RTO AVT ‘advisory report. Topics addressed in this report include (1) introduction to the problem, justification for the WG and definition of the technical approach, (2) review of burning rate fundamentals, (3) review of basic analysis methods, (4) discussion of survey of methods used in the intemational community, (5) ‘assessment of methods using simulated motor data, (6) assessment of methods using real motor dala, (7) discussion of small motor data quality, and finally (B) summary, conclusions and recommendations. ‘The basic technical approach used to address the overall and specific objectives cited above has involved a) Surveying the NATO solid propulsion community for subscale and non-intrusive test methods, analysis, and scaling methods, and b) Analyzing “simulated” and “real” motor pressure-time data using multiple thickness/time and ‘mass conservation burning rate analysis methods taken from the survey. Time-consuming survey and analysis support was solicited from a wide range of facilities within the NATO solid propulsion community during the course of this effort. Four separate Analysis Round Robins were conducted with solicitations for support made to NATO propulsion industry contributors. The influence of various burning behaviors was examined in these round robins, such as progressive or regressive burning, constant and random bore offset variations, constant and random L* variations, and different rate equations. Results of the surveys on analysis methods and results of the round robins are reviewed in this report. Trends in observed differences in calculated burning rate for the different analysis methods were evaluated for these cases with a WG goal of making recommendations on preferred analysis methods. Detailed recommendations on solid propellant test motor scaling are reported in JHUICPIA CPTR 73." Detailed recommendations on current burning rate measurement test techniquesand subscale test hardware for accurate prediction of internal ballistics of a full-scale solid propellant motor are reported in JHU/CPIA CPTR 74. A complete report of the NATO RTO AVT WG 016 activities is available.’ Recommendations on non-intrusive burning rate measurement methods are to be reported ina companion Chemical Propulsion Technology Review to be released at a future date. 1.32 WG 016as a Catalyst for Change ‘Analysis Round Robins were used in analyzing “simulated” and “real” motor pressure-time data for the ‘purposes of involving the participants in the data analysis, review and discussion, and conclusion process. ‘The WG considered ths critical f either voluntary change was to be expected as an outcome of this study, or if resistance to changed procurement specifications was to be forestalled. One concern was, if the participants were not iniimately involved in this process and were only acquainted with the finished ‘comparisons/conclusions, that inertia and conservatism would likely dominate any subsequent actions. Steps to involve the participants in this process take time. Clearly participation and peer review ‘was integral to activities as members of this WG. Peer review of the comparisons and the resulting conclusions was sought outside this WG on a selected basis throughout the period of this effort. Continuing efforts in this regard will be beneficial to understanding the merits of, or for taking further action (on any conclusions drawn from this study. 4.4 NATO RTO AVT Working Group 016 Membership and Participants Dr. P. Kuentzmann of ONERA, France, initiated the WG in 1997 under the former AGARD Propulsion and Energetics Panel (PEP), now the Applied Vehicle Technology (AVT) panel of the Research and Technology Organization (RTO), a Working Group 016 with the charter of "Evaluation of Methods for Solid Propellant Burning Rate Measurements." His early vision as advocate for this effort is appreciated. ‘The primary AVT Working Group 016 membership included: ‘Table 1 NATO RTO AVT WG 016 Membership ‘Me, Ronald Fry (Co-Chairman) | JHUICPIA us. ‘Dr. Robert Frederick Univ Alabama in Huntsvile | U.S. Mr. Rene Couturier (Co-Chairman) | SNPE. France Mr. Dominique Ribereau ‘SNPE France Mr. Jean-Paul Reynard ‘NERA France Mr. Jean-Claude Traineau_ ‘ONERA France Dr. Hans-Ludwig Besser Bayer-Chemie ‘Germany Dr. Rudiger Strecker Bayern-Chemie ‘Germany Prof. Luigi DeLuca Palitecnico di Milano. Italy Dr. Guy MHJL. Gadiot "TNO PML. Netherlands ‘Mr. Tony Whitehouse Royal Ordinance UK ‘The WG016 members most gratefully acknowledge the significant contributions of Mr. Richard Hessler, independent consultant to the WGO16 from the U.S. Additionally, the author the WG members are sincerely grateful to the NATO international propulsion community (facillies and their representatives) for their contributions included in this report. Over 50 contributors participated from over 35 companies, universities and agencies. These contributions included information on test hardware, analysis and ‘scaling methods, and support for multiple analysis round robins of simulated and real motor data. A complete list is provided in Appendix A.2.0 BURNING RATE FUNDAMENTALS. 2.1 Burming Rate Physics and Features Knowing burning rates of solid propellants, whether steady or unsteady, under a variety of operating conditions is of critical importance both for applications (due to their sensible influence on performances and cost of propulsive devices) and fundamental reasons (understanding of combustion processes). Furthermore, since no available theory/model is capable of predicting burning rates with accuracies within 1% and including the effects of rate modifiers, they must be measured experimentally. However, while experiments measuring steady burning rates are reasonably robust, those measuring unsteady values are fragile and stil a matter of research. Since a variety of experimental hardware and procedures are in use today, even for the common steady-state operations, the need arises to understand and perhaps standardize the different approaches developed among the NATO countries. 21.1 Burning Rate Physics* 2.1.4.1 Background Energetic materials in general are capable of a dual reacting regime: © Supersonic regime: a combustion wave preceded by a strong shock wave brings about a detonation wave, propagating at a speed on the order of several km/s and limited by the total thermochemical energy content of the reacting material; * Subsonic regime: a combustion wave brings about a deflagration wave, propagating at a speed on the order of cm/s and limited by heat and/or mass diffusion. For a more detailed background, which lies outside the scope of this writing, the interested reader may wish to consult."**"° Here it is enough to remark that deflagration is the common operating mode for the vast majority of engineering applications. Thus, only subsonic combustion waves (or deflagration waves) ‘are considered in this report. ‘Whether steady or unsteady, deflagration waves in energetic solid materials in general consist of an initial condensed phase and a final phase, and in most cases essentially gaseous reaction products. The interface between the condensed phase and gas phase is called the burning surface. The propagation rate of this interface is called buming rate; physically, this can also be seen as the regression rate of the condensed phase. For many studies itis convenient to define, more precisely, a linear burring rate (or delagration rate) as ‘the web thickness burned per unit time in the direction perpendicular to the burning surface. 2.1.1.2 Internal Ballistics Design and operation of solid rocket motors strongly depend on the combustion features of the propellant charge (burning rate, burning surface, and grain geometry) and their evolution in time. Intemal ballistics is the applied science devoted to these problems.Burning Rate In general, burning rates depend on: Nature of energetic material (basic ingredients and their mixture ratio); © Details of chemical composition (catalysts, modifiers, additives, etc. usually present in small or fractional percentages); Physical effects (particle size distribution, presence of wires or staples, etc.); © Details of manufacturing process and other miscellaneous factors (see Sections 2.2 and 3.0); ‘© Operating conditions (pressure, initial temperature, natural and/or external radiation, heat losses, {925 flow parallel to the burning surface, acceleration, etc.); ‘© Mode of operation (steady vs. unsteady). This report is primarily concerned with the measurement of steady burn rates, implying a steady set of operating conditions and equilibrium combustion. For propulsive applications, the influences of pressure (typically, in a range from 1-30 MPa) and initial temperature (typically, in a range from 219-344 K for airaunched missile motors) on burning rate are of paramount importance. Natural radiation is important for heavily metallized compositions (15-20% metal addition), while external radiation still is a matter of laboratory experiments; heat losses are important only under special circumstances. High velocity gas flowing parallel to the buming surface can seriously increase the local burning rate (causing the so-called erosive burning phenomenon), due to increased heat transfer from the adjacent turbulent boundary layer, especially in the aft-end portion of the motor ‘avity. Motor acceleration larger than 10 gp, whether longitudinal or lateral or due to spinning motion, directed into the burning surface and within an angle of 60 to 90 degrees with respect to it, perceivably increases buming rates. Other peculiar ballistic effects, due to details of manufacturing process, may be important for motor operations but are sensibly dependent on the actual configuration. Detailed comments are discussed in later sections of this report, with further reading in References.°""*? Notwithstanding impressive progress, combustion theory is not yet capable of predicting steady or unsteady rates with sufficient accuracy for routine use in motor predictions. Thus, propulsion designers and engineers require experimental measurements. Burning Surface ‘The burning surface of solid energetic materials regresses in a direction essentially perpendicular to itse. In other words, solid propellants are considered to burn by parallel layers and the grain “tends to retain its original configuration until the web has burned through” (Robert's law, 1839); for details.” Notice that this law, originally proposed for homogeneous compositions, can be extended to the modern heterogeneous compositions if the propellant heterogeneity is limited to a “sufficiently small scale.” ” The actual burning surface and its evolution in time depend on the initial grain geometry and overall combustion processes. Grain Geometry ‘The initial grain geometry of a solid propellant strictly depends on the propulsive mission. See Figure 1° fora variety of shapes commonly employed. The folowing nomenciatue is curently used: ‘Grain configuration: the designed shape ofthe intial burning surfaces of a grain in a motor. ‘* End-burning grain: the propellant grain is @ solid cylinder ideally burning, ike a cigarette, only in the axial direction. '* Cylindrical grain: a propellant grain in which the internal cross section is constant along the axis regardless of perforation shape. '* Perforation: the central cavity port or low passage of a propetiant grain. '* _ Inbibitor: a layer or coating of slow- or non-burning material covering parts of the grain’s propellant surface to prevent buring, ‘+ Restricted surface: a grain surface restricted from burning by the bonding of an inhibitor layer. © Sliver: unburned propellant remaining (or lost because ejected through the nozzle) at the time of web bumout.Bonses insulation Chamber Enasburmer (ease bonded, nearal burn Interval baring tbe, propessive ©oO0 ‘Sts and tbe, neural burn Rada grooves and ube, neal burn © ‘Sterner Wagan wee! Mutipertorated (ewra (orogressiesegesie) Dag Bone Dena (esse bonded Figure 1. Simplified Diagrams of Several Grain Configurations.” Motor Pressure Let us assume uniform pressure and buming rate throughout the combustion chamber of a solid propellant rocket motor filled with a perfect gas bumed mixture. All properties are considered constant. Transient mass conservation requires ale.) o m, + ‘where the mass production of gas due to combustion is, m, vAsts the mass flow rate exiting the nozzle is‘and the mass accumulation rate in the combustion chamber is MOL) op Te vy Bem pA rl, where c* 1 [En eae is the characteristic velocity. By substitution in the mass conservation equation, one finds the transient equation of the internal ballistics dpe Te?) Det 2) By PrP) Aare Ape Under steady operations, one oblains the equilibrium pressure of the rocket motor combustion chamber co) where the steady burning rate has been taken as 21.2 Buming Rate Features Under any circumstances, ideal one-dimensional steady-state combustion waves, if not impossible, are at least very rare. It is important to realize that, besides the main factors summarized in Section 2.1.1, a variety of details conspire against the establishment of an ideal combustion wave. Even for simple strand burners these factors can include size of the sample, lateral surface inhibitor, ignition, and nature and flow rate of the ambient gas, radiative environment, and other factors. Under actual motor operating conditions, further effects worsen the situation even for the simple end-burner configuration. These additional factors include grain processing details, aging, mechanical stresses, contacts with walls, interfaces with inhibitors, migration of various propellant ingredients, rate of polymerization, and propellant state of cure for composite propellants.’® The high strain grain surface near the bond line of targe end-burning grains, for ‘example, encourages the burning surface to become conical from its initially planar shape, as shown in Figure 2° In larger end-buring grains (above approximately 0.5 m diameter) the buming surface does not remain perpendicular fo the axis, but gradually increases and assumes a conical shape. The burning is larger than in the center. The lines in the grain indicate successive burning surface ‘most cases, including small motors for ballistic evaluation, one-dimensional steady- state combustion wave is only an ideal picture. The factors influencing non-ideal buming are reviewed in ‘more detail in later sections relative to the issue of burning rate scaling. Only the general classes of non- ideal burning are reviewed in this section.Equiitvam, iti oneal burning uring sutace Wy Figure 2. Bu ‘Surface Does Not Remain Perpendicular to the Axis in Larger End-Buring Grains.* 2.1.2.1 Quasi-Steady Burning For heterogeneous compositions, the combustion wave is by definition the result of local 3-D and unsteady effects depending on ‘the initial loading fractions and particle size distributions of solid ingredients. On a microscale, unsteady effects are due to the arrival of different ingredients at the burning surface and changes of local thermophysical properties of each ingredient. Even for homogeneous compositions (double-base and triple-base solid propellants), different chemical reactions of the ingredients produce unsteady effects due to changes of chemical reaction rates and possible ‘accumulation of carbonaceous residues on scattered sites of the burning surface. In general, the ideal uniform burning surface is rare to achieve due to foam, bubbles, hot spots, material dispersion, etc. ‘occurring for @ variety of reasons."