THE PROPAGATION OF ZIRCONIUM RUNAWAY OXIDATION OF SPENT FUEL CLADDING FOLLOWING LOSS OF WATER DURING STORAGE by Nicola Anthony Pisano B.S. University of Notre Dame, Notre Dame (1980) Submitted to the Department of Nuclear Engineering in Partial Fulfillment of the Requirements of the Degree of MASTER OF SCIENCE IN NUCLEAR ENGINEERING at the MASSACHUSETTS INSTITUTE OF TECHNOLOGY August 1982 ©) Massachusetts Institute of Technology 1982 Signature of Author Department of Nuclear Engineering August 1, 1982 Certified by_ Thesis Supervisor Certified by I Thesis-Reader K-) 'Acc e p y Chairman, De # JAN I W'o'nmittee on Graduate Students 16i " Archives THE PROPAGATION OF ZIRCONIUM RUNAWAY OXIDATION OF SPENT FUEL CLADDING FOLLOWING LOSS OF WATER DURING STORAGE by Nicola Anthony Pisano Submitted to the Department of Nuclear Engineering on August 1, 1982 in partial fulfillment of the requirements for the Degree of Master of Science in Nuclear Engineering ABSTRACT A phenomenological model of kinetic and diffusion controlled reaction is used in the modelling of the energetic oxidation of zirconium cladding at elevated temperatures in air at one atmosphere. Employing models developed in NUREG/ CR0649, a numerical analysis of the propagation of vigorous oxidation in a spent fuel storage pool following hypothetical drainage is performed. Energy transfer via conventional heat transfer mechanisms, levitation and convection of particulate, and combustion of zirconium vapor is addressed. Numerical simulation of data obtained at Sandia National Laboratories is performed. The results of the pool-wide analysis indicate that propagation of runaway zirconium oxidation from recently discharged to older fuel assemblies is highly dependent on the fuel storage configuration and minimum spent fuel decay time. It is shown that the minimum decay time to prevent carryover from spent fuel stored adjacent to recently discharged fuel (\90 day decay time) is approximately 2 years. Propagation via zirconium particulate is found to be a minor effect. It is concluded that the generation of zirconium vapor is highly unlikely, and that at worst it could only intensify the oxidation reaction within a given fuel holder, without leading to widespread burning. Numerical simulation of experimental data yields unsatisfactory results, for a variety of reasons related to both computer modelling and experiment conditions. Thesis Supervisor: Dr. Frederick Best Title: Post Doctoral Research Associate Department of Nuclear Engineering DEDICATION To the memory of my father NICOLA PISANO, 4 June 1921 - SR. 6 July 1982 ACKNOWLEDGEMENTS The author wishes to thank Dr. Allen Benjamin of Sandia National Laboratories who acted as sponsor and coordinator for the research performed in this report, Dr. K. T. Stalker of Sandia National Laboratories for his assistance in pro- viding the experimental data of Chapter 5, Dr. I-Wei Chen and Dr. John Meyer for their assistance in various areas of computer modelling, Dr. Frederick Best for his indefatigable patience and guidance in this research, my wife, Raisa for her assistance in preparing the illustrations in this report, and Wynter Snow for a timely and expert typing service. 5 TABLE OF CONTENTS page ABSTRACT......... 2 DEDICATION....... 3 ACKNOWLEDGEMENTS. 4 TABLE OF CONTENTS 5 LIST OF FIGURES.. 8 10 LIST OF TABLES... CHAPTER 1. INTRODUCTION..... 1.1 Preface............. 1.2 Background.......... Problem Statement... Summary of Contents. CHAPTER 2. POOL-WID E PROPAGA TION OF ZI RCONIUM BURNING. ......... 21 21 2.1 Description of SFUEL Program. 26 Zircaloy Clad Oxidation....... Extension of SFUEL Program to Assess 0 Pool-Wide Burning............. ....... .. 0.....0 0... 2.3.1 Inter-Holder Conduction. 2.3.2 Revised Fluid Dynamics Calculations...... 2.3.3 Maximum Driving Force Approximations ..... 2.3.4 Revision of Clad Thickness Model......... 2.3.5 Limit to Maximum Rate of Change of Temperature.............................. TABLE OF CONTENTS (Continued) page 2.3.6 Fuel Relocation Models................... 2.3.7 Computer Operating System Changes........ 2.4 Modified SFUEL Analysis of Burning Propagation.................................... 2.4.1 Characteristics of Revised SFUEL Code.... 2.4.2 Pool-Wide Propagation Results............ 2.5 Propagation of Zirconium Burning Via Conventional Heat Transfer Mechanisms: Summary........................................ CHAPTER 3. PROPAGATION OF RUNAWAY OXIDATION VIA PARTICULATE TRANSPORT....................... 83 3.1 Introduction................................... 83 3.2 Particulate Propagation Potential Algorithm.... 85 3.2.1 PARTICLE Model Description............... 85 3.2.2 Analytical Evaluation of PARTICLE Results.................................. 93 3.3 Energy Transfer Via Burning Particulate: Results........................................ 97 3.4 Propagation Via Particulate: Summary........... 101 CHAPTE 4. ZIRCONIUM VAPOR ANALYSIS.................... 105 4. Mechanisms for Formulation of Zirconium Vapor.......................................... 105 4.2 Quantitative Analysis of Vapor Generation...... 110 4.3 Zirconium Vapor Analysis: Summary.............. 112 TABLE OF CONTENTS (Continued-3) page CHAPTER 5. COMPUTER SIMULATION OF SANDIA EXPERIMENTS................................. 113 5.1 Introduction................................... 113 5.2 Sandia Experiments on Zirconium Burning........ 113 5.2.1 Experimental Configuration............... 113 5.2.2 Experimental Data: Air Test #4........... 115 5.3 Computer Simulation of Experimental Tests...... 121 5.3.1 Introduction............................. 121 5.3.2 Experimental Simulation: CLAD............ 121 5.4 Experiment Simulation: Results................. 126 5.5 Experiment Simulation: Summary................. 126 CHAPTER 6. PROPAGATION OF ZIRCONIUM BURNING: CONCLUSIONS................................. 130 APPENDIX A. HEAT TRANSFER COEFFICIENTS AND SKIN FRICTION CORRELATIONS USED IN SFUEL........ 133 APPENDIX B. SFUELlW INPUT, OUTPUT AND PROGRAM LISTING.................................... 136 APPENDIX C. PARTICLE INPUT, OUTPUT AND PROGRAM LISTING................................... . 173 APPENDIX D. THE VAPOR PROGRAM, INPUT, OUTPUT AND PROGRAM LISTING........................ 182 APPENDIX D. CLAD INPUT, OUTPUT AND PROGRAM LISTING..... 196 REFERENCES............................................. 217 LIST OF FIGURES page CHAPTER 1 1.1 Typical Design Characteristics of PWR Fuel Analyzed in this Report............................ 17 CHAPTER 2 2.1 Elevation View of Spent Fuel Pool Showing Modes of Heat Transfer Addressed in SFUEL........... 22 2.2 Spent Fuel Storage Racks Analyzed in this Report............................................ 24 2.3 Correlations for Zirconium Oxidation in Air....... 28 2.4 Cubic Plot of Oxidation of Zirconium -1.5% Tin Alloy at 600* -800C........................... 30 2.5 Cubic Plot of Oxidation of Zirconium -1.5% ...........................31 Tin Alloy at 8250 -900C 2.6 Finite Difference Mesh for One-Dimensional Fluid Flow.......................................... 41 Maximum Clad Temperature for Original and Revised SFUEL.................................. ..... 57 Comparison of Original and Revised Codes during Runaway Oxidation............................ 61 Peak Clad Temperature During Runaway.............. 63 2.10 Comparison of Peak Clad Temperature, Melt Options 1 and 4..................................... 64 2.11 Peak Clad Temperatures in Pool Sections 6 .................... and 5, Melt Option 4........... 67 2.12 Peak Clad Temperatures in Pool Sections 6 ....... ....... and 5, Melt Option 1......... 68 2.7 2.8 2.9 ... 2.13 Temperature in Pool Section 6, Axial Location 16........................................... LIST OF FIGURES (Continued) page 2.14 Temperature in Pool Section 5, Axial Location 16........................................... 2.15 Temperature in Pool Section 4, Axial Location 16........................................... 2.16 Temperature in Pool Section 3, Axial Location 16........................................... 2.17 Composite View of Pool-Wide Propagation Results........................................... 2.18 Pool-Wide Propagation Sensitivity, Case 3......... 2.19 Results of Propagation Sensitivity Analysis, High Density Storage.............................. CHAPTER 3 102 3.1 Spatial Variation of Effective Heat Flux.......... 3.2 Temperature Rise due to (q/A)f.................. 103 CHAPTER 4 4.1 Postulated Mechanism of Zirconium Vapor Generation........................................ 109 CHAPTER 5 5.1 Configuration of Zirconium Burning Apparatus...... 114 5.2 Experimental Data for Zircaloy-2 Oxidation in Air............................................ 117 Experimental Data for Zircaloy-2 Oxidation in Air............................................ 118 5.4 Experiment Simulation Model....................... 123 5.5 Comparison of Pre-Oxidation Heat-Up of Experimental Assembly............................. 127 5.3 LIST OF TABLES page CHAPTER 1 1-1 Decay Heat Generation Rates versus Time from Discharge.................................... 18 CHAPTER 2 2-1 2-2 Oscillatory Instability of Interholder Space Interface Temperatures...................... 39 PWR Spent Fuel 17 x 17 Array, Cylindrical Storage Rack Configuration, Full Discharge Loading, 3" Baseplate Hole........................ 58 2-3 Steady-State Nitrogen Mass Flow Rates............... 59 2-4 Comparison of SFUEL Results Under Runaway Conditions........................................ 62 2-5 Maximum Clad Node Temperature for Different Fuel Relocation Options (Modified SFUEL)............ 66 2-6 Decay Heats Used in Assessing Pool-Wide Propagation....................................... 77 Assessment of Sensitivity of Pool-Wide Propagation....................................... 79 2-7 CHAPTER 3 3-1 Terminal Velocity of Settling Particulates.......... 91 3-2 Particle Size Frequency Distribution................ 95 3-3 Input Values to the PARTICLE Program................. 98 3-4 Particle Distributions for Ramp in Figure 2.3..... 99 3-5 Effective Energy Fluxes to Horizontal Structures........................................ 100 3-6 Thermophysical Properties of 5% Cr-Steel........... 100 11 LIST OF TABLES (Continued) page CHAPTER 4 4-1 Thermophysical Properties of Zirconium and its Oxide......................................... 107 CHAPTER 5 5-1 5-2 Experimental Data for Zircaloy Oxidation in Air............................................ 119 Comparison of Pre-Oxidation Assembly Heat-Up...... 128 12 CHAPTER 1 INTRODUCTION 1.1 PREFACE The majority of commerical power reactors currently in operation were designed with the expectation of a viable fuel reprocessing system being available. In consonance with this fuel life-cycle, spent fuel storage pools were designed to accommodate the amount of spent fuel expected to be awaiting shipment to a reprocessing facility. The prohibition on spent fuel reprocessing in this country has resulted in the accumulation of large inventories of spent Light Water Reactor (LWR) fuel. While the fate of fuel re- processing is as yet undecided, the need for increased storage capacity both on-site and away-from-reactor has continued to grow. The interim solution of this dilemma has been a re-design of the spent fuel pools into longer-term storage facilities. This has been achieved primarily by re-designing the fuel storage racks, thereby allowing for denser packing. Public acceptance of high density storage arrangements and licensing of interim on-site fuel storage has been difficult to achieve. Demands have been made to assure that the modified storage arrangements do not increase the health hazard to on-site workers or the community. At hearings conducted to determine whether the reactor licensees should be granted permission to modify their spent fuel storage pools to accommodate higher density storage arrangements, a common concern is in regard to the potential risk to the public if the water were inadvertently drained from the pool. It is known that, at elevated temperatures, zirconiumthe major constituent of LWR cladding-undergoes vigorous exothermic oxidation upon exposure to air. One scenario advanced by citizen-advocate groups contends that a loss of pool water could lead to a widespread burning of the zirconium cladding, eventually engulfing the entire pool area. The radioactive fission product release from the degraded fuel rods would then be quite large. The purpose of the research described in this report is to examine, refine, and extend existing analyses of this topic and to investigate aspects of the problem not previously addressed. 1.2 BACKGROUND The investigation by A. S. Benjamin, et al., (Bl), NUREG/CR0649, indicates that a self-sustaining clad oxidation reaction could initiate if a high density storage configuration were to lose its coolant, provided that some fuel had been recently discharged from the reactor. The analysis addresses various pool racking arrangements, fuel types, and fuel pool building characteristics. The report indicates which storage configurations are likely to result in clad temperatures susceptible to vigorous exothermic oxidation. Several recommendations are made relating to stor- age rack design and minimum cooling times required prior to fuel storage in the various racking configurations. However, the analysis does not predict whether a self-sustaining zirconium cladding oxidation reaction, once initiated, will either extinguish within a localized region, or propagate to other sections of the fuel pool. The computer code developed to analyze the spent fuel behavior up to the initiation of runaway oxidation, SFUEL, has as its parameters of primary importance the fuel-cladding temperature and extent of Zircaloy oxidation. The code is based on phenomenological models for zirconium clad oxidation. In particular, the transient non-isothermal burning of zirconium in gas of variable oxygen mass fraction is modelled by a parabolic oxidation kinetics formula derived from isothermal constant oxygen mass fraction experimental data. A search of the current literature reveals that there is a dearth of information on zirconium oxidation in a dry air environment, and all of that which exists is obtained under isothermal conditions. Additionally, it is to be noted that the Sandia study is the only one available in the open literature which addresses numerical simulation of large-scale zirconium oxidation in a dry air environment. 15 1.3 PROBLEM STATEMENT The scenario for burning propagation of zirconium clad- ding, and the associated processes investigated in this report, may be outlined as follows: 1) At the initiation of the accident, the spent fuel pool is instantaneously drained of coolant. 2) The resulting heat-up of the fuel rods,due to the reduced heat transfer in the absence of liquid coolant,initiates a natural convection circulation of air throughout the fuel pool. 3a) For low density storage arrangements, the natural convection cooling is sufficient to allow the fuel rods to achieve a steady-state temperature below the runaway oxidation value. 3b) For certain high density storage arrangements, the natural circulation is insufficient to achieve a steady-state temperature below the value for vigorous oxidation ( %800*C). 