* Thus, steady-state combustion processes have to be seen more properly as quasi-steady in time, and measurements of linear burning rates must be taken over distances ‘much farger than the corresponding thermal wave thickness (as well as the distance required to establish), 2.1.2.2 Mean Steady Burning / Neutral In general, steady burning rates have to be seen as a mean value occurring over an appropriate time span. In particular, ignition and extinction transients have always to be excluded from the measurement procedure; but in general this is systematically done only in strand burners and ultrasound burners or other specialized rigs. Under these circumstances, a reacting propellant often shows a mean burning rate that is constant in time. With reference to 2 motor, the combustion process is said to be neutral if chamber Pressure or thrust behavior are maintained constant in time; but for neutral burn rate, pressure only is Fequired to keep constant in time. Note that (slowly varying) excursions within a typical but arbitrary fraction of 15% of the average value are accepted.” Under actual operating conditions, however, peculiar motor effects may affect buming rates yielding unwanted consequences (see later sections). For example, hump effects for cast composite propeliants manifest an excess burning rate of 3-7% at about halfway through the web."" Thus, neutral burning is a very convenient configuration but not easy to obtain. ‘Most small motors for ballistic evaluation are meant to be neutral 2.1.2.3 Transient Burning / Non-Neutral Under transient conditions, burning rate may differ greatly from the equlibrium or steady rate. The degree Of the effect depends on instantaneous operating conditions and their time rates of change, past history, ‘and propellant type (primarily through thermophysical properties). In most applications, pressure cannot be held precisely constant. In motors, transient burning commoniy contributes to a pressure peak at the beginning of operation, and may also cause extinction of remaining slivers during the depressurization at the end of operation. Through the middie portion of operation, the pressure may also vary with timebecause of the grain configuration, nozzle size changes or manufacturing variables, such as the hump effect above. In general, a reacting propellant features a burning rate variable in time. With reference to a motor, the combustion process is said to be non-neutral. In particular, progressive or regressive processes are defined if chamber pressure (or thrust) is overall increasing or decreasing in time (causing excursions wider than 15% of the average value®). The following definitions illustrated in Figure 3. although arbitrary, are commonly accepted: '* Neutral buming: motor burning time during which thrust or chamber pressure remain approximately constant, typically within a corridor of + 15%; '* Progressive buming: motor burning time’ during which thrust or chamber pressure increase (beyond the + 15% corridor); ‘+ Regressive burning: motor buming time during which thrust, chamber pressure, and burning surface area decrease (beyond the 15% corridor). Pressure (or thrust) Time Figure 3. Classification of Grains According to Pressure-Time Characteristics® Even the simplest burners experience dynamic burning effects, at least during the ignition transient, caused by the abrupt hot gas production. For well-designed bumers, the associated rapid pressure change is a minor feature. But even if the pressure transient is avoided, the thermal transient still needs to be dealt with. In strand burners, this is accomplished by allowing the strand to burn some distance after ignition before staring the measurement period; likewise, the measurement period is stopped some distance from the strand end to avoid extinction transient. Also in the so-called non-intrusive burners (ultrasound, microwave, laser recoil, and x-ray) in which “instantaneous” (or very short-term averaged) burning rate measurements are obiained, both the starting and ending transients can be avoided. However, in motors the transients are essentially unavoidable, and are necessarily included in the data analysis because the only length known is the total thickness. In principle under no circumstances, including neutral burning, measurements should be taken during transient operations (ignition and extinction). But should diagnostic techniques capable of direct measurements of burning rates (e.9., ultrasound) be available, high-frequency measurements during transient operations can provide a corresponding time-resolved burning rate history.2.2 Buming Rate Laws 2.2.4. Empirical Burning Rate Laws Under steady conditions and for a given initial temperature, the Vieille or de Saint Robert law? is empirically used to describe the burning rate dependence on pressure Fe=asp « where the two parameters (, and n) are constants experimentally defined over some limited measurement range. Figure 4 compares various burning rate pressure relationships. Propellants showing a region of markedly reduced or zero pressure exponent are known as “plateau” propellants (for example double base propellants with small amounts of lead compounds). Propellants showing small negative values of n over short pressure ranges are called “mesa” propellants. Ofien itis Possible to represent burning rale as a series of straight segments, with different a, and n for various Pressure ranges. To establish a, and for one range of pressure and initial temperature, itis industrial Practice to use 7 runs (3 at the nominal pressure, 2 at the higher and 2 at lower pressure) at normal initial temperature and 5 runs each at expected temperature extremes. og 26 Figure 4. Various Burning Rate versus Pressure Relationships“ For years, the industry standard technique to acquire these data has been based on the so-called Crawford bomb.” The particular form with 1 =1 ap ©) has been used in the past to represent steady burning rates of simple single base (nitrocellulose) gun propellants with some accuracy.” An alternative form borrowed from arillery and called Muraour law"* re=byt+aep 6) describes” the behavior of many double base propellants for pressures above 200 atm as well or better than Eq. (4) and also provides’ good estimates of the constants b, and a,” Basically it yields results similar to Eq. (4) for the pressure interval of interest in rocket propulsion. Another alternative form'® To=b,t+a,p a) 10gives accurate results for many double base propellants over a wide pressure range, but it difers Iitle from both Eq, (4) and Eq, (6) and is anyway inadequate for most rocket propellants.” The classical Granular Difusion Flame (GDF) theory, developed by Summerfield and coworkers **'* ® can be applied to AP-based composite propellants burning at moderate pressures (0.2-0.8 MPa’) leading tothe standard expression Poash-p” @) re where the constants @ and b respectively measure the importance of chemical kinetics and mass diffusion in the gas phase" For lack of better knowledge, the use of Eq. (4) is recommended. ‘As an allerative for some particular compositions, the ‘normal” ballistic law (first proposed by Zeldovich™**in 1942 and much used in the Russian literature” * can be implemented. This burning rate law, of exponential form, under steady conditions is usually written as imz.(Ts) = M, exp(—- E; ) © where Miz = ),7»,2e_ is the steady surface mass buming rate and the pre-exponential factor M, is the (asymptotic) maximum mass burning rate. The relationship of Eq. (9) was experimentally shown fo hold as an “universal” law for particular compositions, over a wide range of pressure and initial temperature, by taking proper values of the relevant constants: for all known® double-base propellants (DBP) and nitrocellulose (NC) the activation temperature E,¥/%t =5000 K and the pre-exponential factor M,=1.8 10° glom’s* In this report, for convenience, the steady burning rate of Eq. (9) is called the normal or Zeldovich buming rate law. 22.2 Pressure and Temperature Sensitivity If 7» is the steady linear burning rate, the commonly accepted definition (see Figure 4) for the steady burning rate pressure sensitivity is v-(28i) -2() «o Olnp ),, tp J, while for the steady burning rate temperature sensitivity® (11) Practically, over the appropriate pressure and initial temperature intervals, the familiar empirical steady relationships? are used (Tue) P-exaler (ToT) 12) being a4()=24 (Tey) explo, (To ~The] (3) where Try is the reference ambient temperature, Ty is the actual initial or ambient temperature, n and 0, are conslants over some limited operating range. In this report, for convenience, the steady buming rate of Eq, (12), obtained from experiments, is called the generalized power burning rate law also referred to as Vieille (or Saint Robert) buming rate law. The steady burning rate laws of Eq. (12) and Eq. (4) are convenient widely used over the appropriate range of operating conditions. Notice that Eq. (12) implies the "simplifying assumption that n=n(p) while a,=a,(T,), as often observed due to the limited range of industrial burn rate testing, But this is not necessarily true in general. However, ifthe assumption is kept, then (14) (15) from which over some limited operating range PAT) = P(Trer)- © (To ~ Trr)] (16) Moreover, combination of Eq. (12) and Eq. (16) yields (7) valid # the relevant parameters are constant and small, Le. o,(TeTel<* The buming process can be seen as one-dimensional, but several zones can be distinguished where chemistry plays @ dominant role. In the gas-phase a fizz-zone (strong thermal gradient), dark-zone (varishing thermal gradient, flame temperature around 1500 K, intermediate products), and luminous zone (strong thermal gradient, final flame temperature, final products) are identified. In the condensed-phase a rate controlled concentrated surface decomposition is assumed * or, more likely, a distributed foam zone involving exothermic processes and partial gasification of the solid propeliant.* For increasing pressure, the overall flame thickness decreases. However, for pressures below 0.15 MPa the luminous zone cannot take place for kinetic reasons and the final temperature is that of the dark-zone. 12BURNING: __. ———_—— ——— Figure 5. Basic One-Dimensional View of Double Base Propellant Burning **" Several mode's were proposed for heterogeneous compositions as well, but rarely based_on first principles due to the intrinsically 3D and unsteady nature of the associated burning processes. The GDF {two-stage flame model by Summerfield and coworkers "* "(Figure 6) and the BOP * multiple flame model (Figure 7) are prominent examples. In the GDF model the burning surface includes dissociative sublimation of NH,CIO, into NH,+HCIO, over the oxidizer surface and endothermic zero-order pyrolysis of the solid fuel, A’ double flame structure is then portrayed in the gas-phase: a primary premixed monopropeliant flame (between NH, and HCIO, provided by the oxidizer dissociation) followed by a final diffusion flame between gaseous fuel pockets and the oxidizing atmosphere of the premixed flame products. The premixed flame is seen as very thin as compared to the diffusion flame; for pressures below approximately 0.1 MPa the gaseous premixed flame collapses to the burning surface. Thus, the resulting fiame structure is seen as one-stage for most operating conditions (from 0.1 to 10 MPa). In the BOP model, the premixed monopropellant flame (between NH, and HCIO,) over the oxidizer surface takes place simultaneously to a primary diffusion flame confined at the edges of the oxidizer crystal. The main or final diffusion flame subsequently follows both flames between fuel and oxidizer intermediate products. ENDOTHERME ZEROTH ORDER PYROLYSIS OF SOUD FUEL AND EISSOCATIVE SUBLIMATION OF AP TO [AUMONGA AND PERCHLORIC ACID KUN RRR NN ed THIN PREWXED EXOTHERIC NEB + HOOK REACTION ZONE (PRROLYSED FUEL VAPOR ‘SERVESAS DLUENT) Figure 6. Schematic of Granular Diffusion Flame (GDF) Two-Stage Flame Structure for AP-based Composite Solid Propellants"® * 3a OF Fusen FLA Figure 7. Schematic of Beckstead-Derr Price (BDP) Multiple Flame Structure of AP-based Propellants' 23 Combustion Stability Propellant and/or combustor stability problems may be encountered during any experimental burn rate testing. Both are anomalies to be avoided. 2.3.1 _ Intrinsic Stability of Propellants This matter concerns the capability of a reacting solid propellant to recover its initial value of burning rate when perturbed. Its also known as intrinsic stability because It is stricly dependent on the nature of the ‘buming solid propellant and operating conditions (typically, but not exclusively, pressure and initial temperature). Following the pioneering work by Zeldovich® in 1942, two main approaches, known as Zeldovich-Novozhilov (ZN) method and Flame Modeling (FM) method, have emerged to study intrinsic stablity of solid propellants. Both share the basic assumptions of QuasiSteady gas phase, Homogeneous condensed phase, and One-Dimensional propellant strand (QSHOD framework). Within this framework and for pressure perturbations only, linear stability analyses were first presented by Denison and Baum ® in 1961 for premixed flames and Novozhilov * ®** in 1965 by the ZN method. Both works, in the linear approximation of the problem, relaxed the assumption of constant surface temperature until then used. The linear stability boundary so deduced is the same; this boundary was shown later to hold true even under nonlinear conditions Just on the stability boundary a reacting solid propellant is expected to reveal self-sustained oscillations of the burning rate. The combustion behavior beyond this stability boundary stil is a matter of speculation, but likely setf-sustained oscilatory buming are observed Until no steady solution whatsoever is allowed *” 2.3.2 Buming Stability of Motors Catastrophic, high frequency combustion instability became relatively uncommon with the advent of aluminized propellants in the late 1950's. As a result, related research and development dwindled. However, compelling need for reduced visibility or opacity of rocket exhaust has dictated elimination of significant concentrations of aluminum from many tactical rocket propellants. The instability problem has reemerged at great cost to many motor development programs. Burning stability concems the capability of the combustor to recover its initial configuration when perturbed. There are many ways of classifying instabilities, but usually the frequency of the oscillation Clearly reveals their source. The instabilty of concern to the one interested in measuring the burning rate of a propellant and motor configuration arises from coupling ofthe rate-determining combustion processes with acoustic oscilation modes of the combustion chamber. A rocket motor with a simple center- 4perforated grain of reasonably high length-to-diameter ratio acts acoustically as a closed cylinder. The simple acoustic modes of such a cylinder are illustrated in Figure 8. * Burning instabilities manifest as high frequency oscillations of the relevant motor variables (burning rate, pressure, temperature, etc.) superimposed on the corresponding average values, accompanied by corresponding vibrations of the motor case. Pressure histories of heavywall test motors containing an unstable, reduced smoke propellant (ie., non-aluminized HTPBIAP) often reveal pressure excursions far more than sufficient to rupture a flight weight motor case. Note that in actual industrial practice ‘combustion pressure excursions are not considered serious if below some limiting value (5%° or 2.5%") These fluctuations are always undesirable; even though catastrophic failures are not necessarily observed, failure of the mission often stil can occur. In general, vibrations are set up and transmitted to the whole propulsive system and vehicle, including payloads. Performances are modified due to shifting of the average chamber pressure, burning time is modified, and mechanical andlor thermal failures may ‘occur. Typically, bulk mode instability occurs in the low frequency range (up to 150 Hz), axial mode instability for combustor cavity lengths between 0.3 and 5 m occurs in a larger frequency range (100 to 2,000 Hz), transverse mode instability for combustor diameter between 0.01 and 1 m occurs in an even larger frequency range (500 to 50,000 Hz). Bulk mode instability throughout the combustor is associated with low values of L" typical of early buming in space motors where high mass fraction and low combustion pressure (below 20 atm) are usually met. it is a nonsteady mode of operation of rocket motors, involving growing low frequency oscilations possibly leading to a succession of quenching and reignition (also known as chuffing mode) without ever reaching a steady-state operation regime. This peculiar mode of operation is not common, it may last for the entire mission or naturally disappear in time due to growing values of L"*" “The high frequency range, whether axial or transversal, is much more common and usually associated with combustion details. "When oscilatory behavior occurs ... the oscillations would be more correctly attributed to instability of the entire combustor. The phenomenon results from a very complex interaction of the combustion, the combustor flowfield, and the combustor cavity walls." Burning instabilities take place when perturbations excite any of the many acoustic oscilation modes of the chamber cavity. This problem is strongly dependent on the details ofthe fluid dynamics (interaction of oscillations with the mean fiow, vorticity, viscosity, flow turning, multiphase flow, etc.) and 3D geometry of the combustion chamber (acoustics) as well as their interactions with the burning solid propellant. The balance of the various Contributions (amplifying or damping) is currently assessed by means of a linear analysis; but some nonlinear aspects are also discussed in literature. Although much progress has been made, this problem is far from being understood in its generality. Further comments can be found in References® #3. ; some practical stabilty problems are also discussed in References.” 152.4 — Burning Rate Measurement Methods 241 Test Devices Early descriptions of burning rate measurement methods were given in a previous AGARD publication by Young “ and several classical textbooks ’'° "***; more recent descriptions are reported elsewhere.*"2 In general the burning rates obtained by different techniques are not the same; even using identical specimens and the same technique at different facilites, the measured burning rates are different due to a variety of details not fully controllable or controlled. A host of methods, ranging from reduced scale rocket motors to the simple strand bumer, is today implemented to measure steady burning rates. In general, small scale motors are preferred to evaluate the burning rate of actual rocket motors (ballistic evaluation), while strand bumers are used for quick assessment tests or quality control of large propellant production; other methods are mainly used for special purposes (interrupted burning, high pressure combustion, etc.). ‘The experimental results from the various methods used today are in general accurate to within + 2-3% both for small motors and strand burners", but accuracy less than of 1% is sought for actual rocket motor design."* In most cases, strand burners reveal burning rates below the values of small motors, which in turn fall below full-scale motors. Nitramines seem to be a notable exception by scaling somewhat higher than full-scale motors.** Therefore, it is not only important to understand and standardize the technical procedures implemented by different users, but also to estimate the scale factor with respect to the full size engine if the burning rate value under the actual motor operating conditions is desired. In general, motor burning rate increases with the motor size; other possible differences between data collected in ‘motors and strands (changes in slopes or plateaus) are discussed in later sections. 24.1.1 Subscale Motors While the development of gun propellants relies on precise control of the manufacturing procedure and chemical compositions, this is impossible for rocket propellants due to the much more complex chemical composition and dependence on parameters difficult to control (particle size, minor additions of catalytic ingredients, etc.). Thus, the most satisfactory method to evaluate steady burning rates is to fire a certain number of rockets loaded with the actual propellant under test. For practical reasons, reduced cost, and Improved safety, different rocket motors of reduced size (typically, 2 to 6 inches - or about 5 to 15 cm - diameter) were specifically developed for ballistic evaluation purposes at many facilities. These reduced scale rocket motors are usually made with heavy case and fitted with nozzles of different sizes to provide ‘a number of convenient operating pressures (see Eq. 18). Typically, small ballistic evaluation motors are radial burners providing a neutral pressure trace in time (within 10%), 2 sharp tall-of, port area / throat area ratio AYA,> 6 and grain length / diameter ratio < 2 to minimize erosive burning, short burning duration (2-10 s) fo minimize heat losses and nozzle erosion, ‘small grain web thickness to minimize thermal shrinkage, conical nozzle geometry with 18° + 0.5° half ‘angle of divergence and no flow separation, The motor nozzle size, A, is estimated" from the burning rate ‘established from initial strand burner rate measurements based on the mass conservation equation Auk A a fehePe (18) Cop. In some cases, non-neutral pressure traces in time are used to reduce the number of tests, but the determination of the pressure exponent n is less accurate. In the industrial practice, over a pressure range for which @ and n remain constant (see Eq. 4), a minimum of seven motors at the nominal operating initial temperature and five at the expected initial temperature extremes are fired.*” Tests with small motors provide better correlation with full-scale motor burning rates, but are considerably more time- and money-consuming than tests in strand burners. Tests in small motors are normally performed only after the neighborhood of the final propellant formulation is reached, in order to obtain & ‘more accurate full-scale motor rate prediction and determine the temperature sensitivity of the motor ‘combustion pressure 7, 16Many specific configurations of subscale motors are used in different countries. The term micromotors or the aoronyms BEM for balistic evaluation motors, BTM for batch test motors, BCM for baich check motors, SSTM for subscale test motors and others are randomly found in the literature to identify this ‘specific but loosely defined class of motors. In this report, only the broad expression "subscale motors" is used. Although several suitable motor designs may be implemented, the most common configuration is a neutral burning grain providing a relatively constant combustion pressure p,. Detailed recommendations, ‘on current burning rate measurement test techniques and subscale test hardware for accurate prediction Of internal ballistics of a full-scale solid propellant motor are reported in JHU/CPIA CPTR 74.° Trends in ‘observed differences in calculated burning rate for the different analysis methods were also evaluated with ‘2 goal of making recommendations on preferred analysis methods. These results, including surveys of analysis methods and results of the round robins are reviewed in the following sections of this report and References "%. Further modifications, or complementary tools, of this basic setup are briefly described below. 2.4.1.2 Vented Vessels In the simplest version, vented vessels are actual rocket motors abruptly extinguished by sudden release of pressure by blowing off the nozzle or by waler injection. More sophisticated designs were also developed in which sticks, or slabs, of propellant are burnt and are quenched with water afier about half the sample has been consumed. The pressure of operation is controlled by a much larger tubular charge ‘of some faster burning standard propellant. By measurement of the burning time and the dimensions of the propellant sample before and after fring, the rate of burning can be determined directly.“* This method to obtain burning rates, which is laborious, is no longer practiced. But the technique of vented vessels is used still today for other purposes, such as interrupted burning to examine the conditions of the propellant ccharge during combustion.” 2.4.1.3 Closed Vessels Several closed vessel configurations are currently available to obtain the buming rate of the propellant from experimental pressure records in time. One option is to bum a small propellant sample in a large closed vessel filed with inert gas, producing a small pressure increase. The burning time is obtained as the time span between the onset and decay of the pressure rise. Another option is to increase the mass of the propellant sample up to a loading density of 0.3 gicm?, producing a very large pressure increase and pressurization rate from which the burning rate is deduced. This is not a direct measurement and the overall approach is a laborious process requiring a number of assumptions, but the method is used still today for very high pressure combustion (gun propeliants).” 157 ‘An alternative technique to assess performances of gun propellants in particular is to measure the so- called heat of explosion in some type of calorimeter. This is a sensitive and quick method, derived from chemistry, capable of detecting any important changes or gross error in chemical composition. But it is Useful in rocket propulsion only if, for the given propellant, the rate of burning is directly related to the heat of explosion, which is not commonly the case. 2.4.1.4 Strand Bumers For about 50 years, the industry standard apparatus for routine measurements of linear burning rates has been the so-called Crawford bomb proposed in 1947. This method, very quick, simple, and economic, is particularly suitable for exploring new propellant compositions or performing quality control of established ‘compositions. Strands of propellant having circular or square cross section, 3 to 6 mm in diameter or side, are employed. The overall strand length usually ranges anywhere from some 10 mm to about 150 mm. ‘These are supported in a suitable holder and inserted into a closed vessel, typically pressurized with nitrogen. The strands are coated with an inhibitor to prevent side buming. In the original configuration, two small holes are drilled, about 5 inches (about 127 mm) apart, along the diameter. Fuse wires are passed ‘through each hole and connected to terminals. The strand is ignited at the top by'a hot wire, and the time taken for burning to pass from the first to the second fuse wire is accurately measured. Itis usual to take several measurements at each pressure."® The burning surface should remain planar and normal to the strand axis, "7‘Over the years, several modifications of this basic setup for solid strands have been proposed. In the most ‘common modification, the whole apparatus can be placed in a thermally controlled environment capable of producing the desired initial temperature range. In another version, called window strand bumer, the burner is equipped with optical windows allowing optical recording of the buming processes (still Photography, movie camera, video camera, etc. both in the visible and infrared ranges). At ‘Thiokol/Huntsville, a bomb holding three strands was used.“* All configurations are easy and quick to operate, use a minor amount of propellant, and require litle instrumentation. Thus, the strand burner method is widely used. ‘Afurther modification was developed at Aerojet Solid Propulsion Company (and occasionally used also at ther locations), where the additional option of testing liquid strands of the uncured propellant instead of the familiar cured propellant solid strands is implemented. Burn rates of liquid strands are used in propellant manufacturing as controls for acceptance of the uncured propellant before casting into the ‘motor. The liquid strands are obtained by casting the uncured propellant into a proper vessel (6.4 mm diameter paper cup or plastic tube or 6.4 x 6.4 x 139.7 mm rounded solid strip) coated with an inert lacquer; see sketch in Figure 9.° The measured burning rate differs with respect to both the solid strands ‘and motor, but values can be correlated, as discussed more fully in later sections. The strand burning rate relationshoip is developed along with other control parameters during propellant development as ilustrated in Figure 10. Once the strand burning rate has been established with its tolerance limits, the motor burning rate can be predicted. Challenges and successes in this correlation are reviewed in Section 6.0. A representative comparison of liquid and solid strand data is given in Figure 11. The liquid strand bum rate is of importance once the propellant reaches a production level. The solid strand burn rate is confined mainly to the development stage, where the composition versus bum rate is being established. ‘An updated description and discussion of various strand burners is given in Reference *; further useful comments are reported in References "1.22, igure 9. Modern Crawford Bomb (Solid or Liqt Strands) 8BURNING RATE turning Rats, in/s Figure 10. Traditional Power Law Burning Rate Behavior’? 