4) The oxidation reaction energy leads to even more vigorous oxidation, potentially resulting in local clad melting, vaporization and spallation with subsequent levitation of particulate. The natural convection heat-up of the fuel is in the study mentioned earlier. However, analyzed the SFUEL code was limited by numeric stability,and other problems, to calculations prior to the initiation of runaway oxidation. The research described in this report details steps taken to extend the code to analyze the propagation of the oxidation reaction throughout the fuel pool. Propagation in this analysis occurs by conventional heat transfer mechanisms: conduction, convection, and radiation. In addition, two other propagation mechanisms, involving simultaneous mass transfer as cited in (4) above are examined. In performing these analyses, a host of phenomena occurring concurrently in the fuel pool and impacting on the propagation potential needed to be examined. Typical design characteristics for fuel analyzed in this report are shown in Figure 1.1, while the decay heats for discharged fuel are given in Table 1-1. Experiments have been performed at Sandia National Laboratories to observe the behavior of heated Zircaloy-2 clad exposed to air in a forced convection environment. Concur- rently, a computer code employing the oxidation models in SFUEL has been developed to allow analytical prediction of the experimental results. 1.4 SUMMARY OF CONTENTS As outlined in the Problem Statement, the SFUEL code has been revised and extended to analyze pool wide propagation of zirconium runaway oxidation. This objective neces- sitated the refinement of many of the SFUEL models, as well 176.8" FUEL ROD PWR ASSEMBLY GRID Figure 1.1 Typical Design Characteristics of PWR Fuel Analyzed in this Report (after NUREG/CRO649). Design Properties of Fuel Assemblies Used in the Analysis Older PWR Newer PWR 15 x 15 17 x 17 Number of fuel rods per assembly Numer of non-fuel rods per assembly 208 264 17 25 Active fuel height (in.) Rod center-to-center pitch (in.) 144 144 0.558 0.496 0.420 0.026 0.374 0.023 0.456 0.461 177 193 Rod array Fuel rod outside diameter (in.) Clad thickness (in.) Channel thickness (in.) Metric tons uranium per assembly (MTU) Number of assemblies per core, typical reactors 18 TABLE l-l* Thermal Decay Power of PWR Spent Fuel as a Function of Decay Time and Discharge Cycle Standard Case: 3 cycles @ 3.3% wt. enrichment. Operating power = 37.3 MW/MTU. Total Burnup = 33,000 MWD/MTU. 30-day down time between cycles. 35-day down time within cycles. 295 operating days per cycle. Decay Power, KW/MTU Cycle 1 Cycle 2 Cycle 3 10 days 75.4 81.1 86.6 30 days 43.4 48.6 53.2 90 days 21.3 26.0 30.0 180 days 11.4 15.6 19.2 Decay Time 11.0 1 year 5.05 8.20 2 years 2.22 4.07 5.90 3 years 1.25 2.46 3.76 5 years 0.607 1.30 2.12 10 years 0.368 0.779 1.28 *After NUREG/CRO649. as incorporating several new analytic capabilities. Fea- tures of the current SFUEL code which were examined and/or revised were: 1) mass flow calculations 2) oxidation kinetics models 3) all modes of heat transfer 4) methods of fuel rod and fluid temperature calculation The revised code was validated for low temperature performance by comparison with the original SFUEL results. The SFUEL program, revisions and results are explicated in Chapter 2. The spallation and levitation of burning zirconium particulate entails a two-part analysis. The first portion in- volves the numerical solution of the equations for particle temperature and transport. The second portion employs a semi-infinite body conduction model to assess the ability of the burning particle to ignite a receiving surface. The model equations and evaluation algorithm are described in Chapter 3. The particulate model requires a number of assumptions, yet for these cases, quantitative as well as qualitative information can be obtained. However, for the zirconium vapor- ization analysis, there are too many unknowns to allow a quantitative assessment at this time. Rather, the zirconium vaporization analysis, described in Chapter 4, gives the conservative estimates for the intensity of propagation of vigorous zirconium burning via this mechanism relative to those discussed above. The fifth chapter of this report concerns the simulation of the experiments conducted at Sandia. The computer code is described and predictions are compared to the experimental findings. Discrepancies are explored and ramifica- tions are discussed. The final chapter of this report summarizes the significant results of the previous chapters. The relationships of the various propagating mechanisms are discussed and the results superimposed to give a global view of the pool-wide behavior. The relevant observations obtained from the ex- periment numerical simulation are applied, and an overall summary is given for the propagation of zirconium runaway oxidation in a drained spent fuel pool. CHAPTER 2 POOL-WIDE PROPAGATION OF ZIRCONIUM BURNING 2.1 DESCRIPTION OF SFUEL PROGRAM The SFUEL computer code developed at Sandia is a two- dimensional model of a spent fuel pool. The model corresponds to the elevation view of the fuel pool, as shown in Fig. 2.1. The code models six adjacent pool sections, where the first section is bounded by the pool liner and the sixth section corresponds to one-half of the center pool section. The primary assumptions embodied in the pool analysis are as follows: 1) The water drains instantaneously, leaving the pool completely devoid of water. 2) The geometry of the fuel assemblies and racks remains undistorted. 3) Temperature variations across the fuel rods are neglected, and all rods in a particular pool section have the same axial temperature distribution. 4) The air flow patterns are one-dimensional and involve a Boussinesq approximation. 5) Radiation view factors are based on projected areas. All radiating surfaces are gray bodies. 6) All decay heat emanates from the fuel rods. Oxida- tion of fuel pool structural materials is not addressed. HOLDER CHANNEL (IF PRESENT) qdec chem - DECAY HEAT INPUT TO FUEL RODS - CHEMICAL OXIDATION HEAT INPUT TO FUEL RODS - AXIAL CONDUCTION IN FUEL RODS - RADIATION FROM OUTERMOST HOLDER TO SIDEWALL LINER CONVECTION FROM AIR STREAM 3 TO qa I 3 1 SIDEWALL LINER qlwW - CONDUCTION INTO CONCRETE SIDEWALL gh RADIATION FROM RODS TO CHANNEL WALL - CONVECTION FROM RODS TO AIR STREAM 1 ach - CONVECT ION FROM AIR STREAM 1 TO CHANNEL WALL RADIATION FROM CHANNEL WALL TO HOLDER qca - CONVECTION FROM CHANNEL WALL TO AIR STREAM 2 CONVECTION FROM AIR STREAM 2 TO HOLDER ghh . RADIATION FROM HOLDER TO ADJACENT HOLDER qha - CONVECTION FROM HOLDER TO AIR STREAM 3 qra I qa 1c a2 qa h'3 ty 1 -RADIATION FROM UPPER TIE PLATE I - RADIATION FROM LOWER TIE PLATE 2 2 TO FLOOR LINER CONVECTION FROM LOWER TIE qta2a4 PLATE TO AIR STREAM 4 CONVECTION FROM AIR STREAM 4 q12 TO FLOOR LINER CONDUCTION INTO CONCRETE FLOOR ci2 - CONVECTION FROM AIR STREAM 3 TO ADJACENT HOLDER Figure 2.1 Elevation View of Spent Fuel Pool Identifying Heat Transfer Modes Considered in the SFUEL Program (after NUREG/CRO649). 7) The spent fuel is arranged such as to have the hottest elements in the center pool section, and the cooler elements progressively toward the ends of the pool. 8) Axial heat conduction is negligible for channel walls, holder walls and liner. Axial conduction is consid- ered for the fuel rods only. 9) The spaces between adjacent racks are closed to air flow; heat transmission through these regions is by conduction through stagnant air. As observed in NUREG/CR0649, the assumption that temperature variations within a pool section are negligible (3) is found to be adequate, since the heatup time is on the order of hours. When vigorous oxidation is in progress, the as- sumption may still be assumed to hold on an assembly-byassembly basis within a given pool section. This topic is discussed later in this chapter. The assumption that the inter-holder spaces are closed to air flow is a conservative representation of the fact that the air flow in these spaces is retarded by support structures. Typical spent fuel holder designs analyzed in this report are shown in Fig. 2.2. The philosophy adopted in this research is to investigate the propagation of burning in configurations where it is thought most likely to occur and not duplicate the results of the earlier (NUREG/CR0649) analysis. For similar reasons, the analyses performed here are carried out under the assumption that the fuel pool building ventilation rate remains 12.75" HIGH DENSITY CYLINDRICAL SQUARE 0.25" SS 12.5"1 8.4" 8.4"1 0.13" SS 0.25" Ss 5 3. 5 OPEN O31 HIGH DENSITY Figure 2.2 9.5 8.4" CYLINDRICAL SQUARE Spent Fuel Storage Racks Analyzed in this Report 25 constant throughout the duration of the accident, such that the room air conditions above the pool remain at the ambient those external to the fuel pool building. conditions, i.e., As suggested in Assumption (4) above, the air flowing through the heated channels is treated using a Boussinesq approximation. In essence, the air is a thermally-expandable but incompressible flow. The fluid dynamics calculations are thus performed in a steady-state fashion, where the compressibility effects in the energy equation and the mass accumulation term in the continuity term are neglected. Thus, the rigorous set of fluid conservation equations (R1), Eqn. 2.1. 1 Dp + 2.la p Dt + VV = 0 PC DT (Pvx) =-7(pvxV) [ = V(kVT) D + [y ( - VP + V 6V 6 . DP + q + TSP 5v P[+ [2 2.lb + p 2 6v + 6v 5v )~ 6v x ( 6v+ + + 6 + X l 2.lc and similar equations of the form 2.lc in the other two coordinate directions are reduced to the one-dimensional flow equations (2.2): 6u 6x 0 2.2a ST 6ST - w Tt f) 2.2b = pg + orifice and friction pressure loss pg+6r±.c 2.2c P~ cpupcpgg+ qwhere q= hA(Tw -T)22 In equations 2.2 the molecular effects of heat and momentum transfer are described by phenomenological engineering models. The SFUEL Program is formulated as a semi-explicit code. In particular the fluid dynamics calculations are performed in an implicit fashion, employing a Newton-Raphson iterative method to obtain consistent values of temperatures and mass flows. All other parameters in the code are calculated in a fully explicit manner. The requirement for an explicit re- presentation of the fuel clad temperature arises from the highly non-linear character of the zirconium oxidation reaction model. 2.2 ZIRCALOY CLAD OXIDATION Oxidation of zirconium in air at elevated temperatures occurs by the following highly energetic exothermic reaction: 02 + Zr -* ZrO 2 liberating approximately 262 kcal per mole of Zr. The oxida- tion reaction may be either kinetically controlled or diffusively controlled. The kinetic-controlled reaction is modelled using a parabolic reaction rate law: 27 2w where dw = KO exp (-E/RT) 2.3 2 w = weight gain per unit surface area (mgO2 /cm t = time (seconds) Ea = activation energy (cal/mole) R = gas constant (*K) KO = reaction rate constant ([mg/cm2 2/s) T = temperature (*K) The parabolic kinetic model is derived from experimental data obtained under isothermal, constant free stream oxygen mass fraction conditions. Thus, the parabolic formulation inherently incorporates these assumptions. As suggested earlier, the actual processes are neither isothermal nor occur with constant oxygen mass fraction. The empirically derived coefficients for Eqn.2.3 are: KO=1.15x103 (mgO2 /cm2 ) 2/s Ea= 2 7 4 3 0 (cal/mole) (T<9200 C) K 0 =5.76x103 (mgO2 /cm2 ) 2/s Ea =52990 (cal/mole) (920 0 C<T11550 c) K 0 =6.20x103 (mgO2 /cm2 ) 2/s Ea =29077 (cal/mole) (T>1155 0 C) This data is presented graphically in Fig. 2.3. The data of Hayes and Roberson (Hl) and White (Wl) is obtained for pure zirconium oxidation in moist and dry air, respectively. The data of Lestikow (Ll) is obtained from oxidizing heated Zircaloy-4 tubes in dry air. There is little data available +6 +4 6. 20 x 10 4 exp (-29077/RT) +2 U 0 C) 0 5. 76 x 107 exp (-52990/RT) -2 E MONO-TETRAGONAL PHASE CHANGE OF ZrO -4 2 1. 15 x 10 3 exp (-27340/RT) -6 a- -8 B PH ASE CHANGE OF Zr -02 SOLID SOL UTIONS 0 LEISTIKOW -10 -12 (1975) (1967) 0 WHITE > HAYES AND ROBERSON (1945) 5I I 8 5 7 8 6 9 10 /T Figure 2.3 10 11 12 (*K) Correlations for Zirconium Oxidation in Air (After NUREG/CR0649) 13 in the open literature which has been derived for Zircaloy-2, the alloy from which reactor-grade LWR fuel cladding is made. It is suggested in Biederman, et al., (B2) that the alloying of zirconium has a profound effect on the oxygen diffusivity in the oxide phase. The typical composition range for Zircaloy-2 according to American specification PDS.11538-4 (Ml) is given in weight per cent as: 0.03-0.08 Tin 1.3 -1.6 Nickel Iron 0.07-0.20 Average (Fe+Cr+Ni) 0.23-0.32 Zirconium Chromium 0.05-0.16 0.68-0.77 Investigation of high temperature oxidation of a 1.5 wt % tin alloy by Mallett and Albrecht (1955) (M2) showed that the oxidation rate followed a cubic law, Eqn.2.4, in the temperature range 600-900*C at 1 atmosphere oxygen (Figures 2.4 and 2.5): 2dw 3Wdt = K0 exp( Ea/RT) 2.4 where K 0 = reaction rate constant [(ml/cm2 3S] Ea = activation energy (cal/mole) and K0 = 5.34x10 K = 87.2 4 Ea = 38400±1100 600 < T < 800 0 C Ea = 22600±1400 825 < T < 900 0 C A study of oxidation of zirconium and zircaloy in dry 0.50 0.45 0.40 U 0.35 0.30 0.25 Co 0 0.20 0.15 >1 X 0 0.10 0.05 0 Figure 2.4 5 10 15 20 Time, min 25 30 35 Cubic Plot of Oxidation of Zirconium-1.5% Tin Alloy at 6000 - 800*C (after Reference M2). 38 36 34 32 CN 30 >428 '- 26 900 0 C 24 22 20 -2 91 18 Q) T o m16 14 U S12 13 10 o 8 6 4 2 Ally82* a 0 Figure 2.5 20 40 nd 00* 60 80 Time, min (ate 100 Reerece 120 2) 140 Cubic Plot of Oxidation of Zirconium- 1.5% Tin Alloy at 825' and 900'C (after Reference M2). air performed by Kendall (1955) (Kl), covered the temperature range of 500-700 0 C. His study indicated that the reaction proceeds in two stages: initially an approximately cubic dependence with time, and at higher exposures, a linear The rate constants calculated from his function of time. data and applicable with equation 2.4 for the initial reaction are: Zircaloy-2 KO = 1.1x109 Zirconium KO = 1.8x9 (mg/cm2) 3/hr Ea = 39400 cal/mole (mg/cm2 3/hr Ea = 41400 cal/mole In the linear regime which conforms to Eqn.2.5: dw K exp(-Ea/RT) 2.5 the appropriate constants obtained from his data are: Zircaloy-2 K = 8.5x106 (mg/cm2)/hr Ea = 31000 cal/mole Zirconium KO = 7.4x6 (mg/cm2)/hr Ea = 29800 cal/mole In the research performed in this report, only the first three correlations have been implemented. Additional tests performed by Mallett and Albrecht for the oxidation of Zr -2.5 wt % Sn alloy in 1 atmosphere of 02 showed a parabolic rate 0 dependence in the temperature range of 500-900 C. While there is evidence that alloying components have a profound effect on the oxidation rate, the parabolic formulation was employed because it is the only formulation for which high temperature (>900*C) data are available. Biederman, et al., (B2) showed that the empirical formu- lation Eqn.2.3 given above may be used for the calculation of oxide thickness under transient heating conditions (assuming 02 concentration at the wall constant) by dividing the time interval into a series of small intervals and calculating the oxide formation between specified initial and final temperatures. The calculational method of Biederman, et al., does not attempt to reconcile the energy liberated during the interval by chemical oxidation with the specified temperature at the end of the interval. All of the codes proposed in the present report do account for the synergistic effect of chemical reaction energy and temperature. The spent fuel holders are closed channels in which air is admitted at the base and exits at the top of the assembly. The oxygen mass fraction of the gases is depleted as the gas moves up the channel. At some point in the channel, the chemical oxidation reaction may become limited by the quantity of 02 available for reaction; this is a diffusion-limited condition. components: In this regime, the reaction is limited by two 1) the ability of the oxygen to diffuse through the gas to the oxide surface, and 2) the ability of the oxygen to diffuse through the oxide layer to the unreacted zirconium metal. by Eqn.2.6: This is represented for a planar geometry d W0 C dW 2 = where W = 02 W - C - 2 ( O 2.6a 2 weight gain of oxygen per unit surface area 2 (mgO2 /cm ) C 0 2 (x), C0 2(o) = Oxygen concentrations at the oxidation boundary and free stream, respectively (mgO2 /cm ) HT = overall mass transfer coefficient and HT is defined as: HT where H D +h 2.6b D HD = mass transfer coefficient for 02 through the oxide layer (cm/s) hD = mass transfer coefficient for 02 through the gas to clad outer surface (cm/s) The value of HD is determined primarily by the diffusivity of oxygen in zirconium oxidation. Depending on temperature, the zirconium dioxide can be in either a monoclinic (800 0 C) or tetragonal (1200*C) crystalline form, with either preferential or non-preferential orientation. The diffusivity of oxygen is profoundly influenced by these factors; however, because of this complexity (and as a conservative assumption), no attempt has been made to model the oxygen concentration The value of hD is deter- gradient through the oxide layer. mined for laminar flow using the Chilton-Colburn analogy (C2): f h . where H 2.7a T D IH Pr2/ PCP 2.7b p hD D 2/ 2.7c (Sc) h = heat transfer coefficient (W/cm2 p = gas density (gm/cm3) c = gas specific heat V = gas velocity (J/gm *K) (cm/s) Pr = Prandtl number = v/a Sc = Schmidt number = v/D f = friction factor and D = diffusivity of oxygen in air. The values of Schmidt number are well documented for air, and the heat transfer coefficient is lations of Appendix A. of oxygen in For lack of data on the diffusivity the various forms of zirconium dioxide, transfer coefficient is the mass composed entirely of the oxygen through air diffusion portion. sponds calculated using the corre- This approximation corre- to the physical situation of the oxide layer spalling as quickly as it appears. The reaction rate used in the code is determined by the relative magnitudes of the values calculated for the reaction rate: if the rate calculated by using Eqn.2.3 is greater than the result of Eqn.2.6, the reaction is kinetically controlled, otherwise it is diffusion limited. The oxidation thickness in either reaction regime is calculated explicitly using the temperature of the clad at the beginning of the time interval for which the change in thickness is to be calculated, thus: W= where [ dt(At) = KV 0 exp (-E/RT) A+ p2 p = density of Zr K0 = Aw 22.8 (gm/cm3) (mgZr/cm2 ) 2/s = Kxf f = stoichiometric ratio of Zr to 02 in ZrO 2 Aw = reduction in clad thickness at the beginning of the time step (cm) However, since for the non-isothermal case the temperature is also a function of time, Eqn.2.3 should be: w = J-t K exp(-E/RT(t)) dt dt / + Aw2}, 2.9 Equation 2.9 must be solved using non-linear techniques, since neither the new oxidation thickness nor temperature are known. Additionally, the clad temperature is a function of decay heat generation rate, and external flow parameters. While the resulting system of equations can be solved implicitly, the computational effort is large and must employ an iterative approach by making successive approximations to the new-time clad temperature. The computational complexity as well as potential for non-convergence are the basis for the use of old-time temperatures in calculation of new-time oxidation thicknesses in this report. References (Gl) and (W3) describe implicit evaluations for stiff systems of equations such as that presented above; however, this refinement was beyond the scope of this project. 2.3 EXTENSION OF SFUEL PROGRAM TO ASSESS POOL-WIDE BURNING Calculation of fuel pool conditions subsequent to initia- tion of runaway oxidation at one or more locations in the pool necessitated the revision of the SFUEL computer code. A careful review of models in the code was undertaken to delineate both the physics-related and computational limits of the original code. The revised code was developed in an evolu- tionary manner, such that only models or programming structures which led to non-physical results or computational instability were adjusted. Models were then implemented for mechanisms such as burnout of entirely oxidized clad, fuel/ clad relocation, etc., which had not appeared in the original program. The following is a discussion of the individual models which were revised or implemented: 38 2.3.1 Inter-Holder Conduction Initial program runs with the original SFUEL code indicated that the code is numerically unstable when any of the fluid passages are blocked, such as occurs in the interholder air spaces. The oscillatory instability, reflected in the interfacial node temperatures of the inter-holder spaces, is initiated at the lowest boundary node for the non-flow channels, and leads to the appearance of negative temperatures which progress up the channel, increasing in amplitude with increasing holder temperature. tions are shown in Table 2-1. Examples of these oscillaThese oscillations exist at all times, although they do not halt the computation until the equation: TBARNC = TW - 0.38*(TW- TA) takes on negative (Kelvin) values. 2.10 Examination of the finite differencing scheme for the fluid dynamics calculations indicates that an oscillatory instability is to be expected for non-flow channels. This will be shown by performing a Von Neumann stability analysis on the linearized discretization of the energy equation employed in SFUEL. The essence of this method of stability analysis is to express the error of each term of the discretized equation with a finite Fourier Series representation. The decay or amplification of each mode is then considered separately to determine stability or TABLE 2-1 Oscillatory Instability of Interholder Space Interface Air Temperature Case Analyzed: Time = 550 sec. PWR, CYLINDRICAL STORAGE RACK 3.0" BASEPLATE HOLE Interface Air Temperature, C POOL SECTION 2 POOL SECTION 4 POOL SECTION 6 0.388 0.623 1.183 19 0.355 0.564 1.051 18 0.732 1.168 2.199 17 0.674 1.069 2.000 16 1.022 1.627 3.056 15 0.923 1.466 2.740 14 1.231 1.958 3.672 13 1.080 1.717 3.210 12 1.340 2.130 3.989 11 1.130, 1.796 3.360 1.134 2.128 3.979 1.407 1.697 3.173 1.235 1.956 3.645 0.897 1.426 1.031 1.631 2.665 2.243 0.634 0.752 1.005 1.877 1.186 0.467 2.186 0.671 -0.1384 1.232 Axial Location (top) 0.298 0.424 1 (bottom)-0.074 0.869 -0.280 The values presented above are the difference between the temperature at the location and the initial room temperature. Note that negative values correspond to temperatures below room temperature. The air interfaces in pool sections 1, 3 and 5 behave in a similar fashion and have been omitted for clarity. instability. The one-dimensional fluid conservation equations are solved in discretized form in SFUEL on a grid as shown in Fig. 2.6. The dotted lines represent the outlines of the control volumes of fluid with indices I-1, I, and I+1. The points (O's) located on the boundaries of the control volumes represent interfacial fluid values. The energy equation is solved for temperature on a grid such as Fig. 2.6, employing the value of temperature at the previous time step and the value of temperature at the fluid interface behind it. The equation for the average fluid node temperature in SFUEL is: 2*GCPA*TA3 n+ TAVE3 n+1 I I-- _ 2*GCPA + + PCA + HAT PCA TAVE3n T + WKl 2.11 1 The variables in Eqn.2.11 are defined as: (The final equality states the equation in more traditional variables.) GCPA = GOXO*CPOX + GNI*CPNI| = 'n CPOX + 1 2 CpNI mC PCA = PROOM*CP*AXA3(J)*DELX/(RA*DELT) = P0 Cp AAX/(RAt) 2.12 2.13 41 /////////////////////////// I- 6 I-1i o0 Figure 2.6 I- 2 I+7 I+1 0 Finite Difference Mesh for One-Dimensional Fluid Flow 6 42 Employing the ideal gas law p0TO = PO/Ro Eqn.2.13 becomes: PCA = p 0 T0 CpAAX/At 2.14 WKl = HWA3(I,J)*AWA3 + HWJMl(I,J)*WZDX = hA 1 2.15 HAT = HWA3(I,J)*AWA3*TW(I,J) + HWJMl(I,J)*WZDX*TWJMl(I)=hA 1 T 2.16 In Eqns.2.14 and 2.15 the wall temperatures and heat transfer coefficients have been assumed constant. 2.12 - 2.16 into 2.11, we obtain: 2C TA3 TAVE 3 n+1= I + p0 C A + hA,T 0C A Ax p At + hA TAVE3 n 1 PO T 2mC + p where A Substituting Eqns. 0 2.17 = heat transfer area and A is the flow cross- sectional area. The temperature of the interface fluid node is calculated by using: TA3I+ 2 - TAVE31 - TA3q = 2.18 Since the temperatures showing the oscillation are the interface nodal temperatures, expand Eqn. 2.18 using 2.11. (The nomenclature TA3 and TAVE3 will be dropped in favor of T, where the location of the point is denoted by the subscript as in Fig. 2.3.) n+1 TIn+ = 2mCT 2 PT0 CA 1+ p I--_ Ax + hA Tw - n Tn~ 2.19 PO TOC A A I+4 2mC p + n +At hA 1 Eqn. 2.19 is nonlinear, incorporating variables m, Cp, p0 ' T 0, h, Tw, nator. and the old-time level value of T in the denomi- In order to perform a stability analysis, Eqn. 2.19 To do this, let the ratio Tn/T0 must first be linearized. Also, let p0 = p; m= puA; and 1, or equivalently, T = T p, u, CpI Tw, The analysis may be A, A 1 and h be constant. further simplified by neglecting the assumed constant heat source hA Tw Tn+l I+42 Equation 2.19 reduces to: 2 2puACT n p J + pT nCC A An+1 + pC A 2puAC p + hA At 1 T 1 Dividing pC pA through the brackets in the above equation 20 yields: ' 2 n+l +4 2uTn+l + Ax Tn At I 12u + + hAl/pC A At ip 1 Tn+ Tn+ 1 - 2.21 The Fourier component of the solution may be written as: n ei(mIAx) Tn _ 2.22 where $ n is the amplification function at time-level n of the particular component whose wave number is m, and i = V1. Substitution into Eqn. 2.21 gives: Ax AtL 2.23 hx A A+ hCA cos + 2iu sin- p The amplification factor $ is given by the complex Eqn. 2.23. For stability : 1, then: l Ax 2 $ 2 hA1 At+ pCCA p 2 1 2m cos 2 2.24 + 4u sin 2m s 1, the choice of m will rehA For stability , and u. strict the values of Ax, At, pC p 1 It is apparent that for I$l S< 1: S< 2 hAn + pC A 22 os 2 + 4u sin m 2.25 For This equation must be satisfied for all wave numbers m. = 0, Eqn. 2.25 becomes: the case sin }= 1, cos 22 2 At 1 < I 2A )[ 4u 2 Ax u > 2At 2.26 Equation 2.26 represents the stability criterion for the SFUEL finite difference scheme. Note that this condition is not met for any of the SFUEL cases which involve blocking one or more of the flow paths, since for these cases u = 10-10 cm/s. Because the air in the inter-holder spaces is actually stagnant, the old SFUEL scheme was replaced by a fully implicit two-dimensional conduction scheme, assuming that there are no buoyancy-driven recirculation flows in this region. The conduction equation is: Tn -+1 T fTI At 1 +T - 2T I + TW+ (Ax) 2 TW 2 -2T 2 2.27 (Ay) 2 where Tn is the previous time interval value and all other values are determined as the new time step. The new-time values of the holder wall temperatures, TW1 and TW 2 , are evaluated explicitly as in the original SFUEL, while the thermal diffusivity is evaluated at the previous node-average air temperature. The interface air temperatures are determined as the arithmetic averages of the average air temperatures determined using Eqn. 2.27. The set of simul- taneous equations (2.27) for all axial nodes are solved with a standard tridiagonal matrix inversion routine. For the steady-state cases reported in NUREG/CR0649, where the amplitude of the inter-holder oscillations is small, this conduction model produces essentially identical results. How- ever, the model's utility is not in confirming previous work, but in allowing extension of the code to the calculation of conditions which could not be assessed previously. 2.3.2 Revised Fluid Dynamics Calculations Examination of the code results subsequent to the revision described above revealed another difficulty with the assembly out-flow boundary evaluation, independent of the interholder non-flow approximation. This problem was mani- fest in the appearance of large negative values of air temperature at the top-most node. The air temperature differ- encing scheme presented in Sec. 2.3.1 incorporates an upwind scheme for the first half of the node control volume, followed by an extrapolation to obtain the node out-flow interface temperature (see Eqns. 2.11 and 2.18). This differenc- ing scheme is thought to have undesirable characteristics, especially with regard to boundary conditions. The stabi- lity criteria of the upwind-extrapolation at the boundary conditions cannot be obtained with Von-Neumann analysis and requires other analytical techniques. Consequently, a for- ward differencing scheme was substituted, which provides physically real values of air temperatures at all points in the calculational domain. The forward differencing in the revised SFUEL program consists of the use of Eqn. 2.11 over n+1 is replaced by the node the entire node length, where TA3 n_ average temperature of the previous node divided by 2: TAVE3n 1 I Sn+l =GCPA * TAV+ PCA+ HAT =CA2.28 GCPA + PCA + WK1 TAVE3n where: TAV = TA4AVE if I= 1 average temperature of base node = TAVE3 if I > 1 average temperature of pre- vious node This scheme requires a fictitious top-most convective node in order to determine the top-most interface air temperature. The interfacial air temperatures are calculated as the arithmetic average of the node average temperatures of the two adjacent nodes, as for the non-flow channels. The change of differencing scheme was motivated partly by the appearance of the large negative outflow temperatures, and partly by a non-convergence of mass flows. While this re- vision did not resolve the mass flow divergence, it does result in physical temperature fields and has also been retained for its better known computational characteristics. 48 2.3.3 Maximum Driving Force Approximations The mass flow divergence alluded to above comes about due to the orifice pressure drop at the base of the fuel assembly holders. Since the pressure drop is calculated in terms of reduced pressure, under certain conditions the pressure drop calculated at the holder inlet is sufficiently large to be outside the range of convergence of the Newton Raphson scheme. This obstacle has been overcome by using an approximation based on purely physical grounds. All cases investigated during this research achieve steadystate mass flows while the temperatures of the fuel assemblies are still relatively low. Further increases in gas temperatures produce decreases in the channel gas density which is already very low compared to the room air density. Therefore, the buoyancy force, which is proportional to the difference bewteen the room air density and the channel air density, approaches a constant value: Buoyancy Force - lim pchannel 0 P room g(l _ channel proom _ = room g 2.29 This is the physical basis for making the approximation that the mass flows do not change from their steady values once runaway oxidation initiates. In the SFUEL code, this appro- ximation is implemented by by-passing mass flow calculations after the initiation of runaway oxidation, typically several hours after the steady-state mass flow distribution has been attained. For fuel pool analyses in which the holder base- plate holes are greater than 5 inches, it has been observed that this approximation is unnecessary, since the mass flow calculations do not diverge subsequent to the initiation of runaway oxidation. Cases run in which the mass flow calcu- lations are continued during pool-wide burning indicate that the constant mass flow approximation is slightly nonconservative, since it provides larger mass flows (lO10%) after fuel/clad relocation than would be sustained by the buoyant driving force. The use of steady-state mass flow values for pool-wide burning calculations is thought to be better than the alternate approximation of assuming that all channels are connected to an infinite reservoir, since in the physical situation, the total pool-wide mass flow is limited by that which can flow between the liner and the first holder assembly. It is thought that this approxima- tion results in a slightly reduced mass flow relative to the infinite reservoir approximation and is therefore the more conservative of the two approximations. 