2.07 "are 3834.16 5.526.995 2.30 nazz 0.80 1.02 x 260 tn] wotor 3 toltstsr 2.30 ie 0.76 PERG 330 Zoo 500 S00 B00 SoS Pressure, psi Burning Rate vs Motor Size Figure 11. Behavior of Solid and Liquid Strand Burning Rate Relative to Motor Size’? ‘57u9 *oyey BuTuing2.4.2. Nonntrusive Methods ‘Several diagnostic techniques are used to deduce buming rates. The well-known fuse wire technique set up for strand burners"” allows only discrete measurements and under steady state burning. It is an intrusive method and thus suitable only for operating conditions far from the intrinsic stablity boundaries. Other diagnostic techniques, notably non-intrusive and continuous in time, are needed for motors. Nor-intrusive methods were developed with the aim of measuring buming rales while minimizing disturbance of the combustion processes independently of the experimental apparatus. Several techniques are available with a different features and degree of maturity: fim or video recording®, X- rays®, microwaves®', ultrasonic®, acoustic emissions®, radiation recoil (typically, but not necessarily by a laser source™), plasma capacitance. The X-rays technique, while applicable to any burning apparatus, is recommended for full-scale motors. Fiim or video recording, radiation recoil, and acoustic emissions fechniques are more suitable for strand burners. Microwaves and ultrasonic techniques are suitable for both small-scale motors and strand bumers. Radiation recoil and acoustic emission techniques do not detect the burning surface position and thus provide an indirect measurement; all other techniques are direct. Several of these techniques (in particular microwaves and ultrasonic) are also apt to measure transient burning rates; in addition, the acoustic emission technique is apt to provide information as to ‘the burning rate nonuniformity (due to localized and intermittent burning rate variations), 2.5 Burning Rate Measurements in Subscale Motors ‘When testing motors, the buming rate measurement is actually deduced from the observed pressure-time or thrust-time history. Thus, following Hessler ®, it is convenient to recognize from start that appropriate definitions are required for burning rate in motors. 2.51 Buming Rate Definitions ‘Two basic classes of empirical burning rate definitions are in use for motor applications. These two families of fundamentally different burning rate determination each have their advantages and disadvantages. One definition is based on propellant thickness and the burning time and is referred to as the thickness/time (TOT) method, and the second is based on the conservation of mass in the ballistic test ‘motor and is accordingly termed mass conservation or mass balance (MB) method, ‘The conventional burning rate definition is the fundamental TOT rate, ror ror ais = eb thickness < Wy a ‘burning time t, requiring the appropriate but elusive value of thickness besides that of the related time. Real world effects such as non-uniform web and non-instantaneous burnout make accurate measurements of burning rate > p., quasi-steady buming rate is evaluated as A, nate (28) ct ays Neglecting gas storage in the combustion chamber due to density change and/or volume change, a (average) mass balance rate ys may be written as =e GyPtt Way {Pelt tents [pdt te [pdt Several variations of Equation 27 have been used, primarily corrections for the neglect of mass storage, Which causes the main error (bias low) in use of the Mass Balance rate, definition. Note that fy ‘compensates for the inconsistency between w,-W and Way by letting fe= f °°” Ne @ ‘The burning rate for the mass balance method (rys) defined in Equation 26 is based on the balance in the mass flows developed during steady state burning. Due to a developing boundary layer in the nozzle, the occurrence of a “borda effect” for the case of a poorly manufactured nazzle, due to thermal expansion of a nozzle throat insert (tungsten or molybdenum) causing a contraction of the nozzle throat, or due metal ‘oxide deposition on the nozzle throat, the actual nozzle throat may be smaller than the geometric value measured before the test. To account for these effects the factor n, is included. Equation 26 requiresaccurate knowledge of the average combustion pressure (p.), the effective nozzle throat area (A,.n,), the propellant density (p,), the burning surface (A,), and of the characteristic velocity (c’). The effective nozzle throat area may be determined by measuring the geometrical nozzle throat diameter using an appropriate measurement procedure. The factor 7, may be based on experience, and may be different for each nozzle, propellant and combustion pressure and its choice may be ambiguous. This can produce ‘additional errors. The propellant density may be either measured or calculated from the propellant ‘composition, when 2 measured value is not available. The burning surface area may be calculated based ‘on accurate’ measured grain dimensions. The burning surface is constant for the case of an EB grain (coning not considered), but will generally change with time when a CP grain is used. The characteristic velocity is a thermodynamic property, which relates to the efficiency of the combustion and is essentially independent of the process-taking place in the nozzle. The theoretical characteristic velocity is fully determined by the ratio of specific heats, the molar mass of the combustion products and the equilibrium combustion temperature. Mass balance rate shown in Equation 27, differs from the thickness/time rate of Equation 21 by a correction factor. Mass balance rate has the reputation of being more rigorous, but also incorporates implicit assumptions. For example, it is assumed that negligible mass is stored in the combustion ‘chamber, and also that the propellant is burning all the time that pressure is nonzero. Mass Balance burning rate definitions contain the implicit assumption that the average buming surface during burning time is the same as during total time. Although this assumption is not correc, its effect is to approximately correct for early burnout because of misalignment or distortion, and accounts for the improved reproducibilty reported for the mass balance methods as compared to the thickness/time methods. Representative mass balance methods currently in use by facilities surveyed by WGO16 include: 1) Common Mass Balance Method 2), Vellacott’s Method 3) Brooks’ Improved Method 4) Jordan's Combined Mass Balance with Thickness/Time Method 3.2.1 Common Mass Balance Method ‘The Common Mass Balance method neglects all storage terms. In Equation 28 the mass balance buming rate (for an exact web) equals the mass balance buming rate (for a nominal web) multiplied by a correction based on pressure integrals for web burned during the burning time f, = (te- te), and fu, = (to — t,). Time definitions are discussed in Section 3.4 (See Figure 14 and Table 3). (28) Development of this equation begins with the fundamental rror, thicknessitime rate relationship in Equation 24. Using the average pressure defined as (29) where Z, is an average value. When rewriting c* and p, as Arf Pat 30) W, (30) and 26Pp cc) Amy Where Wp is the propellant weight used for the test. When replacing c* and i rate becomes the following Equation 29 the burning woe (32) ny 2 fra Brooks’ Common Mass Balance Method is also expressed by Equation 33, but with the end of bum (or {) defined by Equation 40 or Equation 9.1 of Appendix B-9. Alternatively integrating over the total action time t and using Equations 41 and 42 (or Equations 9.2 and 9.3 of Appendix B-9) for a, MP ; (33) a, [Par ne In essence, Equation 30 and 31 are identical to replacing the time in Equation 21 by the ratio of the pressure integral and the average pressure. When considering a typical subscale test with a neutral CP grain, Equation 21 gives the most accurate results from tests yielding the least amount of impulse in the tail-off. When assessing the influence of the {otal tail-off impulse on Equation 21, this wil become explicit. When taking for example two aft-tangent points, one at t, and another at i,+8, burning rate obtained with Equation 21 will differ by a fraction (1.+5)/. ‘The average pressure determined over two intervals in Equation 33 wil differ much less than this fraction (about half). Therefore, the burning rate as given in Equation 33 shows less variation. For the case when there is no tail-ff, the two methods yield identical results. Equation 21 becomes decreasingly accurate pulse in the tail-off ‘See Appendix B-25 for a summary of the common mass balance method, and Appendix B-9 for a more thorough discussion of the Brooks’ common mass balance method, as well as the Appendixes for many other facilities that use the fundamental mass balance method as summarized in Section 4.0. 3.22 Vellacott’s Method Its relatively easy to account for the volume increase (second term right-hand side of Equation 25), asi shown in Equation 22. This term results from the chamber volume freed by the buried propellant. a = pn 34) PeG = Pe 4) ‘The MB burn rate including mass storage due to volume increase is: (35) Note that the term between brackets does not relate to the actual volume change, but merely is determined by the ratio of the density of the combustion product, which is determined by the combustion pressure, and the propellant density. aTo assess the effect consider a conventional AP/HTPB propeliant with a solid loading of 85%. The correction due fo the volume increase can be calculated theoretically using a thermodynamic code, e.g. the NASA-Lewis code to determine the density of the combustion products as a function of pressure. The results for a pressure range of 2 - 10 MPa is given in Table 2. The addition of aluminum to the propeliant formulation does not substantially affect the correction factor. Table 2. Evaluating Effect Chamber Volume Increase ‘Combustion Density of Correction Factor Pressure Combustion X10" {MPa} Products Ikgim3] 2 247 4.28 4 4.49 2.55 6 6.27 3.82 8 8.35, 5.09. 40 40.42 6.36 When the volume change is assumed to be important, the following result is obtained, which is known as Vellacot's method or "equilibrium burning rate": fd pale 7 i, Jr 77 a, ‘An additional correction term is added to the rate equation neglecting storage terms given in Eqn. 28, ‘The bum rate correction due to a changing pressure is more difficult to account for because the pressure generally varies during the test. When the pressure increases, the contribution is positive while when the Pressure decreases the contribution is negative. The contribution is proportional to the free chamber volume; hence, when the free chamber volume is small and the pressure variation during an experiment is small as well, the third right-hand side term may be neglected. This should be evaluated for each smal- ‘scale burner. 3.23 Brooks’ Improved Method Brooks’ improved mass balance method approximates all storage terms on {Joa 14 2e—p) 1 © napa JO RT (7) Poke, Equation 37 equals the common mass balance rate of Equation 28, but includes an exact web thickness\time correction to the nominal thicknessitime ( w,/ , ), @ storage correction for density change due to the influences of pressure variation during the test, and a correction for volume change in the ‘chamber due to propellant consumption. See Appendix B-9 for a more thorough discussion of Brooks improved method.3.24 Jordan's Combined Method Jordan's combined method simultaneously solves for the mass balance and thicknessftime burning rates, ‘assuming only one burning rate properly defines the propellant, regardless of the method. The method ‘aloulates a thickness/time burning rate curve using the beginning and ending times. This burning rate ‘curve and the pressure integral fractions at a defined time are used to calculate a mass balance surface versus web burned curve (SW). This SW curve and time-defined fractional pressure integrals are iterated Until the caloulated SW curve matches the real grain geometry. This surface matching technique assures conservation of mass, and is relatively insensitive to ignition spikes and tail-off anomalies. Coefficients of variation from 0.09% to 0.23% were obtained using Jordan's Method on groups of 12 high quality (nominally replicate) motor firings. Frank Jordan was active in the past in complex solid propellant rocket motor firing analysis methods. The methods developed were implemented as computer codes for various different solid propellant rocket motor manufacturing companies. For example at Atlantic Research Corp., a data reduction procedure called Static Firing Analysis (SFA) was developed, at Talley Defense Systems, the procedure was called Talley Rocket Analysis Code (TRAC), while at Aerojet, a procedure called Aerojet Rocket Motor Analysis Code (ARMAC), was written for the Advanced Solid Rocket Motor (ASRM) project 7.” All these computer codes were essentially based on the same analysis procedure, while in addition, some codes had special capabilities that were unique to the company for which they were written. Jordan's method applies a web sliver correction to mass balance wy te te (38) to yield an instantaneous approximation and mean values of burning rate in an iterative procedure fn J PA, PpAre ‘The most expanded and complete version ofthe code could analyze any type of solid propellant rocket motor firing. While reportedly automated, successful implementation requires review of the data by an experienced analyst. See Appendix B-17 for a more thorough discussion of Jordan's method. See ‘Appendix B-11 for a description of Atlantic Research Corp.'s analysis method. #4, 39) 3.2.5 Iterated Mass Balance Method The iterated mass balance burning rate (Fye,) procedure differs from the common mass balance rate procedure (Appendix 8-25) in the use of the rate-averaged pressure pa, as the associated pressure and the iteration to determine exponent n. A variation on this method ‘involves an iterative process to determine the burning time while still using time-averaged pressure. Further details are provided in Appendix B-26. 