2.3.4 Revision of Clad Thickness Model It has been observed that the original SFUEL model did not account for total consumption by oxidation of the clad in a node. That is, it was possible for clad to "regener- ate," especially when the node experiences repeated transi- tions from the kinetic-controlled to diffusion-controlled regimes. Several minor programming changes were required to assure that a fully oxidized clad node did not continue to generate chemical oxidation energy. 2.3.5 Limit to Maximum Rate of Change of Temperature The fuel temperature calculation scheme in the original SFUEL is fully explicit. By definition then, the new time step fuel temperature is given by the sum of the previous temperature plus the difference bewteen the new-time energy input minus the old-time energy losses, multiplied by the ratio of the time-step size to the thermal capacitance of the fuel/clad node. When the energy input to the node dur- ing the new time step is far greater than the previous time step energy input for which the losses were calculated, i.e., when runaway oxidation initiates, the new time temperature can become very large in a very short time. Use of Hamming's stability analysis (H4),(S3) for the relevant equations illustrates this point. The fuel/clad temperature is determined as: TRI = TRI + QRIJ* DELT/CR where: TR = rod temperature at axial location I CR = thermal capacitance of fuel/clad node = (RHOF*AF*CR+ RHOC*AC*CC)*DELX = (pcPA)TAx 2.30 RHOF density of fuel RHOC density of clad AC = cross sectional area of clad AF = cross sectional area of fuel CC = specific heat of clad CF = specific heat of fuel DELX = length of fuel/clad node = Ax DELT = time interval size = At and QRIJ = QDECAY I +QCHEMI + QCOND - QCONDI+1+ QR 2.31 where: QDECAY1 = heat generation due to radioactive decay QCHEM = chemical oxidation energy input QCOND = energy conducted into node from node below 1 = _1 - TRI)/DELX energy conducted out of node from node above QCOND I+l = (AF*SMKF (AF*SMKF + AC*SMKC)*(TR (AF*SMKF + AC*SMKC)*(TRI - TRI+ 1 ) /DELX + AC*SMKC) = (kA)T SMKF = thermal conductivity of fuel SMKC = thermal conductivity of clad QR = radiative and convective loss from the node ) + radiative losses = HRAl*AR (TAVEln Iz-TR IR 52 HRAl = convective heat transfer coefficient = h AR = surface area of the clad node = A TAVEl = temperature of the air at node I All of the above heat fluxes are determined using temperatures at the previous time step. To simplify the analysis, neglect QCHEM and the radiative losses. Then QRIJ is given by: QRIJ = hA(TAVEl - TR ) + kAT(TR + + TRn_ T n~ I It 2TR IT~/D~ )/DELX 2.32 follows that TRn+l is equal to: I TRn+1 = TR + + _ phAt p 1lT Ax 2 (TR 1+ TR -2TR) (TAVEl2- TRn) 2.33 For TAVEln constant, the difference equation will be convergent only if the spectral radius of its matrix of coefficients exceeds unity, but is less than 1+ O(t): A = 1+ ( At (Ax)2 hpAt pA PT k pA sin 2 (N;-l)lr T 2N 2.34 The term (hAAt/pcpA2 )T allows the convergence criterion to be satisfied for the restriction: At (Ax) 2 - 2 2.35 pcpA k 2.35 Equation 2.35 is the stability criteria for conductive and convective stability of the rod temperature scheme outlined above. Similar, although more restrictive stability cri- teria result from the non-linearities introduced by radiative heat transfer and chemical oxidation energy input. The strategy adopted to maintain stability has been to limit the temperature rise during any interval to a value in concert with the fuel numerical stability limits. This value is not attainable in closed form for the chemical oxidation energy, since this varies non-linearly. Instead, a value of temperature rise which conservatively satisfies the calculable stability criteria is used (100*C/s) and the results carefully monitored for the appearance of irregularies. If the temperature rise during an interval fails to meet the above criterion, the time interval is reduced, all values initialized to the previous values, and the calculations repeated. This strategy is quite successful in containing the exponential growth observed in the original SFUEL program. When the clad experiences vigorous oxida- tion, the time steps are typically reduced by 2 orders of magnitude to satisfy the stability limit. 2.3.6 Fuel Relocation Models Several possibilities exist for modelling fuel relocation when a node reaches the melting point of zirconium dioxide, approximately 2740*K, assuming that molten zirconium, if any, will be contained by the oxidized clad: 1) Ignore clad change to liquid phase; allow fuel/clad temperature to continue to climb. 2) Hold node temperature at melting point for all successive calculations. 3) Hold node temperature at melting point; calculate energy deposited in melting zirconium dioxide; zirconium solid remains intact. 4) Let node disappear (rod relocate) when node clad temperature reaches the melting point. Since only a few rods are restrained at the top, it is reason- able that all nodes above the melted node should relocate as well. 5) Hold node temperature at melting point until mass is completely melted; followed by relocation as in (4). Options (1) and (4) have been investigated in this report, as they represent the extreme situations. Options (2), (3) and (5) require a kinetic model for melting, which is a function of the grain structure at a given radial location in the clad. Such a refinement is beyond the scope of this project. 2.3.7 Computer Operating System Changes In addition to the modelling revisions noted above, programming changes related to system compatibility with the researcher's computer operating system have been made. A restart option has been added to the program to allow the program to be interrupted and restarted at a later time to complete the calculations. The feature was required to keep the computational costs to a minimum, as well as to allow parametric evaluation of certain code variables. 2.4 MODIFIED SFUEL ANALYSIS OF BURNING PROPAGATION 2.4.1 Characteristics of Revised SFUEL Code Following the series of revisions described above, a sample case was analyzed with both the original SFUEL code (except for operating system re-programming) and the revised version of SFUELlW. The case chosen is one which achieves steady-state: PWR SPENT FUEL 17 x17 ARRAY CYLINDRICAL STORAGE RACK CONFIGURATION FULL CORE DISCHARGE LOADING, 3" BASEPLATE HOLE The plot of temperature vs. time for this case is displayed in Fig. 12 of NUREG/CR0649. (1 8 th node, Figure 2.7. Table 2-2. 6 th The maximum clad temperature pool section) for the two codes is shown in Values depicted in Figure 2.7 are tabulated in A comparison of the steady-state nitrogen mass flow distribution in the pool sections is presented in Table The agreement is good and it indicates that the re- 2-3. vised code is functioning properly and also that the prerunaway temperature calculations obtained with the initial SFUEL code are satisfactory despite the objections noted earlier. The variation between the two codes is expected to become for pronounced at higher rates of change of temperatures. This is shown for the case of: PWR SPENT FUEL 17 x 17 ARRAY CYLINDRICAL STORAGE RACK CONFIGURATION FULL CORE DISCHARGE LOADING, 1.5" BASEPLATE HOLE 350 2 300 :94 4 6 6 66 250 original SFUEL 200 A revised SFUEL 150 100 50 0 0 5 10 15 20 25 30 35 40 45 Time from initiation of accident x 103 seconds Figure 2.7 Maximum Clad Temperature for Original and Revised SFUEL. TABLE 2-2 PWR Spent Fuel 17 x 17 Array Cylindrical Storage Rack Configuration Full Discharge Loading, 3" Baseplate Hole MAXIMUM CLAD TEMPERATURE (18 NODE, 6 SECTION) x103 secs Original SFUEL Modified SFUEL 0.0 10 10 3.6 93.9 92.0 7.2 197.0 188.9 10.8 276.6 264.0 14.4 311.2 302.7 18.0 317.9 316.5 21.6 317.4 320.2 25.2 316.8 320.9 28.8 316.7 321.1 32.4 316.7 321.1 36.0 316.7 321.2 39.6 316.7 321.2 43.2 316.8 321.2 Time TABLE 2-3 Steady State Nitrogen Mass Flow Rates Pool Section Fuel Assemblies Original SFUEL (gm/s) Modified SFUEL 1 6.79 6.54 2 33.40 33.74 3 36.31 36.68 4 37.92 38.25 5 41.03 41.35 6 43.68 43.93 Liner Holder Space 199.1 200.5 Figure 2.8 illustrates the effect of the revised fluid dynamics and holder heat transfer calculations during runaway, and suggests that the results obtained with the original SFUEL are accurate up to runaway. The effect of the fuel temperature stability limit is particularly well depicted by the behavior of the original SFUEL subsequent to sustaining oxidation in the 2nd kinetically controlled regime (correlation in the center of Figure 2.3). The revised version is not as subject to the accumulation of fuel temperature error and growth as the original version, by virtue of the time step reduction capability. Consequently, the temperature rise of the peak clad node predicted by SFUELlW is smoother and occurs over one and one-half hours later than predicted in the original code. The values graphed in Figure 2.8 are tabulated in Table 2-4. The re- sults presented thus far indicate that results calculated by the original SFUEL are conservative with respect to both temperature rise during oxidation and time to initiation of runaway oxidation. The results given above validate the new code's ability to predict fuel pool conditions up to the point of initiation of runaway oxidation. Figures 2.9 and 2.10 illustrate the code's predictions for the maximum node clad temperature (node 18, section 6) up to the time at which the node either burns out (upper curve) or melts and falls unimpeded 2200 + original SFUEL 2000 A revised SFUEL 1800 1600 1400 1200 (1) 1000 800 ++ A + 600 (1) Revised SFUEL reaches runaway conditions at t = 42500 sec. 400 200 5 10 15 20 25 30 35 40 Time from initiation of accident x 10 3 seconds Figure 2.8 Comparison of Original and Revised Codes during Runaway Oxidation. 45 62 TABLE 2-4 Comparison of SFUEL Results Under Runaway Conditions Time Original Version Modified Version secs TR (18,6) *C TR (18,6) *C 10.0 10.0 1800 43.89 43.61 3600 77.94 76.59 0 5400 115.4 112.4 9000 200.2 192.7 10800 246.2 235.9 12600 293.6 280.9 14400 341.7 325.4 16200 389.9 370.4 18000 437.8 414.7 19800 485.4 458.3 21600 533.8 501.4 23400 583.9 544.7 25200 636.4 588.5 27000 691.4 633.1 28800 748.4 678.1 30600 806.8 723.5 32400 865.2 768.7 34200 921.3 813.5 36000 995.2 857.4 37150 2012.3 37800 x 899.5 39600 code breaks down 942.0 41400 1014.0 3500 + 3000 clad melt, Option 4 0 clad melt, Option 1 2500 u 0 5 2000 ci) ro1500 P 1000 500 Expanded View Fig. 2 - 0 50 100 150 200 250 300 350 Time from initiation of accident x 102 seconds Figure 2.9 Peak Clad Temperature during Runaway. 400 450 3500 +clad melt, Option 4 Oclad melt, Option 1 3000 U 2500 w 02000 4I 1500 1000 - +-. ---- +-+ 500 0 412 414 416 418 420 422 424 426 428 430 432 Time from Initiation of accident x 102 seconds Figure 2.10 Comparison of Peak Clad Temperatures, Melt Options 1 and 4 434 65 to the pool floor. Note that Figure 2.10 is an enlarged view of the oxidizing time interval depicted in Figure 2.9. The numerical values in Table 2-5 corresponding to Figures 2.9 and 2.10 reveal that the total time from initiation of runaway to clad melting is approximately 8n,10 minutes. The decrease in clad temperature for option 1 occurs because the clad has been completely oxidized. The purpose of the foregoing discussion was to demonstrate the capability of the revised SFUEL code to predict the behavior of the peak clad node temperature from initiation of runaway oxidation to melting/burnout. With this tool available, the question of propagation of the vigorous zirconium oxidation from the fresh fuel pool section to the older fuel sections of the pool is addressed. Figures 2.11 and 2.12 present the peak clad node temperatures' histories for the fuel assemblies in pool sections 5 and 6 for the 1.5 inch baseplate hole case described in Figures 2.9 and 2.10. Figure 2.11 illustrates the impact of the "total" fuel relocation model (option 4)-where all nodes above the melted node disappear-while Figure 2.12 gives the result for the case where the high temperature node neither melts nor relocates (option 1). The lower curve in each figure depicts the behavior of the peak clad node temperature in the adjacent fuel pool section. In Figure 2.12, it is evi- dent that sufficient energy is transferred to the adjacent TABLE 2-5 Maximum Clad Node Temperature for Different Fuel Relocation Options (Modified SFUEL) Time x10 3 secs Peak Clad Temperature (*C) 0.0 10.0 3.6 76.6 9.0 192.7 12.6 280.4 16.2 370.4 19.8 458.3 23.4 544.7 27.0 633.1 30.6 723.5 34.2 813.5 37.8 899.5 39.2 942.0 41.4 1014 42.3 1117 42.54 1656 42.61 2117 42.67 2465 42.73 2747 42.79 994 2981 42.86 42.98 43.97 1002 1002 3194 43.11 1009 Option 4 3428 3361 43.13 3293 43.19 4237 43.20 1014 1 Option 1 2800 - 2600 . Clad Melt Option 4 + peak node, Section 6 peak node, Section 5 2400 2200 -p 2000 a) 1800 - 1600 - .1400 a> 1200 - 1000 -- 4- --- ---- 800 600 424 423 425 426 Time from initiation Figure 2.11 427 428 429 430 431 432 of accident x 102 seconds Peak Clad Temperatures for Pool Sections 5 and 6, Melt Option 4 - 3500 Clad Melt Option 1 + - 3000 O peak node, Section 6 5 peak node, Section 3 W 2500- 0- 2500 1 000 100 423 424 425 426 427 428 429 430 431 432 Time from initiation of accident x 102 seconds Figure 2.12 Peak Clad Temperatures for Pool Sections 5 and 6, Melt Option 1 pool section to engender runaway zirconium oxidation. The fuel relocation model, Figure 2.11, not only results in little energy transfer to the adjacent pool section, but the subsequently cooler air temperatures at the burned out axial position cools the wall and, indirectly, the adjacent pool section. Unless a physical situation can be envi- sioned in which the fuel behaves as in Figure 2.12, the propagation of runaway zirconium oxidation to the adjacent fuel pool section for this configuration might be ruled out. Examination of the approximations involved in calcu- lating Figure 2.11 (instantaneous phase change and relocation upon achieving the melting point of ZrO 2 ) indicates that the actual situation may yield a curve for the adjacent pool section peak clad temperature which lies slightly above that presented. 