3.3. Grain Web Thickness Definitions ‘The propellant grain configuration can be end burning (EB), circular centrally perforated (CCP), or may have a more exotic shape (e.g. a star shaped grain). It may be relatively easy to measure the web for an EB configuration, While for a CP configuration, the web may vary in the grain length and tangential directions, making an accurate measurement very complicated in practice. In all cases, the web measurement procedure should be well defined. The method used for the production of the propellant ‘samples largely affects the geometrical acouracy of the sample dimensions.3.3.1 Web Thickness Determination - End-Burning Grains Before the experiment to measure the burning rate of a particular propellant can be carried out, the propellant web has to be determined. A measured web thickness is preferred over a thickness taken from a drawing. Measuring the propellant web at three places at 120 degrees interval and arithmetically averaging the values may do this. In a practical situation the propellant will be cast into a plastic (e.g. rrylon) cylinder, and propellant discs are milled from this cylinder on a lathe. This yields relatively flat discs. ‘The nylon functions as an inhibitor. For most conventional composite rocket propellants good bonding between the nylon and the propellant may be obtained, however, for propellants that may yield poor bonding, the propellant may start burning between the inhibitor and the propellant. When this occurs, the ‘experimental results will deviate (may be observed from the pressure vs. time trace). The deviation will be larger for tests tuned to higher pressures. When the effect is small it may not be observed but will influence the results to some extend. ‘Another unexpected increase in pressure may be due fo a non-uniform regression of the grain surface, resulting in a burning rate enhancement along the inhibitor called “coning” "2", Explanations as to why on the propellant; 2. Migration of curing agent or other propellant ingredients (from within the propellant and/or from the inhibitor into the propellant) that may increase the burning rate, 3. A different propeliant composition near the inhibitor (e.g. a local concentration of fine particles), 4. Deviation from the 1D heat transfer situation due to a better heat conduction along the inhibitor. All these effects will change the burning rate locally and the overall pressure vs. time trace of the test; and through this the measured burning rate. 3.3.2. Web thickness Determination - Centrally Perforated Grains ‘Atypical web shape for a CP grain is given in Figure 13, URLLZZZZZLLRLLLLL LLL LLL LLL Propellant oe Figure 13. Typical Web Shape of CP Grain Inaccurate results will be obtained when mandrel or drawing dimensions are used to determine the web thickness’? as the changes that the propellant web undergoes between casting and curing are completely ignored. Shrinkage due to the curing process, thermal deformations (e.g. propeliant shrinkage during cool down) and deformations due to motor pressurization all occur and result in web changes that can result in burning rate errors of up to 5 %."* In the end, the number available to use for burning rate calculation is usually either: 1. Drawing dimensions, 2. Drawing dimensions corrected for theoreti 3. Ameasured “average” shrinkage, or deformation Note that each choice yields some kind of average thickness. The difference between the choices on the results can be considerable. Improved accuracy can be obtained by measuring grain bore and case inside diameters. Diameter measurements 1 — 2 web thickness into the grain bore will essentially eliminate ‘most of the web bias caused by taking measurements at the grain ends (compare Figure 13), but the 30thinnest point is usually near the center. Moreover, mandrel roundness, straighiness and alignment imperfections can result in up to 10 % web bias over drawing dimensions and result in inaccuracies for the burning rate of up to 3 %. When a grain is cast into phenolic sleeves or into a steel case with relatively thick liners on the inside surface, the grain outside diameter will be rather difficult to determine and simultaneously the web thickness, again enhancing inaccuracy. Methods for compensating for shrinkage, hardware variation, grain distortion and misalignment are offered by Hessler and Glick 3.3.3 Web Thickness Determination - Other Grains For all grains shapes the propellant burning situation is considered as one-dimensional (burning velocity vector pointing perpendicular inwards). This assumption is most realistic for the end-burning grains, it is less realistic for the centrally perforated (CP) grains, while it is even less realistic for the star shaped propellant grains. The more exotic grain shapes are not considered here in order not to further complicate the discussion. Also important for an accurate determination of the web are deviations from the desired ‘grain shape due to the propellant grain production process. CP grains generally have some taper due to the production process (mandrel taper). Taper normally is not applied to the mandrels of small grains. 3.4 Burning Time Definitions 3.4.1. Beginning and End Burning Times ‘The motor action time follows from the pressure versus time trace, which starts at the beginning of motor operation, and ends at the end of motor operation (see Figure. 14). The burning time is determined from this trace as the period from the moment that all propellant is considered burning till the moment the web is considered consumed. For burning rate determination, the burning time is the appropriate time period to use. The two periods define four moments in time on the pressure vs. time trace. These time points are often referred to by the use of diferent number or letter subscripts. Two such examples are presented in Table 3. Initia axiom Pressure or thrust Figure 14, Definitions of the Burning Times 3Table 3. Burning Time Definitions Time point Variant 1 Variant 2 Beginning of motor operation te te Beginning of burning t, te Ending of buming ty te Ending of motor operation t, te When Miler and Barrington® conducted their review in the late 1960's a number of methods existed for determining the beginning and ending burn times for solid propellant systems. These definitions, listed below in Table 4 for convenience, are illustrated in greater detallin Figure B-23 of Appendix B-22. Miller & Barrington indicated that no one definition seemed to have proved superior in all applications up to the late-1960s. Table 4. Burning Time Definitions Summarized in Miller & Barrington’s 1969 Review Surface Ignition Time Definitions (Identified on the Pressure Rise of Figure B-23) First pressure raise (point ©) The inverse tangent bisector (point A), A fixed pressure or a fixed percentage of the average or maximum pressure (point B), The initial inflection (point C), The forward tangent bisector (point D). Web Burnout Time Definitions (Identified on the Pressure Decay of Figure B-23) 4. The aft tangent bisector (point E), 2. The point to maximum rate of change of curvature during tailoff (point F), 3. A fixed pressure or fixed percentage of the average or maximum pressure (point G), 4. Point when pressure returns to zero (Point H). ePeNe In the US there exist a few basic references”*’5”, which for years have provided a basic set of reference burning time definitions. These references and their respective definitions are summarized in Table 5. Recent trends identified by the WGO16 will be compared with these earlier findings in Section 4.2 Table 5. Historical Sources of Characteristic Time Definitions in US Time References os CPIA Pub 1747 Mil Std 2920 Mil Std 292C, CPIA Pub 807° te Web Burnout Time Tangent Bisector CPIA Pub 80, CPIA Pub 174 0.95 | pat CPIA Pub 174 & ‘Action Time Terminus 0.05 P, CPIA Pub 174 0.10 Phase Mil Std 292C, CPIA Pub 80 100 psia CPIA Pub 80‘The following methods for defining burn time determination are discussed below, viz. 4) Constant P or %P 2) Tangent-Bisector 3) Brimhall 4) Brooks’ 5) Hessler-Glick ‘These well established methods (and variants) have been used to reduce different sets of pressure-time data within the context of evaluations using simulated motor (discussed in Section 5.0), and real motor (discussed in Section 6.0) behavior. 3.42 Constant P or %P Burning time, t, = (t,— t,) (Figure 14), may follow from the pressure-time trace by taking the time values at a certain pressure level or a constant percent of some characteristic measured pressure, such as P,.,. Typical %P,n.. Values used by facilities surveyed include 10%, 50%, 60% and 75%, with 10 % of the maximum pressure being used most often. While it is standard in many facilities, experience has shown ‘that 10% may be too soon to start burn time in some test cases. The P.,,, occurring during the test is generally the pressure due to the ignition transient when the combined action of the igniter and the burning Bropellant yield higher than normal buming pressures. The igniter size will have a considerable effect on Pman however, since the ignition transient is generally fast the effect on the action time may be substantial due to the tail-off. Constant P or %P definitions are often considered ad hoc, and will yield results biased to different degrees. However, burning times based on such definitions do give reproducible results when ‘the definitions are in the steepest portions of the pressure rise or decay. ‘When a CP propellant grain is used, the maximum pressure may occur somewhere during the experiment, the point which is determined by the L/D ratio of the particular grain. There are many variations to this relatively simple burning time definition, ie. a different pressure level may be chosen. It is clear that different choices yield diferent results for the burn rate, particularly due to the tai-off transient. Different results are also obtained for the burn rate vs. pressure relation. For example, if the locations of t, and ty move respectively to the left and the right, an apparent increase for the burning time results yielding a ower burn rate and a corresponding slightly lower average combustion pressure. This causes the burn rate vs. pressure relation to shift downwards, while possibly affecting the burning rate exponent (e.g. Vieile's relation, Equation 4) as well. Hence, effects occurring during the tall-off transient strongly affect the measurement, as illustrated further below by variations in the grain manufacture process. An advantage of this simple definition is its ease of implementation in software as part of a data reduction procedure. Unfortunately, the burning time does not begin at the start of web burning since a large portion of the ignition pressure rise is due to ignition material buming. Also the end of the action time does not coincide with the end of web consumption as the tall-off pressure trace includes contributions from stored ‘combustion chamber gases, from burning insulation and liner material and residual propellant sliver as a results of mandrel misalignment or non-ideal grain geometry. Motors can experience an effect, called “hump” or BARF (Burn Anomaly Rate Factor) effect in motor pressure-time behavior. This effect can create varying challenges for the burning rate analysis methods. This effect is the result of radial variations in burning rate caused by the theology of the g1 manufacturing process, e.g. casting the grain and subsequently plunging the mandrel. The racial variations of burning rate across the propellant web can influence burning rate bias by 4.5%.”"The “hump” cffect will affect the pressure trace, and in particular the value of P,., and hence, affecting any burning time definition relying on Pp, Casting with the mandrel in place can essentially eliminate the hump effect, as ilustrated in Figure 15, which shows typical pressure-time traces for two manufacturing methods. The influence of grain manufacturing on the scaling of burning rate is discussed more fully in JHUICPIA CPTR5C3-9 Plunge Cast 5C3-9 Cast with Mandrel in Place Figure 15. “Hump” Curves for 5-Inch CP Grain with 3-inch Bore, 9 inch Length 3.4.3. Tangent Bisector Method Defining burning time using the Tangent Bisector method (Figure 16) begins with the identification of the start of burning (typically t, is taken at the first 10% Png, point). The end of burning is determined in an effort to minimize the effect of the tailoff integral. This is important since the talloff integral is commonly {our to ten time larger than the ignition integral. However, the method neglects tailoff buming, which is the part of the web that continues buming after the anticipated end of bum. While viewed as arbitrary, also a historically “consistent” method of evaluating the end of bur. Q Assumes Ist point lit is Ist to burn out off Integral Pressure D Tail-off imegral 4-10 times ignition intergral © Does not account for sliver/web relation + Bum Time ——— ' os ® t Figure 16. Definitions of Tangent-Bisector Method ‘Actually, two definable points exist in the burnout process, intial and final Burnout. This process be; the vicinity of the knee of the pressure trace. The Tangent Bisector method really determines the inital bumout point, as do most end of burn time definitions. The method originated in the 1940s when recorders were slow and the knee of the pressure trace looked sharp. As faster recorders came into use and interest in rate accuracy increased, it became obvious the knee was not am sharp comer. As a result, several variants of the bisector method came into being including, bisector of the angle, bisector of the area between trace and tangents, and normal to the trace from the tangents intersection. The Tangent Bisector method usually defines a point in time that represents an arbitrary web burn out point and not the actual burn out point. In particular, non-neutral pressure histories are dificult to reduce due to the varying judgment of the analyst, and can cause up to a 1/2% bias in burning rate variations for the same motor.Different motors might indicate up to a 3% bias in rate because of different grain web shape The Tangent Bisector method may be more difficult to implement in software as part of a data reduction procedure than the constant pressure (or constant %P) definitions. However, the maximum intercept approximation of web bisector is relatively easy to program. 3.4.4 Brimhall Method Defining burning time using the Brimhall Method again begins with the identification of the start of burning (typically t, is taken at the first 10% Pye, point). The end of burning is determined by identifying an inflection point on the tail-of curve when the second derivative of the pressure, d°P_/at? = 0 or dP/dt is a positive maximum during tall-off. Typically, this point occurs after the time determined by the Tangent Bisector method. This defiition of burning time excludes consideration of the tailoff integral. Talloff burning with this definition causes problems simitar to those experienced using the Tangent Bisector method. Brimhall is an ad hoc definition, with more variation than Tangent Bisector if much variation in web thickness exists. For a given fixed motor manufacturing process with tight grain port alignment, itis ‘more repeatable than the Tangent Bisector. The Brimhall method may be easily implemented in software 2 part of a data reduction procedure for data that is not noisy. The method fails when pressure histories have multiple inflections or spikes in the tailof transient. Using a definition involving dP/dt = minimum may avoid this failure, 34.5 Brooks’ Method Defining burning time using the Brooks’ Method (Figure 17) again begins with the identification ofthe start of burning (typically ¢, is taken at the first 10% Pay, point). The end of burning is determined from the pressure integral as defined in Equation 40 below. The end of burning point (t,) is @ correction of the time point determined by the Tangent Bisector method (t). This point is corrected using the total pressure integral and a correction term c,. This definition of buming time does not consider the entire taloff integral. Talloff burning with this defiiion causes problems less than those experienced using either the constant pressure or the Tangent Bisector methods. @ Assumes Ist point lit is Ist to bun out Tail-off Integral Giver) Pressure © Tail-off integral 4-10 times ignition intergral © Does not account for sliver/web relation ———s ' my on ee ft 1——— Bum Time Figure 17. Definitions of the Brooks Method While the Improved Brooks Method accounts for stored chamber gases and some of the talloff pressure integral, the Common Brooks Method does not account for the stored gases. Both methods are based on the ratio of the total pressure integral to average pressure over the Tangent Bisector time. Both methods improve burn time calculations as the sensitivity to inaccuracy of the aft-tangent location is reduced.” 35b G@-H.e (40) [ra ‘The normal tail-off in a motor varies with pressure at burn out. Therefore, a, is available and is essentially linear function of pressure ”, and is defined by a=m.P, +b 1) ‘Where, Py is the pressure at web burn out, b is an empirical constant based on data. The slope mis a theoretical vaiue based on the empty chamber volume (V), the molecular mass of the gas (M), the gas constant (R), the flame temperature (7,) and the total propellant weight (W,) M RT.W, 83.10? Pa" and b = 0.97. The Brooks t, of Equation 40 approximates the = (ta* te)/2 discussed in the next section. The Brooks’ method is more fully (42) ‘Typical values are m = Hessler-Giick time of te, described in Appendix B°2. 3.4.6 Hessler-Glick Method For determination of the burning rate, the web thickness based on the design dimension, ay is used”®, ‘The reason is that ri, is accurately defined, even if it is biased slightly. ‘As to the burning time, the following definitions are suggested: ty= tebe (43) tas ter te (44) ‘where the relevant time points are defined as: ‘* Beginning of Burning t, : The midpoint of the time interval immediately preceding the first perceptible rise in dp/dt on the last sustained pressure rise to equilibrium motor operation. ‘* Initial Burnout f.; : The midpoint of the time interval immediately preceding the negative step to negative value of d?p/at? during the blow down period after the end of equilibrium motor operation. ‘+ Final Burnout ty: The midpoint of the time interval immediately preceding the positive step to Positive value of dp/dt? at or following the end of equilibrium motor operation. ‘The burning rate can than be defined using the two burnout times and the design web thickness, as Tor = Teg! (te - ts) (45) Tog = gl (ter ta) (46) ‘The reference rates for intial and final burnout represent two independent estimates of rate. They are averaged together to yield the HG burning rate +n (47) "Ho ‘The Hessler-Glick method is more fully described in Appendixes B-21 and B-27. 363.5 Average Pressure Definitions ‘The pressure usually associated with a measured burning rate is the time-averaged pressure [ipa (48) However, any measured point [fness: Plfmess)] Must simultaneously satisfy also the Vieille®® burning rate equation r = ap", which requires use of rate-averaged pressure™-*2.®° (‘ra PC meas) = Pas = (49) tp ‘The rate-averaged pressure Py is the pressure that should be associated with measured rates. For exponent m less than unity, rate-averaged pressure is less than time-averaged pressure. Consequently, use of time-averaged pressure p, resulls in rates corrected to reference pressure that are low. Both the rate correction to reference pressure, Equation 50 and the rate-averaged pressure Eqn. 49 require a value of exponent n. This is usually accomplished by performing a least-squares fi of the data to the Vieille rate equation, using the form In rps = Ena + 10 PA aes) 61) For methods using time-averaged pressure, the least-squares fit (Equation 51) is performed one time. For methods using rate-averaged pressure, it is necessary to solve Equations 49 and 51 simultaneously by iteration, if multiple motors with the same rate equation are available. An iteration beginning with the time- ‘averaged pressure as the starting point (which corresponds to an initial guess of n= 1) typically converges in three to five steps. ‘The average pressure for each bumout time for the HG buming rate method is the rate-averaged pressure, viz.: Pres = UF) = Uh Pt) / tal!” (62) Pret = (ad) = [Ue P°A)/ td (83) ‘Actually, to carry out the above analysis, an estimate of the pressure exponent is required to determine both rate-averaged pressure and for carrying out the correction to the reference pressure. The value for the pressure exponent may be obtained from historical data, or from a rapidly converging iterative procedure when at least two motors are available at the same intial temperature but different pressure. 373.6 Comparison of Analysis Methods Many rror methods do not explicitly account for non-instantaneous bumout. In fact, rror methods typically define end of burning as the knee of the curve (web burnout), when the experimental pressure trace begins to fall rapidly near the end of motor operation. However, specific choices of time points may make the correction implicitly. Methods that define end of buming near 50% pressure implicitly assume burning ‘continuation and thus partially avoid non-instantaneous burnout error, but not as well a5 an fy, definition that actually uses the integral ratio. Due to transient operations, these rror methods tend to behave essentially like rye methods. While use of 50% pressure time points for start of burning only has small effect on burning rate, the choice of 50% or more for end of burn during pressure decay can be a source of higher rate bias. A drawback of a 50-50 definition is that the time-averaged pressure differs much more from the rate-averaged pressure because the ending points are much lower down the tailoff curve than for ‘an equilibrium or web-knee definition. More bias in buming rate is introduced, as the rate-averaged pressure is seldom used. ‘ys Methods yield rates that are systematically low by a mass storage error. In turn, mass storage error also introduces a systematic nonlinearity in measured ri(p). Methods essentially behaving lke yp are likewise low by a mass storage error and generate similar nonlinearities. HG fully avoids the mass storage error, so it will be linear and systematically yield higher rates than nie methods. ror methods with instantaneous bumout also avoid the mass storage error yielding negligible nonlinear errors but high bias due to nor-instantaneous burnout. 4.0 INTERNATIONAL SURVEY OF ANALYSIS METHODS AND DEFINITIONS ‘The burning rate analysis methods and definitions used by the international solid propeltant rocket industry ‘are summarized and discussed in this section. Over 20 facilities from 7 NATO member couniries were surveyed to identify the methods and definitions used and their characteristics. This represents a complete survey of all the international facilities involved in burning rate measurement, and is a thorough representation of the fundamental methods used in the solid propulsion community today. Two conceptually different families of methods for burning rate determination exist, as indicated in Section 3.0. Each family incorporates a number of different definitions and methods to determine the burning rate of a solid propellant from a small-scale rocket motor firing. These analysis methods were examined using multiple round robins involving simulated motor data (Section 5.0) and real motor data (Section 6.0). Comparison of the methods and their relative accuracy are discussed. Small motor data quality, including uncertainty and error analysis, is reviewed in Section 7.0. ‘Subscale rocket motors and batch check motors are used in a number of countries to measure the burning rate of experimental propellants and commercially produced propellants. These motors differ in construction and means of operation. The reader is referred to JHU/CPIA CPTR 74° for an account of the different types used by the various faciliies and countries. 4.1 Review of Historical Surveys ‘Our understanding of analysis methods can be improved by first examining the trends in practices ‘employed within the solid propulsion industry. The examination of three historical surveys can contribute to this insight, one conducted by Miller and Barrington® in the late 1960s, a second conducted by Brooks and Hermsen* in the late 1970s, and the last by Fry* in the mid 1990's. When Miller and Barrington conducted their review in the late 1960's the fundamental thickness/time buming rate (ror) procedure for calculating solid propellant burning rates was the widely held practice. While web w, was readily measured, establishment off, involved the identification of surface ignition time and web bumout time on the pressure-time trace, as summarized previously in Table 4. and shown in ‘Appendix B-22, Figure B-19. Miller and Barrington reviewed several definions for f,, but indicated no one method seemed to have proven superior in all applications. Miller and Barrington indicated in their review that, based upon data provided by Brooks, using pressure integrals for determining the burning time may minimize motor-to-motor variation (e.g. due to sliver or nozzle erosion effects on tail-off). This was of ‘course the rationale for the original development of the mass balance burning rate (rug) methods for 38calculating solid propellant burning rates. Fry indicated trends in the U.S. suggest the fundamental TOT methods are being replaced by MB methods or TOT methods incorporating an improved method for ‘accounting for non-ideal tailoff in the burning time determination. Brooks and Hermsen** reported on a survey of U.S. facilities conducted in 1979 for the purpose of establishing standardized analysis methods. Fifteen of the 23 U.S. facilities contacted responded indicating both fundamental thicknessftime burning rate (Fror) and mass balance burning rate (rue) methods were in use for calculating solid propellant burning rates. All respondents except one indicated they used the aft tangent bisector as at least one method, and most used it as their principal method to define end of burn. A constant %Pnq, was the next most common method, followed by other methods such as maximum dPidt in the tailoff.. Web was measured by some and considered constant by some. Results showed more reliance was being made on the use of pressure integrals for determining the burning time. ‘The principle here was the calculated value of average pressure is more insensitive to the judgment of the analyst and the tailof. Two means of defining burning time based on average pressure ‘were in use 1. Normalized burning time (54) at Po (55) ‘This approach produces better results for good pressure-time traces, or when Jae =0 inthe tall 2. Nominal fraction of burning time (66) (7) This approach can accommodate cases where {Pat #0 in the talof. It also produces results identical to the first method for good pressure-time traces. 42° Current NATO WG016 Survey (ur understanding of analysis methods is further improved by examining the current practices employed by the over 20 facilities from 7 NATO member countries surveyed by the WG016. An overview of data analysis methods is given in Table 6. This table summarizes the source of the method by country and facility, the fundamental type and the location in Appendix B of a detailed description of the method. ‘A detailed examination of Appendix B provides insight into the methods and key definitions used by those surveyed, and are summarized in Table 7. This table summarizes the source of the definitions by country and faclly, the fundamental method type, and definitions for start burn time, end burn time and method used to determine it, how web or mass was defined and the averaging method used for the pressure integrals. Table 8 summarizes the frequency of use of the methods and definitions given in Table 6. The data show the European facilities universally employ the fundamental thicknessitime method, while the thicknessitime and mass balance methods are used almost equally in the U.S/Canada. Intemationally, the survey indicates a preference toward the thickness/time method. This is deceiving, as almost half of the European countries survey also use an end of burn time definition based upon the pressure integral in the tailof, thus mitigating inaccuracies that can typically accompany thicknessitime methods for non-ideal pressure-time tailoff behavior. Some facilities use multiple methods for different type of motors or as prescribed by the customer. ‘The vast majority of facilities surveyed define start of burn time by either a 39constant P or a constant %P, as shown in Table 8. Two reference methods from the U.S., Hessler & Glick and the various others surveyed by Miller & Barrington use slightly more rigorous definitions such as those involving dP/dt or forward tangent bisector in the pressure rise. Some facilities use more than one definition. On the other hand, a wider variety of definitions are in practice for defining the end of burn time. Sixteen of the 25 facilities surveyed use a more rigorous definition for the end of burn time, involving the pressure integral, a simple Brimhall d?P/at® (NAWCWD), or a more complex use of d°Pidt” (Hessler-Glick), As such, these facilities are seeking to account of inaccuracies typically associated with non-ideal pressure-time tailoff behavior. Another 8 of the 25 facilities surveyed use an end of burn time defined bya standard tangent bisector procedure or tangent bisector corrected by @ constant close to 1. The remaining 3 facilities simply use a constant %P to determine the end of burn time. A majority of the facilities, therefore, have determined greater rigor should be exercised in determining the end of burn time. Generally all facities use a constant web value, either measured or taken from a drawing. Finally, the vast majority of the facilities use time-averaged, rather than rate-averaged pressure. The advent of faster more capable computers should make rate-averaged pressure a more attractive means of reducing rate bias up to an additional 14%.