2.4.2 Pool-Wide Propagation Results A pool-wide analysis was performed for the case: PWR SPENT FUEL 17 x 17 ARRAY HIGH DENSITY STORAGE CONFIGURATION 90 DAY MINIMUM DECAY TIME, 5.0" BASEPLATE HOLE to determine whether the runaway oxidation would propagate across fuel pool sections. This case was run with contin- ually-calculated mass flows and clad melt option 4 (complete relocation upon reaching melting point of ZrO 2 ). Propaga- tion was observed to occur across four pool sections. The temperature profile for the initially hottest node in pool section 6, the first to undergo runaway oxidation, is shown in Figure 2.13. The arrow in the figure denotes the time at which the fuel/clad relocates; the subsequent temperature profile is for air occupied by the fuel/clad node. The temperature at the location is observed to increase and decrease as the remaining fuel/clad nodes beneath it undergo vigorous oxidation. These oscillations would be expected to be damped in the actual pool, as relocated clad and uranium dioxide block the lower channel and reduce the available air flow and therefore reduce the oxygen available for chemical interaction. Figures 2.14, 2.15 and 2.16 display the temperature-time history for the peak clad temperatures in the three adjacent pool sections which experience vigorous burning. Figure 2.17 is a composite of the four preced- ing figures, and is intended to show the waiting times before the various sections of the pool ignite. It may be inferred from the figure that chemical oxidation energy liberated in pool section 6 has an immediate effect upon the oxidation rate in section 5; however, the decay heat generated within these two sections is approximately the same. Sections 4 and 3, which are modelled as much older fuel, and have lower decay heat generation, require significantly longer times to experience runaway oxidation. 2800 POOL SECTION 6 CLAD RELOCATES 2600 2400 2200 I I I I I 2000 1800 1600 1400 1200 I' 11 21 1000 _ 800 220 230 240 250 260 270 280 290 Time from initiation of accident x 102 seconds Figure 2.13 Temperature in Pool Section 6, Axial Location 16 300 310 2800 POOL SECTION 5 CLAD RELOCATES 2600 U 0 2400 0 2200 1 I -H 4-) 0 1-4 2000 I 1800 I 1600 I -) (1) 4 I 1400 + E-4 I 1200 V 1000 - sa %.. + ..+ .+ f+++ 800 2 220 Figure 2.14 230 240 250 260 270 280 290 Time from initiation of accident x 102 seconds Temperature in Pool Section 5, Axial Location 16 300 310 2800 POOL SECTION 4 CLAD RELOCATES 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 I 220 min 230 240 250 260 270 280 Time from initiation of accident x 10 Figure 2.15 2 2 Temperature in Pool Section 4, Axial Location 16 290 seconds 300 310 2800 POOL SECTION 3 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 220 230 240 250 260 270 280 290 Time from initiation of accident x 102 seconds Figure 2.16 Temperature in Pool Section 3, Axial Location 16. 300 310 PWR HIGH DENSITY, 90-DAY MINIMUM DECAY TIME, LARGE BASEPLATE HOLES 2800 Pool Section Symbol 2600 2400 2200 A 6 + 5 o 4 0 3 2000 1800 1600 1400 1200 1000 -- 800 220 230 240 250 260 270 280 290 Time from initiation of accident x 102 seconds Figure 2. 17 Composite View of Pool Wide Propagation Results 300 310 In an effort to assess the sensitivity of the burning propagation relative to the decay heat, a series of analyses were performed in which the center pool section was given a high decay heat generation rate, and the remaining sections were given a substantially lower decay heat rate; these values are reported in Table 2-6. The results for the third case are shown in Figure 2.18; the values used in this plot are tabulated in Table 2-7. Examination of Figure 2.18 shows that the highest clad temperature in pool section 5 peaks approximately 2 hours after runaway oxidation occurs in the adjacent (6 th) pool section. The peak temperatures in the other pool sections are quite low and achieve steady-state as the remaining clad in the sixth pool section burns out. This trend is clearly evident in Figure 2.18, where the air temperature in the sixth pool section is observed to be generally decreasing (between the intervals in which the remaining clad in section 6 ignites). Of the cases shown in Table 2.6, cases 1 through 4 attained peak temperatures in pool section 5 and subsequently cooled (without experiencing runaway) as the 6th pool section burned out. 5 th Cases 5 and 6 sustained runaway oxidation in the pool section. By plotting the peak temperatures reached in pool section 5 versus the decay heat generation rate, a curve of burning propagation is established, shown in Figure 2.19. This curve indicates that the minimum decay time for TABLE 2-6 Decay Heats Used in Assessing Poolwide Propagation PWR SPENT FUEL 17 x17 ARRAY HIGH DENSITY STORAGE CONFIGURATION > 5.0" BASEPLATE HOLE POOL SECTION (1 fuel assembly per section)* 1 2 3 Case 1 1.0 1.0 1.0 Case 2 2.0 2.0 2.0 Case 3 3.0 3.0 Case 4 4.0 Case 5 Case 6 4 5 6 1.0 30.0 2.0 2.0 30.0 3.0 3.0 3.0 30.0 4.0 4.0 4.0 4.0 30.0 5.0 5.0 5.0 5.0 5.0 30.0 6.0 6.0 6.0 6.0 6.0 30.0 *Units of decay heat are KW/MTU. The decay time for these heat generation rates may be estimated from Table 1-1. 1200 A Pool Section 6 CLAD RELOCATES 1150 - o 1100 - Q) - 0 Pool Section 5 A 1050 1000- (0i a) 950 -g 900- 850 800 18 20 22 24 26 28 30 Time from initiation of accident x 103 seconds Figure 2.18 Poolwide Propagation Sensitivity, Case 3 32 34 TABLE 2-7 Assessment of Sensitivity of Pool Wide Propagation PWR, High Density, Section 6 @ 30 KW/MTU, All Others @ 3.0 KW/MTU Time From Initiation of Accident x103 secs Temperature at Axial Location, *C POOL SECTION 6 NODE 15 18. 853.7 20. 913.6 22. 987.9 POOL SECTION 5 NODE 15 23.29 1081.3 23.42 1140. 23.54 2104. 23.55 2741. 23.67 1090. 23.79 1041. 23.92 1056. 24.04 1074. 24.17 1098. 24.29 1136. 24.42 2263. 24.54 1032. 28.17 1002. 28.29 1007. 851. 28.79 1033. 857.6 29.29 1076. 864.3 29.79 928. 876.8 30.29 929. 931.4 30.79 932. 935.3 31.29 936. 934.5 31.79 940.3 925.9 32.29 944.7 915.8 32.79 949.3 906.0 33.17 953.1 899.0 I I I If I I 1000 Ii 975 4 ' '-- 950 /'I /'I I GLr) 900 - E4J (o 4j 925 - 4 >1) 875 0 0 / 850 _ / / / / / / / A INITIATION OF RUNAWAY OXIDATION It, 825 - / 800 - / I I / DECAY HEAT Figure 2.19 (KW/MTU) Results of Propagation Sensitivity Analysis, High Density Storage. 81 spent fuel to be stored adjacent to recently discharged fuel (,\90 days) is approximately 2 years (see Table 1-1). For the cases presented in Table 2-6, the vigorous oxidation reaction propagated only between pool sections 6 and 5. The peak clad temperatures in section 4, even while section 5 was experiencing rapid oxidation (case 6, Table 2-6), did not exceed 750*C. Thus, any storage configura- tion which uses fuel of the same or lower decay heats than shown in Table 2-6, cases 1- 4, is not susceptible to propagation of vigorous zirconium burning. 2.5 PROPAGATION OF ZIRCONIUM BURNING VIA CONVENTIONAL HEAT TRANSFER MECHANISMS: SUMMARY The SFUEL code described in NUREG/CR0649 has been re- vised and extended to analyze the poolwide propagation of vigorous oxidation of zirconium alloy clad. Correlations for zirconium oxidation have been presented which suggest a variety of reaction rate laws may apply; linear, parabolic or cubic. An explicit formulation of the parabolic reaction rate model is used exclusively in this report, as it is the most commonly cited phenomenological model for zirconium oxidation. It is shown that it is possible for widespread burning of zirconium to occur in a spent fuel pool of high density or cylindrical storage configuration, following loss of water. A curve is given (Figure 2.19) which indicates the 82 minimum decay heat generation rate for spent fuel which is to be stored adjacent to freshly discharged fuel, if runaway oxidation is to be confined to the freshly discharged assembly. Future work on this topic should focus on removing some of the conservative approximations made in this analysis. In particular, the conduction scheme through the stagnant air in the interholder spaces should be re-modelled to allow buoyancy induced recirculating flows. Also, axial conduc- tion of heat in the holder walls would tend to flatten the axial temperatures along the fuel rod length during the temperature excursions engendered by zirconium burning by increasing the radiative heat transfer. Finally, it is recom- mended that a larger database be developed for oxidation of Zircaloy-2 in air, especially with respect to the effect of zirconium alloying on the amount of chemical energy released during the oxidation reaction. CHAPTER 3 PROPAGATION OF RUNAWAY OXIDATION VIA PARTICULATE TRANSPORT 3.1 INTRODUCTION The burning particulate model developed during this project simulates the behavior of an oxidizing sphere of zirconium metal levitated by convective air currents. The burning particle model requires the simultaneous solution of mass, energy, and momentum conservation for the particle. Employing initial conditions determined from SFUEL, the code calculates the temperature, velocity, distance, mass gain and heat generation rates as functions of time and oxidation rate for a variety of allowable particle diameters. Parti- cle mass gain via oxidation, heat transferred by convection and radiation from the particle during flight are accounted for in the computer program. The oxidation model in the program, developed in spherical coordinates, assumes a kinetically controlled reaction, based on the parabolic models presented in Sec. 2.2. Diffusion limited kinetics are not included because the particle is in motion and will always be moving into a region of fresh air. The distributions of particles determined from the PARTICLE code are employed in determining an effective heat flux to the receiving surface (upon which the particulate comes to rest). These evalua- tions are performed using closed form analytic transient conduction equations. The primary assumptions embodied in the burning particulate analysis are: 1) Gas velocity in the subassemblies is uniform throughout a pool section; 2) Air properties for the subchannel are assumed to be those of the hottest fuel node; 3) Interaction of particles in the flow is neglected; 4) The amount of zirconium metal spalling into the air flow is equal to one half of the amount oxidized in place on the fuel rod; 5) The initial temperature of the particle is equal to that of the hottest clad node in the assembly; 6) To account for turbulent convection at the subassembly exit and subsequent lateral dispersion of the particles, a horizontal component of velocity equal in magnitude to the vertical particle velocity is assigned; 7) Subsequent to exiting the subassembly, the air temperature is assumed to be room air temperature; 8) Room air pressure is used in all calculations; 9) Air velocity exiting from inter-holder spaces is assumed negligible, in agreement with Sec. 2.1, thus the particle descent is in stagnant air unless it enters another holder/pool section; 10) All particles are assumed to land on a horizontal surface at the same height as the origin; 11) The heat flux to the receiving surface is approximated as that conducted in from the projected area of the particle, while the rest of the energy is lost by radiation and convection. There are many interacting effects which would need to be quantified in an exact analysis; however, the threat due to particle levitation is insufficient to warrant a more detailed analysis. Assumptions 4 and 5 are approximations, since there is as yet no formulation for the spallation rate of unreacted zirconium nor is there a quantitative database to develop a phenomenological model. 3.2 PARTICULATE PROPAGATION POTENTIAL ALGORITHM 3.2.1 Particle Model Description The PARTICLE code outlined above calculates time-temperature histories for spheres of unreacted Zr metal assumed to spall from the oxidizing fuel rods. The initial conditions taken from SFUEL are: TR(I) = temperature of hottest fuel/clad node TAVEl(I) = temperature of air flow at the assembly exit GNIl(I) = mass flow rate of nitrogen in pool section AXAl = cross-sectional flow area of pool section PROOM = room pressure TROOM = room temperature The range of particle sizes investigated in the code are: D largest possible particle that can be entrained by the air flow (cm) DMIN smallest particle that will not be extinct subsequent to exiting a specified upper plenum distance (cm) The parabolic kinetic oxidation formula is (see Sec. 2.2): W = /K 0 t exp(-E/RT) where W = mass gain in oxygen (mgO2 /cm2 ) T = temperature *K K = Arrhenius pre-exponentiation constant 2 ? (mgO 2 /cm ) '/s E = reaction activation energy (cal/mole) R = gas constant (cal/mole 3.1 4K) Differentiating with respect to time (constant temperature) yields: K 0 exp (-E/RT) dw _ dt 3.2 -r = mass gain per time (mg/cm2)/sec. where 7' For a spherical particle of mass m = 3 3pD , the deriva- tive of mass with respect to time is: pD2 dD dm = 2 3.3 dt Multiplying Eqn. 3.2 by the surface area of a sphere and equating to Eqn. 3.3 yields: dD -f / K 0exp(-E/RT) 3.4 PZr dt- pZr = density of Zirconium (mg/cm3) where f = stoichiometric mass of Zr to mass of O2' and the minus sign indicates particle oxygen mass gain. Integrating (with constant temperature) from D D <D 0 to D, and time from 0 to t gives upon rearranging: D = D 0 2f /K0exp(-E/RT) / 03.5 PZr . (PZrDO) 2 0 < t for < 2 4f K0 exp (-E/RT) D where = outer diameter at time t= 0. The mass of oxygen added to the particle is then: M 02 (t) = f Zr D03_ D0 2f v/K 0 exp (-E/RT) 3.6 PZr during (pD0 ) 2 0 < t < 4f2 K 0 exp (-E/RT) and constant (C) M~~ D3 PZr Tr 0 f02( thereafter. Heat generated in the particle during the interval of oxidation is: QOXID = AHm where 0 f QOXID = heat generated (J) AH = heat of combustion 12.03 J/mg. 3.7 89 As was discussed in Sec. 2.2, the oxidation is evaluated isothermally during a given time step. The temperature of the particle is obtained from an energy balance, assuming the particle is thermally thin: m tCptt dT mtot where tot dt q O0XID ~ UCONV 3.8 ~ ERAD mttCptt dT = heat storage in particle, = total mass of particle at time t, (mg) mtot = total specific heat of particle (J/*K) Cptot = heat generated by oxidation, Eqn. 3.7, qOXID (W) qCONV = hTrD 0 2 Tw -Tair h = heat transfer coefficient, (W/cm2K) Tw = temperature of particle, (*K) T. air = ambient temperature, q RAD = Eo i7DO (T 0 2 (J) 4 4 w - T air ), (*K) (W) = emissivity of particle = Stefan-Bolztmann constant The equations of motion for the particle are dealt with in rectilinear coordinates: d y dt2tot CD air 2 2 y(O) = 0 y'(0) = V 0 3.9 90 d 2X 27D 2 D Pair 2 dt 8m x totrD (0) = 0 x' (0) = 0 3.10 where d 2 = net acceleration in x, y directions dt2' dt g = acceleration due to gravity (980 cm/s 2 CD = drag coefficient as function of particle Reynolds number, based on the magnitude of velocity, V (mg/cm ) pair density of room air, mtot total mass of particle at time t, (mg) y dt' dt = velocities in respective coordinate positions V = = ( dt )2+ ( dx)2 '/2 dtI vector velocity, (cm/s) Terminal velocity of the particle is calculated from one of several formulas shown in Table 3-1 (Pl), and is determined by the particle Reynolds number. The convective heat transfer coefficient from the burning sphere is calculated using the following equation developed by Whitaker (Wl): Nu = 2+ (0.4 Red + 0.06 Re d w) 4 which is valid for the range 3.5 < Re d < 8 x10~. 