‘Table 6. Summary of Data Analysis Methods (Refer to Appendix B) COUNTRY EACILITY ‘METHOD TYPE" CANADA DREV Rror FRANCE SNPE J ONERA Rom GERMANY BAYERN-CHEMIE Rror ITALY FIAT AVIO Room NETHERLANDS TNO-PML Rror UNITED KINGDOM RORM Rror UNITED STATES AEROJET Ror Poo AFRLIPLE Rror ALLIANT TECH SYSTEMS Re AMCOM Re ATLANTIC RESEARCH CORP Rugs Re BF GOODRICH / UNIV PROP Rror GD/ ORDNANCE & TACT SYS. Rror NAWCWD CHINA LAKE Ror Pow Nsw Ree Paw/csD Re ‘SNAP / JORDAN Roan STONE ENGINEERING Roar TALLEY DEFENSE SYS Rrot THIOKOL PROPULSION Ron Poe REFERENCE HESSLERIGLICK (HG) Ree MILLER & BARRINGTON re FUNDAMENTAL METHODS 1) Thickness/Time Rate (Rror) 2) erated Thickness/Time Rate (Rrors) 3) Mass Balance Rate (Ry) 4) erated Mass Balance Rate (Rye,) 5) terated Two-Point Thickness/Time Rate (Rix) 7 Methods Defined in Reference Category, Fundamental Methods a APPENDIX B Bt B2 SaRE B7 BS B10 Bat B42 B43 Bas B45 B46 B47 B48 B49 B21ey sojsesie-veGueL aL nouing ye ounssoig "Ped « Poa cou EER IBY BAT RATT coon araReA | ~PROToMRRNS TTT PaBTaNe OEY Ten WEUED SGT Eo CORDSTTL ara | rah Ped 7 OND /wsSsaH | __SONEUaT paberane ou, | eae e000 er er | Sanne nor | 7 a oberon oU em EVD ca er | Saves exon] IW dO TORO paberone-oun, | esau Weis eL, er SO bed 35Ns50 ITIVE paberoneouL ‘S5eu0 TueISu0D ‘Sui Uang 180 jeiBajuy oan eie TaPON TE HOF =r NIONG NOLS, Pibemne oa | “aren UEIe ay ce ao Boj omeserg 1a POH SO 7 VGUOT7 eVNS- Pebemnrou | ss8iijuRaue] | eRe UTERUS > exBanursineserd | yoneueD >ia | a KGL eon | Pa Bie IW OSOT Med Poe Re HLS SAL a 70n9BOTH aATCTE Ta oH “0h 7 ‘OMEN Fabeane aL ng 7009 BOI INET Te POA S09) ba] a faba ‘on was — | pore of RUN = re TOUTE ee S00) z TA ONORAWA eb eanr aL aon WBE ca #1 a Kor 7 AT HORIOOOS a BE) Ba wa TONG BOTT ORT Ta Pon 0 7 za a anane-sia | ~ sear URED Ta POA S08 7 TOY a 1a =a Hor wr Woon peters oui | seen wee LPH a KO Eine |e peberene-oun | seeun weave e1pon | aon wr TW SAS OSL INGTTWW paberane oun Ton Tueis06 09 | a 00 7 Har wav PESTS gam WeieTeD er =F %0b zi snowy] _saivis aaLiNn PESO on eH = SOF i WRG | WOGONDH GALINT POBEIORE uN om EIS FL POUPON LPR Tange 7 ww OBEN SUL am TOD Tg 0 9 HITS asaor_| a or 7a TNTAGONT | _SONVTSHISN aaa TAT 3 ST SG oro ABH SABRI 7908 "0 r Onv Iv 7 PabeNT AL aon TRIO E10 % BLO 08 Fa 108, a SINSHO-NESAVE RWIS PaBeAe OU, ‘769 OAD ‘uy wang 76n0 jeABo}U] eanssaIg "a HOS 7a 05 ary Bans pabeene oa 2a SRR 3 cng 79R0 ABSI ITER aH Tae a VeaNO Bonves pebasene oan aon Tueisve Tura WF POTTED cL aed ean ZO a 7380 vawnve BO a a einssoig 80m) 49m WE PU Tame wes —] aon.an SNOUT SISATYNY aurova uuNn09 suopuyed pue spoweyy sishreuy evey Bujuing yo Aeumung +2 o1qeL‘Table 8. Summary of Frequency of Use of Analysis Methods and Burn Time Definitions ‘ANALYSIS METHOD. Geographic Region Thickness/Time (Fron) ‘Mass Balance (rus) European 7 0 US. / Canada 8 10 Total 15 10 "BURNING START TIME DEFINITION ‘Geographic Region Constant %P Constant P| Other 10% | 50% | 60% | 75% dPidt | Various European 2,4/- 7 US. / Ganada 13 1 1 Total 6 | 6 1 1 Total 27 2 ‘BURNING END TIME DEFINITION Geographic | Constant | T-5 | T-BCorrected | Pressure ‘Other Region “PF by Intogral Constant | PI (0 __| Brimhant | @?P/dt | Various European 2 2 4 : 3 z : E US. / Canada 4 5 : 2 8 1 1 1 Total 3 7 1 2 1 1 1 1 Total 4 16 435.0 ASSESSMENT OF METHODS USING SIMULATED MOTOR DATA During the working period of RTO/AVT WG016 assessments were made of the various thickness/time and ‘mass balance analysis methods taken from the international survey. The objectives of these assessments ‘were to: ‘+ Clarify distinctions of small motor analysis methods © Identity sources of the differences ‘The assessment approach involved using both simulated and real motor data as described earlier in Section 1.3. Table 9 summarizes the scope of these assessments and the degree of involvement by the volunteer participants. The assessment using simulated motor data involved carrying out four Round Robin evaluations as described and discussed in this section. The assessment using real motor data is discussed in Section 6.0. The Round Robin analyses were carried out with the objective of comparing the results of the various analysis methods as they are commonly used in Canada, France, Germany, Italy, ‘The Netherlands, UK and the U.S. and the various organizations in these countries. Each Round Robin was essentially carried out as @ blind experiment, where individual participants were given minimal knowledge of the motor geometry and no information on the propellant-bumning rate behavior. Simulation Programs were used, or if not available developed, to generate the Round Robin pressure versus time ata. It was argued that an accurate knowledge of the input burning rate data was useful for comparison with the results obtained from the data reduction, ‘The discussion of each Round Robin evaluations is organized beginning with a description of its design, followed by a presentation of the analysis results, and closing with conclusions. Design details of each Round Robin are provided in Appendix C. Typical results are discussed relative to reference burning Tates Fr; and rp (burning rates at respectively 7 and 10 MPa (50 and 70 psi), the burning rate coefficient @ ‘and exponent n of the power law relation, the reduced burning rate data (used to obtain power law relation) and their retation to the input reference values used to generate the Round Robin data. ‘The design and execution of the Round Robin #1 and #2 simulations were found to embody properties that placed limitations on the conclusions that could eventually be drawn from the results. These issues are reviewed individually withthe discussion of each Round Robin. As a result two final Round Robins #3 ‘and #3X were designed and executed to overcome these shortcomings.‘Table 9. Burning Rate Analysis Round Robin Summary _ COUNTRY FACILITY! ANALYSIS | __ ROUND ROBINS PARTIOPATION. METHOD’ | RR#1 | RR#2 RR #3 RR #3x, Groupst2 | Groups 3:10 CAR Dey : z FRANCE one n ra SNE ts wie te CERUANT SAVER STEM i ZT Tay TAO MEM i Eager Pr ao OUI SAVER x 7 z HESBLER/GLCK co: 7 z BAYERN-CHEMIE i Exper Pana as SNE ta Experimental mot eas OLN Wss Banos [ta Ecperienial Ps motor ex FIAT AVIO M8 MD z Experimental mot eas HESSLERIGUGK r Expriental Pmolr aa NETHERLANDS [ROPHEME ta zee M2 r= whe wT oan z TONE M = Eigen Pano aa BC a Experimental mot xs HESSLEWGICR om Experimental Pot xs TRITED eNGDOW | ROR — C217 TRTED STATES | ERO n z ALLIANT TEGH SYS MA ie 7 z AICO ian ? ACM oo 7 BF GOOSRGHUP r 7 NAWED M2 i zo NSE i ze PAW [65D i to SNAP / JORDAN ie ze STONE ENGIN a 7 TALLEY DEFENSE a THIOROL PROP = 7 Z REFERENCE HESSLER / GLICK tr ¥ [vit v L v 7 WGO16 member facilities ONERA, SNPE, BAYERN-CHEMIE, POLIMI, TNO-DML, and RORM. ® Analysis methods described in Appendix B, and Round Robin Designs described in Appendix C ® Participation on volunteer basis, ¥ completed Round Robin *° Politecnico di Milano, Dipartimento di Energetica graduate students analyzed RR#3, RR#3X and Real motor data ‘1 TNO-PML graduate students analyzed RR#3X data using TNO M2 method "= TNO-PML graduate students analyzed Real Motor data using selected methods 455.4 International Round Robin #1 Results Round Robin (RR) #1 is based on GasGen Version 2.0 ('93). GasGen is a spreadsheet program based on 2 relatively straightforward simulation logic developed by WG016 member Dr. Robert Frederick at UAH, US. Propellant burning rate and other propellant data, and propellant geometry data form the input. Detaiis of RR#1 design and input data are summarized in Appendix C-1. A simple endbuming grain as shown in Figure C-1 of Appendix C was simulated. RR#1 consists of four pressure-time traces for the cases listed in Table 10. Table 10. RR#1 Small Motor Ballistic Simulations Examined Group Cases Perturbation 1 1 Neutral - Baseline 2 Progressive 3 Regressive 4 Progressive with noise added Pressure-time behavior for these cases is given in Figure 18. — mete) Figure 18, Pressure versus Time Traces for the Round Robin #1 Cases ‘The first three cases have similar pressurization rates while the fourth case has slower pressurization, Moreover, the noise added to case 4 had peak-to-peak amplitude of about 5%. Routine measurements of different batch check motors revealed a typical noise level ranging from 0.1 through 0.4% of the motor pressure, which is an order less. When due attention is given to the experimental setup through instrument shielding and grounding provisions, a level of 0.05% (two orders less) of motor pressure is achievable ®, 5.11 Round Robin #1 Analysis The RR #1 data was supplied to a number of U.S., French, German, UK and Netherlands companies/organizations. Each participant was asked to analyze the data using their preferred method{s) and determine the average burning rate and average pressure. Analysis results for RR #1 data were received from the following eleven volunteer participants as shown in Table 9, including SNPE, ONERA, Bayern Chemie (BC), TNO-PML, Aerojet, Aliant Tech Systems, (formerly Hercules), Hessler-Glick (HG), F, Jordan! SNAP, NSWC, Talley Defense Systems, and Thiokol. The methods used to analyze the data 46and determine the (average) propeliant burning rates were mainly thickness/time (TOT) methods, with @ Alliant Tech, Jordan and NSWC using mass balance (MB) methods as summarized in Table 9. A typical set of results using the HG method for all cases is summarized in Table 11.° Table 12 compares the correlations of the data as obiained using the commonly employed power law relation for the RR #1 participants. The differences in results are clearly shown in Figure 19 where the burning rate at 7 MPa (rz) and 10 MPa (r,9) relative to the reference (or input) rate are plotted for each organization. Table 11. Typical Burning Rate Analysis Results NATO RTO/AVT WG016 RR #1™ HG Method RR#1 Cases 4 2 3 4 Burn Time, § 5.142 | 5.002 | 5290 | 5.783 Burning Rate, mm/s 19.6255 | 20.1748 | 19.0764 | 17.4502 Iterated Exponent 0.58661 ‘Average Pressure, MPa___| 9.91154] 10.3656 | 9.4599 | 6.1172 Reference Rate at 10 MPa 19.7195 Reference Rate, mm/s 79.7146 | 19.7389 [ 19.6951 | 19.7295 Deviation 0.025% | +0.098% | -0.124% | +0.051% ‘Table 12. RR #1 - Power Law Relations Obtained from Correlating the Received Data [County Company |___ Coefficient nent Burning Rate a a = ™ Tn my ny a an ae Tw Tos ja fa Ta B ma] [Reference e808 | 0.000 | 0.600 | 0.000-| 15.600 | G.000 | 19430 | 0.000, [France - SNPE Bis | 5540 | 0583 | 2767 | 16.030 | ater | 19740 | 1595 TONERA 49593 | 1.608 | 0.600 | 0.000 | 15.940 | 1503 | 19.740 | 1505 [Germany - BC 32178 | 6.905 | 0578 | 3650 | 16.070 | 2422 | 19.750 | 1647 (Netherlands -TNOMT] 52781 | 6911 | 0576 | 4083 | 15990 | 1912 | 10630 | 1029 TINOM2 | 51603 | 5.911 | 0582 | 3017 | 16.040 | 2231 | 19740 | 1595 US -Aeroet ‘55208 | 13207 | 0559 | 6883 | 16.400 | 4525 | 20.020 | 3037 Alliant Tech 50540 | 3567 | 052 | 1417 | 15.980 | 1848 | 19.790 | 1544 ~Hessior ‘E1108 | 4.708 | 0587 | 2233 | 16.000 | 1976 | 19.720 | 1493 TdordanSNAP | 4.1378 | 15.223 | 0.680 | 13333 | 15.540 | 0.956 | 19810 | 1966 TRSWE: 53063 | 8.718 | 0569 | 6.150 | 16.060 | 2358 | 19670 | 1295 TTalley DS aers6 | 4245 | 0624 | 3.950 | 15.790 | 0255 | 19.650 | 1.102 Figure 19 and Table 12 indicate the burning rate is generally over-predicted by 2-4.5%, with the majority of the participants showing a 1-2% deviation. ‘The power law exponent, shown in Figure 20, is typically under-predicted by 1-5%, with a couple outiers. While the power law coefficient, shown in Figure 21, is typically over predicted b'1-7%. The differences. between the 7 MPa and the 10 MPa relative results are due to these observed differences in the power law exponent and coefficient. 47-1} |mrt0irt0 ref 5.000 4.000 22286 8 (4) s0u3 ovey Bwana enneren Countries - Facitties Figure 19. RR#1 - Relative Burning Rate Error (%) at 7 MPa and 10 MPa Reference Pressures r] dvNsiueHor sn sa howe sn |wuano- aa eouevejo s g EI le] 15,000 2 9 1%) ujuouedsa uo 10103 eAneIoN 10.000 Countries - Facilities Figure 20. RR#1 — Relative Exponent Error (%) at 7 MPa and 10 MPa Reference Pressures15,000 5.000 10.000 | —_____— 0.000 5.000 10.000 15.000 ‘expanded section of Figure 22. Countries - Facilites Figure 21. RR#1 - Relative Coefficient Error (%) at 7 MPa and 10 MPa Reference Pressures {All burning rate data from the 4 cases are plotted in Figure 22 on a linear pressure scale, instead of the ‘more common log-log scale to enhance the visbbly of differences among the resus. igure 23 shows an Pres Pa) Figure 22, RR #1 - Burning Rate versus Pressure Behavior for all 4 Cases Compared with Reference 49ma g 5 reser a} Figure 23. RR #1 - Burning Rate versus Pressure Behavior for Case 1 ‘Compared with Reference (Expanded Scale) ‘The 4 cases are grouped from left to right in Figure 22 corresponding to the average pressures shown in Table 11, viz: case 4, case 3, case 1 and case 2. Most of the data is grouped together for each case showing small deviations to the reference (line visible inthe figure). Each case shows a typical distribution Of the result as obtained from the different organizations. The overall trend is in line with the reference data, although the deviation differs from case to case. In general the absolute value of the difference steadily inoreases from the neutral, to the progressive, to the regressive burning cases. Deviations in the ccaloulated average pressure data are also observed in the range of +0.2 MPa. Small differences in results between Bayern Chemie, SNPE and ONERA are due to similarities in their analysis methods. The small deviations observed are due to slight differences in the implementation method employed. 5.1.2 Round Robin #1 Conclusions. The design and execution of the round robin simulations were found to embody properties that place limitations on the conclusions that may eventually be drawn from the overall RR#1 results: 1) First, the data covered a limited pressure range (8.1 ~ 10.4 MPa). Due to this, conclusions on deviations among the burning rate relations in power law form may be of limited value (Table 12) as effects of errors may be exaggerated 2) By design, burnout in the round robin simulations is instantaneous. This steep decay is normally not observed in real motors and is an artifact of the spreadsheet programming used to generate the data. In reality, the blow-down phase (part of tal-off) exhibits an exponential decaying pressure towards ambient pressure. This exponential decay is modified by heat losses and by the transition from sonic to subsonic flow. This artificially steep decay causes several definitions, ‘for end of burning to yield precisely the same answer. Consequently, conclusions on the relative ‘merits of the end of burning definitions are unwarranted. 3) In execution, an error was introduced in pressure during blowdown after burnout. This causes errors in the pressure integrals, upon which various Mass Balance definitions of burning rate rely. This error will also carry into the time calculations used by any method relying on the 50determination of a total pressure integral. Consequently, conclusions on the relative merits of ‘Thickness/Time and Mass Balance burning rate definitions are unwarranted. 4) By design, one of the simulations included 0.5 MPa of simulated instrumentation noise. This Unrealistic noise level (12 to 50 times more than recent data, 100 times larger than good instrumentation) undoubtedly produces artifical burning rate errors in this particular simulation. This particular simulation is pivotal in pressure exponent calculations because of its relative isolation in rate-pressure space, and error due to the noise will unduly influence the calculated ‘exponent. Consequently, conclusions on the relative merits of time-averaged and rate-averaged pressures in burning rate correlations may be compromised for Case 4. 8) Presentation of the raw simulation data results in a minor (0.01%) ambiguity in web thickness burned. Table 13, RR #1 - Variation of Exponent and Reference Rate with Mean Pressure Definition and Burn Time Definition™ Analysis using Time-Averaged Pressure Bum Time Ref Rate Definition Exponent __at 10 MPa Pimin> Web 0.60088 19.6702 (0% > Web 0.61608 19.6695 10% > Web 0.59337 19.7086 50% > Web 055207 19.82397 Analysis using Rate-Averaged Pressure ‘Bum Time Ref Rate Definition Exponent __at 10 MPa. Pmin>Web 0.58661 19.7194, 0% > Web 058951 = 19.71893 10% > Web 058337 19.73279 50% > Web 054536 19.8056. Despite these limitations however, some conclusions may be drawn regarding beginning of burning rate