3.11 For Reynolds' TABLE 3-1 Terminal Velocity of Settling Particulate Particle Diameter Reynolds Number mL 2 x10 3 - 10 6 5 x10 2 - 10 6 Laws of Settling 2 Newton' s C = Law p gD p tp - 2 x 10 3 2.0 - 5 x 10 2 cr (p - p) s K cr = 2,360 Intermediate Law C = 18.5NRe-0.8 0.153g u 3 - 10 2 10~4 - 2.0 t 0.71 1.14 D (p -p) s p 0.29 0.43 0.71 = p p- K cr = 43.5 Stoke's Law C = 24NRe 1 gD 2(p t 18P See next page for explanation of terms. - p) K cr = 3.3 -13 gp (ps - P) - 0.44 u= 174 102 Critical Particle Diameter TABLE 3-1 (Cont'd.) C = overall drag coefficient D = diameter of spherical particle, (cm) Dp, crit = critical particle diameter above which law will not apply, (cm) g = local acceleration due to gravity (cm/s ) NRe = Reynolds number, DPput/.J p = fluid density, (gm/cm ) p5 = particle density, (gm/cm ) 93 numbers less than 3.5, the value of Nusselt number for conduction from a sphere to an infinite, stagnant medium is employed: Nu = 2.0 3.12 The equations for energy and momentum conservation, 3.8- 3.10, are integrated numerically as simultaneous ordinary differential equations, using a third-order Runge Kutta scheme. A listing of the PARTICLE program is given in Ap- pendix C. 3.2.2 Analytical Evaluation of PARTICLE Results The output generated by particle consists of the time temperature - displacement histories of a series of allowed particle sizes. A spallation rate equal to one-half the oxidation rate at the fuel/clad node is used to estimate the total mass of particulate. A normal distribution is as- signed to the distribution of particle sizes, where the mean is given by: DMAX+ DMIN 2 3.13 and the standard minimum and maximum sizes are taken as representative of a 2-a deviation from the mean, thus: a = (D2- p)/2 3.14 The continuous distribution of particle sizes is approxiThe frequency of occurrence for mated by 5 particle sizes. the individual particle sizes (representing discrete segments of the continuous distribution) are given in Table 3-2. It is estimated that subsequent to arriving on the receiving surface, the particles will lose half of their enerby by convection and radiation to the room. The effective heat flux to the horizontal structures within range of the particulate is then calculated as: (q/A)eff 4 1 x. al l m Cp (Tp - T)f . p 3.15 = mass of particle upon arrival, (gm) m p. C = specific heat of particle, (J/gm *C) p = frequency of occurence, (1/sec) f where T p = temperature of particle of size i upon arrival, (*C) Ax. = range within which particle will fall, (cm) L = width of holder, (cm) Particles of a given size are assumed to have a uniform spatial distribution between zero and their maximum possible range. The total energy input to the receiving surface is equal to the arrival rate times the time during which the clad continues to oxidize vigorously, on the order of 8 r 10 minutes. If the fuel/clad axial location reaches the melting 95 TABLE 3-2 Particle Size Frequency Distribution y= (D MIN + DMAX)/2 a (DMA - p)/2 Frequency P + Ia yi y p -a y - 2a p + a +(7 y + 2a Diameter Frequency Diameter P - y~ - P 2 a < D <p 3 cy - < D < p - < y + 3 0.044 1 0.242 1 3 a < D < p + 2 a y + y + C < D - 3 < D < p + 0.383 0.242 0.044 temperature of ZrO 2 , spallation ceases. The temperature rise in the receiving surface (assumed to be stainless steel) is calculated by using the solution For a to the semi-infinite transient conduction equations. constant surface heat flux, the appropriate equation is (H2): T = T. + 2(/A) k Tr -x 2O4aT (q/A). k X 1-erf 2 /a 3.16 where a = thermal diffusivity of the receiving surface, (cm2/s) x = depth into surface, (cm) T = time, k = conductivity of the receiving surface, (sec) (W/cm *C) (q/A) = surface heat flux, (W/cm 2 The temperature rise is calculated using equation 3.16, where (q/A) is taken as the maximum value of (q/A)eff given by Eqn. 3.15. This maximum value of heat flux is assumed to be constant during the entire interval in which the clad node oxidizes. The receiving surface peak temperature is the variable of interest and it is obtained by setting x= 0 in Eqn. 3.16, giving: T = T=+2 (q/A3 T. + 2 (/) 1 k aT3.17 'IT .1 The values of receiving surface peak temperature thus obtained may be used in evaluating the potential for exothermic oxidation of the receiving structural surface; however, such analysis is beyond the scope of this project. For the purposes of this report, the temperature rises are employed qualitatively, so as to assess the potential for energy transfer via the levitation and convection mechanism. 3.3 ENERGY TRANSFER VIA BURNING PARTICULATE: RESULTS The energy transfer via burning particulate has been analyzed for the pool configuration case illustrated in Figure 2.10. For this case, spallation is assumed to occur for the duration of the "spike" (lower curve) in Figure 2.10. This is the period of the most rapid oxidation of the clad, and is thought to be the most likely time at which spallation could occur. The PARTICLE code is run for a series of discrete values representative of the oxidation period, shown in Table 3-3. The PARTICLE results for the values in Table 3-3 are presented in Table 3-4. The frequencies given in Table 3-2 are used, together with total zirconium oxidized in the interval and Eqn. 3.15 to obtain effective energy fluxes to horizontal structures. These values are tabulated in Table 3-5 for the first and last oxidation intervals presented in Table 3-3. It is apparent that the effective heat fluxes, which were all calculated relative to a receiving surface 98 TABLE 3-3 Input Values to the PARTICLE Program Time From Initiation of Accident (seconds) Peak Clad Temperature (*C) Peak Air Temperature (*C) 41400 1014 1005 4. 57 x 10- 42300 1117 1097 6. 35 x 10 3 42540 1656 1574 1.259 x 10-2 42610 2117 2006 1.194 x 10-2 42670 2465 2344 2. 584 x 10-2 42730 ------- clad relocates-------------- Reduction in Clad Thickness (cm) 3 TABLE 3-4 Particle Distributions for Ramp in Figure 2.3 Time From Initiation of Accident (seconds) Diameter (cm) 42300 .001665 42540 42610 42670 Time (seconds) Distance (cm) Temperature (0 C) 1.11 44.32 10.4 .002773 .475 34.44 379.8 .004617 .250 21.64 916. .007687 .160 12.13 1070. .01280 .001 .612 1170. .001744 1.0 61.11 10.1 564.3 .003041 .500 49.81 .005304 .301 36.72 1395. .00925 .202 21.54 1671. .01613 .007 .641 1767. 1.15 69.63 10. .00324 .650 69.04 419. .00584 .350 50.87 1672. .01052 .251 33.17 2079. .01896 .257 .032 2232. .001801 .001840 1.15 74.7 10. 426.6 .003384 .650 76.37 .006226 .350 57.88 1912. .01145 .251 38.49 2400. .02107 .0016 .022 2582. 100 TABLE 3-5 Effective Energy Fluxes to Horizontal Structures Time from initiation of accident = 42300 seconds Distance (cm) (/A) eff (W/cm ) per assembly 44.32 0.0 34.44 1.2 x 10~ 21.64 3.48 x 10-3 12. 13' 2.39 x 10-2 3.89 x 101 0.625 Time from initiation of accident = 42670 seconds Distance (cm) (/A)ff (W/cm ) per assembly 76.4 3.4 x 10 2 57.9 1.51 38.5 3.09 4.127 x 102 0.022 TABLE 3-6 Thermophysical properties of 5% Cr Steel a = 0.011 cm 2/s k = 0. 4 W/cm 0 C (Hl) 101 temperature of 10*C are relatively low. The spatial energy variation of the effective energy input rate is plotted in Figure 3.1. The amount of decay heat radiated and con- ducted to the upper tie plate by the tops of the fuel pins is approximately 0.103 w/cm 2, where the decay heat generation is taken as 5 kw/MTU. For simplicity, the conservative approximation is made that the maximum heat flux (for TIME= 42670) exists for the entire episode of vigorous oxidation. The maximum possible temperature rise is then calculated using Eqn. 3.17 and values of (q/A)eff estimated from Figure 3.1, and properties of steel given in Table 3-6. The re- sulting temperature rise of the adjacent tie plates for the dispersion of burning particulate is presented in Figure 3.2. The large temperature rise apparent within the first 10 centimeters should be viewed in light of the conservative nature of the evaluation, and would be lower if the analysis were extended to a larger number of particles. 3.4 PROPAGATION VIA PARTICULATE: SUMMARY The analysis presented above indicates that it is pos- sible to transfer large amounts of heat within relatively short distances via levitation and convection of burning particulate. Figure 3.2 shows that the majority of the energy will be deposited within a short distance of the burning assembly. At distances greater than the vicinity of the burning assembly, the energy transfer and resultant temperature rise are quite low. First-hand observation (B3) of 20 18 16 14 12 (q/A) eff (W/cm2 10 8 6 4 2 I 0 a I 30 40 a a-II 50 Distance (cm) Figure 3.1 Spatial Variation of Effective Heat Flux 70 240 220 200 180 DISTANCE TO ADJACENT HOLDER U o 160 w uo o 140 Q High Density Storage 0 Square Storage Array 120 -P (a 100 .- 80 - 60 - 40 - 20 - Q) 0 0 10 20 30 Distance Figure 3. 2 50 40 60 70 (cm) Temperature Rise of Steel Tie Plates due to Particulate 80 104 the spallation of vigorously oxidizing clad suggests that the use of a spallation rate of 50% of the oxidation rate is an order of magnitude too high. Additionally, the as- sumption that all clad which spalls from the rod will be buoyed by the air flow is not supported by experimental data, which show that much of the spalled metal leaves the clad in large fragments. Further investigation in this area should examine mechanisms for the formation of unreacted Zircaloy particles. Additionally, the particulate obtained in recent experiments should be sized to provide a basis for the present analysis or future revisions. 105 CHAPTER 4 ZIRCONIUM VAPOR ANALYSIS 4.1 MECHANISMS FOR FORMATION OF ZIRCONIUM VAPOR The objective of this portion of the analysis is to ascertain whether unreacted zirconium in a pool experiencing runaway oxidation could vaporize, and if so, what would be its subsequent behavior. It is believed that a cloud of zirconium vapor, generated in an oxygen depleted region of the burning fuel pool, may be diffuse or be convected to a region of higher oxygen concentration, where it could burn or explode. Experience gained with metal fires indicates that it is indeed possible to sustain dust explosions (H3). Fires of this nature have occurred in the past at manufacturers' scrap piles, where millings from pyrophoric metals have been improperly discarded. For situations such as these, the metal dust is present at (and responsible for) the explosion and ensuing conflagration. The mechanism pro- posed for fuel pool analysis is radically different in two respects: 1) The fuel for the reaction is a gaseous phase, rather than a finely divided solid, and 2) The reacting substance must be generated as a result of the oxidation process, that is, it does not preexist, as metal dust or millings. 106 Pertinent thermodynamic properties of zirconium and its oxide are listed in Table 4-1. compiled from references (B4), (Pl), (Kl), (Ml), (L2), (S7), and (Ql). It is observed from Table 4-1 that the density of zirconium vapor (pZr = 0.311 kg/m ) is greater than three times the density of air at the same temperature (pair,3578K = 0.0948 kg/m3 ). Thus, it is expected that a cloud of Zirconium vapor would fall through the rising hot air. As the zirconium descends down the channel, it will either enter a region of higher oxygen content and oxidize or cool and condense on the surrounding rods and structures. The analysis of the vaporization of zirconium consists of two portions: a part related to the vaporization mechanism and a part related to the subsequent behavior of the vapor. Quantitative analysis of this mechanism is beyond the scope of this report; however, observations related to the following aspects of the problem are pertinent. 1) Any method to assess the ability of zirconium to vaporize must entail a reaction kinetics formulation. Two possible modes of vapor generation can occur: the vapor is generated in the unreacted portion of a piece of clad undergoing vigorous oxidation on its external surface, or the vapor is generated by an unreacted piece of Zircaloy in an oxygen depleted environment. A description of the first mode of vapor generation must include the migration of the zirconium vapor through the oxidizing layer without itself 107 TABLE 4-1 Thermophysical Properties of Zirconium and its Oxide Zirconium Zirconium Dioxide ZrO 2 - Melting Point, *C 1855. 2700. Boiling Point, *C 3578. 5000. Heat of Fusion, kcal/mole Heat of Vaporization, kcal/mole 20.8 5.5 142.150 Heat of Transition ((ct + 6) , kcal/mole 0.920 Vapor Density (kg/m3 @ 3578 0 K, 1 atm 0.31073* N/A Vapor Pressure for zirconium, 1949 < T< 2045 0 K is given by: log P = - (31066/T) + 7. 3351 - 2. 415 x 104 T where P is in atmospheres, T is in degrees Kelvin. At higher temperatures, the following measurements have been made for pure zirconium: Temperature, *K Vapor Pressure, atmospheres 2450 2700 3000 10-2 3850 1.0 *Calculated by ideal gas approximation. 108 undergoing the oxidation reaction. Neglecting considera- tions of meling and boiling points, it might be thought possible that the zirconium vapor could diffuse through the zirconia matrix to the oxide-film gas interface where the reaction occurs; however, studies (M2) show that oxygen is the diffusive ion and that the reaction occurs at the metaloxide film interface. It is thus unlikely that zirconium vapor could be released at the surface by diffusing through the oxide layer. 2) The second possible mechanism, that of vapor generation by zirconium in an oxygen depleted environment, is only thought to be possible in a configuration such as shown in Figure 4.1. The reaction flame front and the depleted oxy- gen region are depicted in the figure. It is conceivable that the clad adjacent to the flame front may experience sufficiently elevated temperatures for the Zircaloy to vaporize. However, a mechanism must then be proposed which explains why the clad remains in the oxygen depleted region sufficiently long to undergo vaporization, while it is possible for the clad to melt and relocate at a temperature thousands of degrees lower than the boiling point. It is considered unlikely that unreacted clad would be subjected to heat fluxes due to chemical oxidation high enough to cause the thin outer layer of clad to sublimate. None of the computer studies performed during the course of this 109 FUEL ASSEMBLY Figure 4.1 Postulated Mechanism of Zirconium Vapor Generation 110 research have indicated that clad in the oxygen depleted region reach temperatures at which vaporization is possible. No temperatures recorded in the experiments performed at Sandia (discussed in Chapter 5) could have resulted in zirconium vaporization. 3) Despite the above observations, it is conceivable that the scenario depicted in Figure 4.1 may exist. It may be hypothesized that the vapor generated in the unreacted region could be carried up to the assembly exit via the convective air flows present in the holder, rather than sinking through the hot air as postulated previously. If oxygen is present at the assembly exit, then a diffusion flame could result, analogous to a Bunsen burner. If for some reason there were no oxygen available at the assembly exit, then the subsequent behavior of the vapor would be governed by the same plume dispersion mechanics as the particulate, as discussed in Sec. 3.3. If it is assumed that a cloud of zirconium vapor has the same ignition characteristics as a cloud of very fine zirconium dust particles, then the minimum explosive concentration of zirconium at 20*C is 0.045 mg/cm 3, as reported in reference B4. 4.2 QUANTITATIVE ANALYSIS OF VAPOR GENERATION A one-dimensional implicit radial conduction model of the cross-section of a fuel rod was developed to assess the feasibility of vapor generation. The code is based upon 111 premise (1) discussed in Sec. 4.1, namely, that the vapor is generated in the inner portion of the clad undergoing vigorous oxidation. In the code, the fuel, gap, and clad temperatures are monitored as the oxidation front sweeps through the clad thickness. The oxidation model employed is that of Chapter 2, a parabolic kinetics formulation which is evaluated explicitly for each time interval, while all other parameters are calculated implicitly. The VAPOR code, listed in Appendix D, is programmed to vary the thermophysical properties of clad elements as functions of temperature and constitution (Zr or ZrO2). A con- stant heat input is assigned, and the external surface of the clad is subjected to radiative and convective cooling. However, the code was not programmed to model the behavior of the oxidation reaction within the clad for the diffusion controlled regime, for the reasons outlined in Sec. 2.2. Restricted by limited thermophysical property data at elevated temperatures, the lack of properly characterized data for fuel and gap properties and the absence of diffusion controlled reaction kinetics, the VAPOR code can offer little quantitative information about the potential for vapor generation. The one-dimensional nature of the code (no axial conduction) ignores an important heat sink for the chemical oxidation energy during a rapid oxidation reaction. 112 4.3 ZIRCONIUM VAPOR ANALYSIS: SUMMARY The discussion of zirconium vapor generating mechanisms in Sec. 4.1 indicates that there is not sufficient data available from which to formulate quantitative models. There is no evidence to suggest that vaporization of zirconium can occur by the mechanisms discussed above. Still, it is con- ceivable that a holder basket may be blocked at a certain axial level, forming a crucible, so that heat input from the oxidation reaction in adjacent regions could vaporize a portion of the molten clad. However, blockage of assemblies would be expected to reduce the amount of oxygen available for oxidation, thereby reducing the chemical energy input and limiting the runaway. 113 CHAPTER 5 COMPUTER SIMULATION OF SANDIA EXPERIMENTS 5.1 INTRODUCTION In conjunction with this research, experiments were performed at Sandia National Laboratories in which Zircaloy-2 tubes, heated in an inert environment, were suddenly exposed to air. High-speed movies and luminosity ments of the oxidation reaction were taken. measure- Additionally, temperatures of various components of the assembly were recorded by thermocouples. This chapter describes modelling efforts to simulate the behavior of the clad tubes observed during the experiment. 5.2 SANDIA EXPERIMENTS ON ZIRCONIUM BURNING 5.2.1 Experimental Configuration The zirconium burning experiments performed at Sandia National Laboratories employed nine silicon-carbide heaters of which three or more were sheathed with Zircaloy-2 tubing. Figure 5.1 illustrates the configuration of the experiment apparatus. The assembly consists of a circular chamber 18 cm in diameter filled witn alumina fluff insulation. A trapezoi- dal duct of 26 cm 2 cross sectional area is located along the axis of the chamber. Inside the duct, nine silicon- 114 \GAS INLET Figure 5.1 Configuration of Zirconium Burning Apparatus. 115 carbide electric resistance heaters of radius 0.635 cm are arranged, several of which are enclosed in Zircaloy-2 cladding of inner and outer diameters of 0.546 and 0.484 respectively. 33 cm. The length of the heater rods is approximately In several tests, a stainless steel holder was pre- sent to restrain lateral movement of the rods. The gas flowing through the assembly, either helium or air at 1 atmosphere, was introduced at the base of the assembly, and was vented to the room atmosphere through a short length of tube. 5.2.2 Experimental Data: Air Test #4 The primary objective of the experiments was to serve as a subject for the refinement of in-pile visual diagnostic methods under development at Sandia. The need to make ex- tensive use of thermocouples in obtaining temperature measurements was subordinate to the need for maintaining good visibility. Consequently, there are few tests which yielded appropriate data. Air Test #4. The most useful data were obtained during In this test, three heaters were clad, while the remaining heaters were left bare. the following procedure The experiment used (Sl): 1) Heaters were raised to a steady-state temperature of approximately 10000C in a helium flow at 12 liters per minute (lpm). 2) Individual heater power was set at 175 watts; the 116 airflow was then changed from 12 lpm He to 12 lpm air; temperatures were recorded at 1 minute intervals. The time temperature history for the central pin in the assembly was recorded at the following axial locations: 1) approximately 3 - 5 cm below the top of the pin 2) at the mid-point of the pin The temperature histories for these locations are presented in Figure 5.2, along with the temperatures recorded at the mid-height on the alumina fluff side of the zirconia liner, which separates the airflow from the alumina fluff insulation. Figure 5.3 shows the temperature histories for the following mid-height locations: 1) the centrally located, clad tube 2) an off-center clad tube 3) an off-center bare-heating element These data are tabulated in Table 5-1. The temperature rise shown in Figure 5.2 for the time 2 to 11 minutes represents the simultaneous occurrence of three transient phenomena: 1) the temperature rise due to increased heater input power, since steady-state was attained for a lower heater input power, 1600 1500 [ 1400 * 1300 $-4 4 1200 0 1100 1000 A top of center pin 0 mid-height of center pin 900 mid-height of inside of liner 80 0 - 2 4 6 8 10 12 14 16 18 Time from Introduction of Oxygen Figure 5.2 22 20 24 (minutes) Experimental Data for Zircaloy-2 Oxidation in Air a a 26 28 - -A p L p_ -A 30 1500 1400 0 o 1300 0) a) 1200 A mid-height, center clad pin 0 mid-height, off center clad pin < mid-height, off center base heater 1100 1 1000 0 2 4 6 8 10 12 14 14 a 16 1 12 20 20 22 22 Time from introduction of oxygen (minutes) Figure 5.3 Experimental Data for Zircaloy-2 Oxidation in Air A 24 26 2 28 119 TABLE 5-1 Experimental Data for Zircaloy Oxidation in Air Time From Introduction of 02, minutes 0 1 2 3 4 5 6 7 8 9 10 11 12 12.25 13 14 15 16 17 18 19 20 21 22 23.5 24 25 26 27 28 29 30 Temperature, *C TOP PIN MID PIN BOT PIN LINER MIDCLAD MIDBARE 1035 1039 1057 1136 1287 1396 1438 1440 1490 1435 1471 1518 1551 1571 1558 1531 1505 1486 1467 1453 1440 1422 1460 1471 1525 1460 1385 1329 1299 1258 1225 1196 1119 1119 1130 1204 1293 1348 1390 1397 1384 1375 1390 1428 1454 1466 1473 1480 1473 1453 1440 1422 1409 1416 1448 1467 1518 1447 1366 1305 1275 1234 1199 1164 932 937 948 997 1057 1102 1136 1147 1142 1139 1139 1153 1164 1171 1182 1193 1199 1187 1176 1168 1159 1153 1171 1171 1199 1159 1096 1052 1024 986 948 911 868 868 874 905 970 1041 1096 1119 1130 1142 1183 1182 1222 1227 1246 1252 1252 1246 1240 1222 1216 1204 1216 1222 1246 1246 1217 1176 1136 1102 1069 1041 1080 1085 1110 1193 1264 1317 1354 1354 1342 1342 1360 1390 1422 1427 1435 1441 1441 1428 1406 1390 1378 1378 1396 1406 1454 1427 1348 1293 1240 1199 1164 1130 1041 1041 1057 1130 1199 1252 1284 1281 1275 1269 1293 1326 1360 1360 1366 1372 1375 1372 1354 1342 1330 1330 1336 1342 1375 1366 1305 1252 1205 1170 1136 1102 120 2) the temperature rise due to energy liberated by the oxidation reaction, 3) the melting of the stainless steel spacer. It has become apparent that the coupled nature of these phenomena limit the usefulness of the data. The temperature rise during the time 20 to 24 minutes is attributed to the fact that the heater input in the operable heaters which remained, was increased during this period. The following observations have been made (B3) with respect to the temperature measurements. The temperatures measured by the thermocouples may be considerably lower than the true temperatures at a given axial location. cies may arise due to two mechanisms.. Discrepan- Since the thermocou- ple is in contact with the outer surface of the clad and not welded to it, the thermocouple is insulated by the ZrO 2 oxide film, which is a much poorer thermal conductor than zirconium. As the oxidation front moves into the clad, the insulating thickness of the oxide layer leads to steeper temperature gradients in the clad. If the oxidized clad beneath the thermocouple fragments, the insulating characteristic is further exacerbated by the presence of a gas filled gap between the fragmented oxide and reacting metal oxide film interface. The second mechanism which may lead to erroneous readings is the fact that the thermocouples are not attached to the clad surface, but merely rest 121 against the clad surface when the apparatus is assembled. Bowing of the clad surface, which is evident in the movies taken of the experiment, could add large resistances to the thermocouple-clad system. It is conceivable that the bow- ing mechanism could result in apparently large, rapid clad temperature variations. COMPUTER SIMULATION OF EXPERIMENTAL TESTS 5.3 5.3.1 Introduction The experiment modelling effort focused on two sets of control volume boundaries. In the first approach, the be- havior of the rods, canister and gas were modelled, using the CLAD code. The second approach employed the VAPOR code mentioned in Chapter 4, in an attempt to simulate the detailed processes taking place in the oxidizing clad. How- ever, this effort was later abandoned because the VAPOR code does not have the capability to calculate the diffusion controlled behavior of the oxidation reaction front within the clad. 5.3.2 Experiment Simulation Code: CLAD The stand alone SFUEL code used to analyze open frame spent fuel holders was revised in an effort to simulate the zirconium burning experiments performed at Sandia. However, the essence of the code is still that of its predecessor: the array of fuel rods is modelled as a single component. 122 In the new code, the model has been extended to calculate the temperature and oxidation potential for a central rod. The principal components modelled by the CLAD code are depicted in Figure 5.4. The major characteristics of the code are as follows: 1) the original air temperature calculation scheme has been replaced by a donor-cell method; 2) axial heat conduction is calculated for the fuel bundle only; 3) in the new code, all temperature calculations, except evaluation of fuel temperature, are performed implicitly; thus, the stability criteria of Sec. 2.2.5 applies to the fuel/clad nodes; 4) the rods are assumed to remain intact while undergoing vigorous oxidation; 5) the gas mass flow and heater power are input and constant; 6) the principal mode of convective heat transfer is by forced convection; assuming that natural convection is negligible, the energy equation is uncoupled from the momentum equation and solved independently; 7) radiation, conduction and convection modes are included; 8) convection heat transfer models are: forced convection for rod bundle, constant Nusselt number and 123 ALUMINA FLUFF ALUMINA OUTER SHELL INPUT GAS FLOWS Figure 5.4 Experiment Simulation Model 124 natural convection from the exterior of the canister; 9) the oxidation thickness is evaluated explicitly, using the parabolic kinetics formulation of Sec. 2.2; 10) The diffusion-limited oxidation reaction is governed by the Sherwood number calculated using the ChiltonColburn analogy for heat and mass transfer. The natural convection heat transfer correlation used for the exterior of the canister is the same as that for the global SFUEL code (see Appendix A). Examination of the ratio of assembly length to hydraulic diameter (approximately 25) shows that the velocity profile is not expected to become fully developed at any point in the assembly. For the Reynolds numbers encountered in this experiment (on the order of 10 % 100), no experimental data on heat transfer coefficients for turbulent developing flow parallel to an Even the use of the Reynolds ana- array of rods was found. logy to obtain a value of Nusselt number from the available fully developed friction factor data for laminar flows parallel to rod arrays is not strictly correct. Kayes (K3) gives the following solution for Nusselt number in the entrance region of a circular tube: Nu = Nu + K.[(D/x)RebPr1 1 + K2[(D/x)Re bPr]n 125 where for uniform wall heat flux and developing velocity distribution: Nu, = 4.36 K1 = 0.023 K2 = 0.0012 n = 1.0 The fully developed friction factor for laminar flow parallel to a rod array with the same pitch to diameter ratio as the experiment rod configuration is obtained from the work of Sparrow and Loeffler (S2). Use of Eqn. 2.7 yields an approximate value of Nu= 10.02, for 10< Re<100. Evaluation of Eqn. 5.1 shows that the Nusselt number does not vary appreciably over the entire rod length. On this basis, a constant value of Nu= 10.02 is employed in the CLAD program. It is quite possible that the variation of Nu with entrance length for a rod array may be substantially different from that of Eqn. 5.1: the fully developed value is thought to be a good approximation given the lack of experimental data. The value of 10.02 is also used for the Sherwood number when calculating the mass transfer coefficient for oxygen diffusion in nitrogen for the diffusion-limited oxidation reaction regime, as described in Sec. 2.2. Simulation of the experiment required that the code be provided with accurate thermophysical property data for the fuel, clad, air and canister materials. Thermophysical pro- 126 perties for which experimental values at elevated temperatures were not available were input to the code as nominal values, or the upper limits of available property-temperature correlations. A listing of the CLAD program is pre- sented in Appendix E of this report. 5.4 EXPERIMENT SIMULATION: RESULTS The CLAD program was run for the heat input and mass flow rates of Air Test #4, performed at Sandia (see Sec. 5.2.2). Results for the steady-state calculations (prior to introduction of air) compare favorably with the time= 0 experimental data. A plot of the predicted vs. observed temperatures for the heatup of the mid-point of an offcenter clad rod is shown in Fig. 5.5, and tabulated in Table 5-2. However, the results for the oxidation portion of the code compare poorly, with the code predicting temperature rises several hundred degrees higher than those observed. The divergence of the observed and predicted values is attributed to the causes discussed in Sec. 5.2.2, and well as the limitations of the phenomenological oxidation model employed in the clad code. Temperatures recorded during other experiments have been observed to exceed 2400*C. 5.5 EXPERIMENT SIMULATION: SUMMARY Experimental tests were performed for the vigorous oxi- dation of Zircaloy-2 in an air environment. For reasons related to experimental assembly configuration and the test 1000 + ++ 800 600 measured values for mid-height, off center clad 400 value predicted by clad 200 1- 2 4 6 8 a a 10 12 M 14 16 a 18 a 20 a 22 a 24 Time Figure 5.5 Comparison of Pre-Oxidation Experimental Assembly Heat-Up. 1 26 ~ A 28 is 30 128 TABLE 5-2 Comparison of Pre-Oxidation Assembly Heat-Up Time (minutes) Measured, Off-Center Clad Rod, Mid Height Predicted Average Rod Mid-Height 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 22 24 26 28 30 93 260 343 443 510 610 650 704 760 810 848 871 882 898 916 927 932 936 940 945 950 955 960 982 982 128 253 369 476 571 654 724 781 827 863 892 914 931 945 955 962 969 973 977 980 982 984 988 988 988 129 objectives, limited data of a quantitative nature have been obtained. Computer simulation of the test results is suc- cessful for the non-oxidizing heat-up period of the assembly; however, satisfactory results for the oxidation results could not be obtained. The cause of this discrepancy is attributable to both insufficient modelling of the assembly (assumptions 1 to 10, Sec. 5.3.2) and the phenomena described in Sec. 5.2.2,which are not amenable to quantitative analysis. There is insufficient evidence to indicate that the oxidation model is itself at fault for the deviation between experimentally measured and predicted values. Future work in computer simulation of the oxidation effect must be based on an appropriate database. It is re- commended that oxidation models, other than the parabolic model, be developed to incorporate non-uniformity of oxygen concentration and variable temperature capabilities. 