You might also like

• Solid Rocket Propulsion Fundamentals: by Robert Geisler, January 2003

  

  No ratings yet

  Solid Rocket Propulsion Fundamentals: by Robert Geisler, January 2003

  15 pages
• Launch Methodology

  

  No ratings yet

  Launch Methodology

  5 pages
• Propulsion System Modeling Guide

  

  No ratings yet

  Propulsion System Modeling Guide

  18 pages
• OUFTI-1 Nanosatellite

  

  No ratings yet

  OUFTI-1 Nanosatellite

  127 pages
• Carbon Carbon Nozzle Erosion and Shape Change Effects in Full Scale Solid Rocket

  

  No ratings yet

  Carbon Carbon Nozzle Erosion and Shape Change Effects in Full Scale Solid Rocket

  11 pages
• Grossi Et Al 2022 Numerical Simulation of Multiphase Flows in Solid Rocket Motors Nozzles

  

  No ratings yet

  Grossi Et Al 2022 Numerical Simulation of Multiphase Flows in Solid Rocket Motors Nozzles

  10 pages
• Practical APCP Motor Design For Amateur & High Power Rocketry

  

  No ratings yet

  Practical APCP Motor Design For Amateur & High Power Rocketry

  50 pages
• Optimal Nozzle Design Using MOC

  

  No ratings yet

  Optimal Nozzle Design Using MOC

  2 pages
• Single Engine Propeller Aircraft Design

  

  No ratings yet

  Single Engine Propeller Aircraft Design

  69 pages
• Design Criteria For Film Cooling For Small Liquid-Propellant Rocket Engines

  

  No ratings yet

  Design Criteria For Film Cooling For Small Liquid-Propellant Rocket Engines

  6 pages
• M

  

  No ratings yet

  M

  72 pages
• A Fist Step Towards The Definition of A Semireusable Launch Syst 2004

  

  No ratings yet

  A Fist Step Towards The Definition of A Semireusable Launch Syst 2004

  10 pages
• Guidance and Control of Artillery Projectile

  

  No ratings yet

  Guidance and Control of Artillery Projectile

  8 pages
• Motore p80 75

  

  No ratings yet

  Motore p80 75

  4 pages
• Vega PDF

  

  No ratings yet

  Vega PDF

  7 pages
• Aerodynamic Coefficients

  

  No ratings yet

  Aerodynamic Coefficients

  18 pages
• High Angle of Attack Aerodynamic Predictions Using Missile Datcom

  

  No ratings yet

  High Angle of Attack Aerodynamic Predictions Using Missile Datcom

  19 pages
• Ballistic Missile Trajectory Prediction

  

  No ratings yet

  Ballistic Missile Trajectory Prediction

  16 pages
• Ounding Ocketry EAM: Critical Design Review

  

  No ratings yet

  Ounding Ocketry EAM: Critical Design Review

  104 pages
• Rocket Propulsion Efficiency

  

  No ratings yet

  Rocket Propulsion Efficiency

  15 pages
• SRM - Xls S R M: Performance

  

  No ratings yet

  SRM - Xls S R M: Performance

  29 pages
• Vega Program Status

  

  No ratings yet

  Vega Program Status

  40 pages
• Green Propulsion Innovations in Aerospace

  

  No ratings yet

  Green Propulsion Innovations in Aerospace

  10 pages
• Design Complete

  

  No ratings yet

  Design Complete

  257 pages
• Green Rocket

  

  No ratings yet

  Green Rocket

  6 pages
• Nozzles: Narayanan Komerath 1

  

  No ratings yet

  Nozzles: Narayanan Komerath 1

  34 pages
• Forces Acting On A Missile While Passing Through Atmosphere PDF

  

  100% (1)

  Forces Acting On A Missile While Passing Through Atmosphere PDF

  2 pages
• Fast Burning Rocket Propellants Based On Silicone Binders New

  

  No ratings yet

  Fast Burning Rocket Propellants Based On Silicone Binders New

  7 pages
• Continuous Detonation Engine and Effects of Different - 2010 - Chinese Journal PDF

  

  No ratings yet

  Continuous Detonation Engine and Effects of Different - 2010 - Chinese Journal PDF

  6 pages
• Prediction of High Velocity Solid Propellant Gun Performance by Gas Dynamic Computer Program

  

  No ratings yet

  Prediction of High Velocity Solid Propellant Gun Performance by Gas Dynamic Computer Program

  15 pages
• PDF

  

  No ratings yet

  PDF

  98 pages
• Hybrid Rocket Paper 01

  

  No ratings yet

  Hybrid Rocket Paper 01

  14 pages
• Rocket Nozzle Design & Analysis

  

  No ratings yet

  Rocket Nozzle Design & Analysis

  18 pages
• Aerodynamics of Projectile With Wrap Around Fins

  

  No ratings yet

  Aerodynamics of Projectile With Wrap Around Fins

  63 pages
• MPMM Ballistic Model for NATO STANAG 4119

  

  No ratings yet

  MPMM Ballistic Model for NATO STANAG 4119

  28 pages
• StuttgartUni - The HyEnD Stern Hybrid Sounding Rocket Project

  

  No ratings yet

  StuttgartUni - The HyEnD Stern Hybrid Sounding Rocket Project

  40 pages
• Design of Axis Symmetric Nozzle

  

  No ratings yet

  Design of Axis Symmetric Nozzle

  127 pages
• Design and Validation of Swept Back Wing

  

  No ratings yet

  Design and Validation of Swept Back Wing

  100 pages
• Ada 800118

  

  No ratings yet

  Ada 800118

  214 pages
• Coning Motion Stability of Wrap Around Fin Rockets

  

  No ratings yet

  Coning Motion Stability of Wrap Around Fin Rockets

  8 pages
• Ijm&sv12n1 12

  

  No ratings yet

  Ijm&sv12n1 12

  14 pages
• Evaluating The Effectiveness of A Guided Projectile: Georgia Institute of Technology/GTRI/ATAS, Atlanta, GA 30332-0844

  

  No ratings yet

  Evaluating The Effectiveness of A Guided Projectile: Georgia Institute of Technology/GTRI/ATAS, Atlanta, GA 30332-0844

  25 pages
• A Computational Tool For Predicting Hybrid Rocket Motor Performance

  

  100% (2)

  A Computational Tool For Predicting Hybrid Rocket Motor Performance

  10 pages
• AIAA-1998-3377-716 Magnus Effects On Stability of Wraparound-Finned Missiles

  

  No ratings yet

  AIAA-1998-3377-716 Magnus Effects On Stability of Wraparound-Finned Missiles

  6 pages
• Rocket Engines Explained

  

  No ratings yet

  Rocket Engines Explained

  30 pages
• Rocket Science Basics

  

  No ratings yet

  Rocket Science Basics

  46 pages
• Time-To-go Polynomial Guidance Laws

  

  No ratings yet

  Time-To-go Polynomial Guidance Laws

  5 pages
• (AMCP 706-283) - Engineering Design Handbook - Ballistic Missile Series - Aerodynamics - U.S. Army Materiel Command

  

  No ratings yet

  (AMCP 706-283) - Engineering Design Handbook - Ballistic Missile Series - Aerodynamics - U.S. Army Materiel Command

  98 pages
• Small Solid Propellant Launch Vehicle Mixed Design Optimization Approach

  

  No ratings yet

  Small Solid Propellant Launch Vehicle Mixed Design Optimization Approach

  10 pages
• Dynamics Re Entry

  

  No ratings yet

  Dynamics Re Entry

  176 pages
• Two Stage Launch Vehicle Trajectory Modeling For Low Earth Orbit Applications

  

  No ratings yet

  Two Stage Launch Vehicle Trajectory Modeling For Low Earth Orbit Applications

  8 pages
• Aerodynamic Design, Characterization and Flight Performance of RLV-TD

  

  No ratings yet

  Aerodynamic Design, Characterization and Flight Performance of RLV-TD

  16 pages
• Nozzle Design and CFD Analysis

  

  No ratings yet

  Nozzle Design and CFD Analysis

  8 pages
• Ada 399211

  

  No ratings yet

  Ada 399211

  256 pages
• ADA405711

  

  No ratings yet

  ADA405711

  22 pages
• 1 s2.0 S2667134423000573 Main

  

  No ratings yet

  1 s2.0 S2667134423000573 Main

  20 pages
• MP-091-34-Evaluation of Methods For Solid Propellan Burning Rate Measurement

  

  No ratings yet

  MP-091-34-Evaluation of Methods For Solid Propellan Burning Rate Measurement

  26 pages
• Bertacin 2013

  

  No ratings yet

  Bertacin 2013

  14 pages
• Setback and Spin For Artillery, Mortar, Recoilless Rifle, and Tank Ammunition by Leo D. Heppner

  

  No ratings yet

  Setback and Spin For Artillery, Mortar, Recoilless Rifle, and Tank Ammunition by Leo D. Heppner

  576 pages
• Burning Rate Measurements From Small-Scale Test Motors: L.T. Deluca, F. Maggi

  

  No ratings yet

  Burning Rate Measurements From Small-Scale Test Motors: L.T. Deluca, F. Maggi

  17 pages