130 CHAPTER 6 PROPAGATION OF ZIRCONIUM BURNING: CONCLUSIONS The SFUEL code has been revised and extended to calculate the propagation of runaway zirconium oxidation in a spent fuel pool following loss of water. Analysis of cylindrical and high-density storage arrangements have engendered the following conclusions: 1) The high density storage configuration is most susceptible to propagation of zirconium runaway oxidation. 2) The minimum decay time for spent fuel to be stored adjacent to recently discharged fuel (%o90 days decay time) is approximately two years (decay power < 4.0 KW/MTU), if propagation of zirconium runaway oxidation is to be prevented for a high density storage configuration. 3) Levitation and dispersion of particulate does not play a major role in the propagation of energetic zirconium oxidation. It can at most increase the likelihood of burn- ing in fuel holders immediately adjacent to that containing fuel experiencing runaway oxidation. Even for this situa- tion, the effective heat flux due to particulate is small relative to the radiated and convected heat fluxes from the assembly experiencing the zirconium burning. 131 4) There are no feasible mechanisms for the generation of zirconium vapor in a spent fuel pool in which assemblies are vigorously oxidizing. In the event that zirconium vapor is generated, it will either suppress the natural convection in the holder, reducing chemical energy release via oxygen starvation, or sink through the rising air and subsequently burn at the oxygen diffusion front. Alternatively, if the zirconium vapor descends to lower, cooler portions of the fuel pool prior to oxidizing, it may cool sufficiently to condense. 5) Fuel relocation as modelled in this report acts to contain runaway oxidation in adjacent fuel holders by removing the heat source to lower, cooler portions of the spent fuel pool. 6) Fuel relocated to the base of the holder may obstruct the baseplate hole causing extinguishment of the oxidation reaction via oxygen starvation. It is suggested in NUREG/CR0649 that coolability of spent fuel during a postulated accident, prior to runaway,may be maintained for a nearly complete drainage of the pool (water blocking baseplate holes) by drilling holes at various elevations in the lower part of the holders. It is apparent that this design change could exacerbate the vigorous oxidation situation by: a) allowing hot reaction gases to enter the inter-holder spaces, resulting in larger convective heat fluxes to adja- 132 cent fuel holders; b) providing more oxygen for the oxidation reaction and thereby reducing the regions of diffusioncontrolled reaction; and c) negating the ability of relocated fuel to cause extinguishment of runaway oxidation via obstruction of the baseplate holes. 133 APPENDIX A HEAT TRANSFER COEFFICIENTS AND SKIN FRICTION CORRELATIONS USED IN SFUEL (AFTER NUREG/CR0649) TABLE A-i Equations Used for Nusselt Number and Skin Friction Coefficient* Flow Geometry 1. Forced Convection Parallel to a Flat Plate Turbulent Flow Laminar Flow (NuD )l= 0.332 Re 0.5Pr 0 (cf) 1= 0.664 Re -33- -0.5 Re (NuD 2 = 7.54 + 0.0234 ReDPr (cf)2= ReD - 24 Pr 0.6 DH 5 x10 5 -- ] (NuD) 2 = 0. 023 ReD 0 8 Pr 0 .4 (cf 2= 0.0014+ 0.125 Red 0.32 /ReD ReD 3000 Poiseuille Solution 0.8 "Power Law" Solution [El] Blasius Solution [El] 2a. Forced Convection Between Parallel Plates (Applied Outside Fuel Element) e (c f) 1 = 0.0592 Re -0.2 5 x 105 Re (N)00296R uDl [El] 3000 Correlation [El,S4] TABLE A-i (Cont'd.) Flow Geometry Laminar Flow Turbulent Flow 2b. Longitudinal Forced Convection Between Parallel Tubes in an Infinite Array (Applied Inside Fuel Element) (NuD)2 = 8 Assumed to be same as (c) 2 = 25 (2a). /ReD 3000 ReD Sparrow-Loeffler [S5] 3. Free Convection Past a Vertical Plate (NuD Gr 3 = 0. 36 Gr 025 1x10 Correlation , DH Pr=0.71 [El] (NuD Gr x- u.33([rD H 0. 116 Gr l x10 9 , Correlation Pr= 0.71 [El] *To obtain Nusselt number for a particular condition, take the maximum of (NuD)2 ' and (NuD)3. and (cf)2 ' (NuD)l' To obtain skin friction coefficient, take the maximum of (cf) 136 APPENDIX B SFUELlW INPUT, OUTPUT AND PROGRAM LISTING CONTENTS page Section B.l SFUELlW Input 137 B.2 SFUELlW Output 140 B.3 SFUELlW Program Listing 142 B-1 Sample Input Listing 139 B-2 Sample Output Listing 141 Table 137 SFUELlW INPUT, OUTPUT AND PROGRAM LISTING SFUELlW INPUT B.1 The SFUEL program was developed on a Control Data computer operating system. To implement this program on the Information Processing System's Multics computer, it was necessary to incorporate a restart capability in the program. A namelist output file, under the heading $RESTRT, is used rather than a BLOC data transfer to facilitate program debugging and allow variables to be altered prior to restart. The first card of the SFUEL1W input contains two entries: IRESRT = 1 0 Program begins calculations from time = 0. Program is being restarted with a $RESTRT data file. TTEST Time at which restart file is written; file is updated during each printout. The format for the first card is: 1 2 00 0 0 or I10,I10. The remaining input is identical to that for the original SFUEL program, and is entered under the namelist heading $INPUT. The interested reader is referred to NUREG/CRO649, 138 Appendix D,for a list of the names, dimensions and units of the input variables. For the code presented in this appen- dix, the following variables described in NUREG/CR0649 have been rendered inoperable in the course of revising the SFUEL code. ASINK DAMP IPLOT TIMWON CSINK DMWTR PRMAX VENT CPS EPS TIMWOF Additionally, it is recommended that the following variables only be input with these specified values: FSTR = 1.0 Use of IBLOCK = 3 other than these values will lead to computational instability. A sample input listing is shown in Table B-l. This particular case corresponds to the curve in Fig. 2.7, run for the modified SFUEL code. 139 TABLE B-1 Sample Input Listing 32000 1 PWR CYLINDRICAL RACK $INPUT DELT = 50. 366. FL NPRNEW = VROOM = 36. 4.25e9 FSTR = 1.0 WS = 21.4 IBLOCK = 3. = 28.0 ICHEM = 1. WW XW NCEND = 1. XWL = 40.6 NSECT = 6. XS = 0.0 NROD = XWW = 3.69 RHOW = 7.82 CPW = 0.30 FMULT = -1.0 0.4614 NDECAY = 6. RCI 0.418 NASS (1) RCO 0.475 SKBOT = RF 0.401 TRDELT = 0.0 UL = TRMAX = EPW POWO = 289. ROWS = TIMEX = 43200. NPRINT = FDECAY(l) = 0.351 0.460 1, 5 * 4 0.0 2232. 0. 6000. -36. 2.72, 3.76, 5.05, 5.90, 8.20, 11.04 For variables not appearing in the above list, the default values are used. 140 B.2 SFUELlW OUTPUT The user has the choice of long or short output format, depending upon the sign of the input quantity NPRINT. The short output format provides output for the hottest pool section only. A short-format output is shown in Table B-2, these results corresponding to the input of Table B-1, with an elapsed real time of 10 hours. The variables shown in Table B-2 are defined in Appendix B of NUREG/CR0649, in order of their appearance in the output. The following variables have been replaced in the printout: has been replaced by TAI+l, which is the tem- TS perature at the interface of the airflow control volumes (*C). has been replaced by TA3I, which is the tem- TAVE2 perature of the interholder air control volume interfaces (*C). is not used. 12 In addition, the output variable IS has been redefined as: IS = 1 oxidation reaction is kinetics rate-limited 2 oxidation reaction is diffusion rate-limited 3 clad is completely oxidized 4 clad has melted and relocated *** KIT- i I 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 i8 19 20 DPFRAC= 9.342E-03 DGFRAC= 5.839E-03 TIME = 3.600E+04 TRMAX= 3.230E+02 *** TROOM= 1.000E+01 PROOM= l.013E+06 DELT= 500.00000 TL(I) J TB(J) TA4AVE(J) PA4AVE(J) FOXAVB(J) GNIAVB(J) GNI(J,1) GNI(J,2) GNI(J.3) 1.122E+01 1.133E+01 1.147E+01 1 2 3 4 5 6 1. 122E+01 1.263E+01 1.264E+01 1.265E+01 1.267E+01 1.271E+01 1.276E+01 4.376E+02 4.375E+02 4.372E+02 4. 370E+02 4.369E+02 4. 369E+02 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 9.697E+01 1.771E+02 1.4 19E+02 1.044E+02 6.460E+01 2. 196E+01 6.544E+00 3. 374E+01 3.668E+01 3.825E+01 4. 135E+01 4.393E+01 1 .000E-10 1.000E- W i .000E-10 1.000E-10 1.000E- 10 1.000E-10 -2.005E+02 -2.OOOE- 10 -2.OOOE-10 -2.000E-10 -2.000E- 10 -2.000E-10 1.165E+01 1.186E+01 1.209E+01 1.234E+01 1.260E+01 1.285E+01 1.310E+01 1.332E+01 1.351E+01 1.366E+01 1.375E+01 1.378E+01 1.374E+01 1.363E+01 1.344E+01 1.317E+01 1.220E+01 1. 151E+01 1. 153E+01 1. 151E+01 1. 159E+0i 1.171E+01 .- 6 TR 2. 101E+01 3. 129E+01 4.473E+01 6.075E+01 7.897E+01 9.903E+01 1.205E+02 1.430E+02 1.660E+02 1.891E+02 2. 118E+02 2. 336E+02 2.541E+02 2.727E+02 2.891 E+02 3.029E+02 3.137E+02 3.212E+02 3.230E+02 2.958E+02 TAVEI i.679E+01 2.396E+01 3.439E+01 4.785E+01 6.404E+01 8.261E+01 1.031E+02 1.251E+02 1.481E+02 1.716E+02 1.949E+02 2. 177E+02 2.394E+02 2.595E+02 2.775E+02 2.930E+02 3.057E+02 3. 152E+02 3.200E+02 3.050E+02 11 TAiI+l 2.038E+01 2.9 1BE+01 4. 112E+01 5.595E+01 7.332E+01 9. 286E+01 1.141E+02 1.366E+02 1.598E+02 1.833E+02 2.063E+02 2.286E+02 2. 494E+02 2.685E+02 2.853E+02 2.994E+02 3.105E+02 3. 176E+02 3. 125E+02 3.050E+02 TA31 2.018E+01 2.846E+01 3.958E+01 5.331E+01 6.932E+01 8.726E+01 1.067E+02 1.272E+02 1.483E+02 1.695E+02 i.904E+02 2.103E+02 2.290E+02 2.459E+02 2.607E+02 2.730E+02 2.825E+02 2.883E+02 2.813E+02 1.413E+02 12 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 TW 1.730E+01 2.489E+01 3.579E+01 4.975E+01 6.645E+01 8.551E+01 1.065E+02 1.289E+02 1.522E+02 1.759E+02 1.994E+02 2.221E+02 2.436E+02 2.633E+02 2.808E+02 2.957E+02 3.077E+02 3. 164E+02 3. 199E +02 3.003E+02 TAVE3 1.678E+01 2.358E+01 3.334E+01 4.583E+01 6.078E+01 7.785E+01 9.666E+01 1. 168E+02 1.377E+02 1.590E+02 1.801E+02 2.006E+02 2.200E+02 2.379E+02 2.538E+02 2.675E+02 2.785E+02 2.865E+02 2.900E+02 2. 727E+02 13 RCT 1.538E-04 1.538E-04 1.538E-04 1.538E-04 1.538E-04 1.538E-04 1.538E-04 1.538E-04 1.538E-04 i.538E-04 1.539E-04 1.539E-04 1.539E-04 1.540E-04 1.542E-04 1.545E-04 1.549E-04 1.552E-04 1.552E-04 i.543E-04 FOX 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.3001-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 2.300E-01 EGEN= 2.301E+09 ELINRS- 7.461E+04 ECONV3' 4.007E+05 2.939E+04 ECHEM= ELINRB- 8.207E+04 ESTAIR= 7.546E+05 EFUEL= 4.358E+08 ECONCS= 9.305E+05 EREMDR= 9.090E+05 ESTR= 0.000E+00 ECONCB= 1.041E+06 EROOM= 0.000E+00 EHOLDR= 1.293E+08 ECONVi= 1.645E+09 ESINK= 0.000E+00 ERAD= 8.631E+07 ECONV2- 0.000E+00 ELOSS= 0.000E+00 PGEN= PLINRS= PCONV3= 1.366E400 PCHEM= PLINRB= 9.773E-01 PSTAIR=-5.045E-03 PFUEL= -5.904E+00 PCONCS= 2.731E+01 PREMDR= 3.425E401 PSTR= 0.000E+00 PCONCB= 2.948E+01 QRooM= 0.000E+0() PHOLOR=-2.256E+00 PCONVI= 6.040E404 PRAD= 3.415E+03 PCONV2= 0.0(X)E+00 OSINK= 01-OSS= 6.391E+04 8.453E-01 1.548E+01 0.000E+00 0 000E+00 IS OM 000 000 OU0 'rrrlrrlr'IN3AI'IbViSI'SI'YONI'GGNI'CGNI'ZONI'IGNI'riliI liwrMH'CVMH'ZVMH'dW3iH' ViH4ZVSH'IVSH'WbH'ZVbH'lVbH'inOH $WHOH'SbGH'NOH' VSH'iVH'inowNoloxoglioaxoolexoolxoolqolND 'MNO /ibiS3b/ IS113WVN ZOIND'IOIND'XVWIND'GIND'SAVIND'INO'NID'19'Nlid3D'9d3D6VdOD 'ZdOD'IdOD'dOD'HISJ'WOObJ'IdJ'OXOJ'SAVYOJ'AVXOJ'XOJ'OJ'M3NJ ,iwrvisovj'J'bIS3'biVIS3'NNIS3'WOON3'bGW3b3'GVb3'Md3'id3'Sd3 'ld3'3d3'SS013'SbNI13'8bNI13'13'HOlOH3'N393'13nJ3'CAN003'ZAN003 'TAN003'SON033'80NO03'W3HO3,xazaxvwioovbjiombasbaiba lebanVidG'qOdG'COdG'ZOdG'lOdG'YVWdG'OVbAdO'da'biMWG'iOVJIG'HO 'INDG'XVWOO'OVbJDO'ZI30'AI30'YI30'01130'il3G'30'dWVG'MMD'tAD SSO'NNISO'SO'bO'MdO'SdO'WOObd3'YOdO'OdO'lNdO'M3NdO'IdO'NO3d34d3 'ldSV'NNISV'SV'bV'ldV'Hd-IV'INDV 'JV'SOV'OV'GV'3WI.L')(VWII'il3G /IHIS3b/ iSI13WVN iO8XOO 'GXOD 'SIND 'IND 'SdOO 'GAVXOJ /MOIJ/ NOWWOO THdlV'C3AVI'iwrmilmi /0018/ NOWWOO NOISN3WIG NOlSN3WICI NOISN3WICI NOISN3WI0 NOISN3WI0 (TZ)SGI'(CZ)bIVO'(CZ)THdlV (T)ldA '(T)GVlA '(I)IdX '(09VIX (iz)x (iz)iwrmi '(9'IZ)Mi '(9'IZ)SI '(9'TZ)Hi (IZ)10111 '(IZ)li '(5)31ili '(9)03WIi '(IZ)10001 090 (9)10ig.L '(9)81 '(9)OAVqVI '(9)3AMI '(L) Vi NOISN3WICI OLCO (9'TZ)CVI '(9'IZ)ZVi '(9'IZ)IVI '(9'TZ)Oi3AVI NOiSN3WIG 090 (9 1Z)OAVI '(9'lZ)OZ3AVI '(9'IZ)Z3AVI '(9'IZ)013AVI NOISN3WICI 050 (9'IZ)13AVI l(L)qVOHb 6(TZ)CVOHld '(IZ)ZVOHb NOISN3Wia O io (IZ)IVOHH '(9'lZ)1Ob '(9'IZ)Mb '(9)dOllb NOISN3WIG Offo (9) 108ib ' (9' 1Z) Sb ' (9' 1 Z)Hb "(I Z)lb NOISN3WIG OZO (9'IZ)AV03ab '(TZ)GNOob '(iz)ioiiob '(iz)iob NO ISN3Wia (9'; 1Z)W31-lob 6 (9) ioisob 1 (9) sob ';(9) ab ' (I Z)b NO ISN3W 10 Olzo 000 (9)AvnVId I(L)nVid MMMcl I(L) Vcl NO ISN3W 10 OLZO (I Z) CVd '(TZ)ZVd '(IZ)lVd '(9'IZ)WXO '(QiON '(9)SSVN NO ISN3W 10 09ZO 9 11 1 Z)SI '(9'IZ)iGNI '(9'lZ)ZGNI ';(9'TZ)IGNI '(9)AV3301 NO ISN3W 10 OLZO (9'iz)iwrMH '(9'TZ)CVMH '(9'IZ)ZVMH 'M VIH NO ISN3W 10 09ZO (9'IZ)ZVSH (9'TZ)lVSH '(9'lZ)ZVHH '(9'IZ)IVNH NO ISN3W 10 05ZO (9'IZ)NGH 6(9) VSH '(P9)IOGXOD '(L)exoo '(iz)xoo NOISN3WI0 WO (1 49) OINO'(P9)COINO '(P9) ZO INONO ISN3W 10 (P9) TOINO ' (L)SINO ' (9) SAVIND " (P9) IND (P9) SdOO NO ISN3W 10 OQO OZZO (9)SAVXOA '(9'TZ)XOA '(TZ)HMd0J '(IZ)AV030A (i z)Hme aj NOI SN3W 10 Ol ZO (9)NNSIVA '(9)SIVA '(9)MSVA '(9)SOVA 1(91 1z)mba NOI SN3Wia OOZO (9'1 Z) SbG "(I Z) IN ' (9) Gba ' (C9)OdCl 4 W9) OdO NO ISN3W 10 OLTO (P9)ZOdG '(P9)TOda '(P9)dG "(Q NiMWG ' W9) INOG NO ISN3Wia 0910 (9) A13 0 ' (9) MO ' (9) SO ' (C)ZAA303 6 K) I AA300 NOI SN3W 10 OLTO (9) SO "(9) Cvxv 1 (9) zVxV 1,(9) 1vxV 1(9) IdMV '(9)MV NOISN3Wia 0910 (9) IV ' (9) 1 dSV ' (9) S V ' (9) b V ' (9) AV I (9)OV '(9)SV NOISN3WIa 0510 OfT 10 *H31VM ON HIM OCTO 100d 13ni IN3dS NMd NO HMS V NI 3SIN 3HnivH3dW3I SNIISIa WVU90Ud MTqanaS US OTLO ooLo 0890 OLo HOI*q '(03S) 3WIiHOI/ GVIX VIVO OZLO /9*01 6501/ six lix vivo /509 I*o 163*1 163*1/ GX 'NOMWI1 'JOMWIi 'INdNi VIVO /63*1 163'T '*0009C '5ZO*/ XVWNi 'i13ONi 'XVWII '8WS VIVO /0 o69o /93151*1 '1 '9C 'TZ/ XVWNd 6*04C 6*1 "05 '*0/ iOldl 'HiMWO 'M3NNdN 'iN1NdN 'N VIVO 410VA10 '1-130 4dWVO VIVO 'C-319,L 'Z-35 *T 'Z-365*z 'Z-3z6oC 'Z-38L* 0590 /*0*5 '5-R *C '5-3oL*5 ' -310*1 ' -319*1 ' -396,z ' -351*5 ' -350*9 'C-31 *l 'C-3ZC'Z 'C-39L*C OL90 0990 'Z-3CC*5 '*0*5 65-366,1 '5-308*L ' -38Z'l ' -3C6*i 090 ' -359*9 'C-3Wi 'C-36Coz 'C-39L*C 'Z-365*z 'Z-R6* 'Z-38L* 'Z-3CC*5 IL399 o6 'L311C*9 6*045 '83551'C '83SL5oi IL3551'C IL3555#1 '939LL*L '93z65*z '53 9*8 '53*1 0190 0090 'C-319*L 'Z-35 *T OC90 OZ90 6 8 L 9 5 C z o65o '3*1 '3*1 'Z3*1 '13*1 '03'T 41-3*1 1 NMdOJ 'Nmeoi 4ioooi Iiinwi 'AV330N VIVO '00*1 69T/ '*0 '*CSZ/ dOINX 'iOSNX '01 VIVO 0850 Po oL5o 0950 0550 050 OC50 OZ50 0150 OM 0 O OM /ZIO* "086 'ZI-3L9*5 '6L*Zl NOONWS 'DWS 'DIS 'MOHH VIVO /599 'zS*L 1 *oi ' C*Z/ SOHN '1OHN 'JOHN 'NOOOHH VIVO /5*9 '931LS*Z '93CTO*1 1 59z65i i*C/ OOHb 'VN 'Od 41d VIVO /51 'z* IL, ILV xvwN 'Mc13 Id3 'Sc13 VIVO /Co 'Le 4C99* '%*/ IcI3 'Od3 'MdO 'SdO VIVO /oWi IoWi 'oW 'LWU XOdO 'INdO 'IdO 'NOOdO VIVO L IND lo.LndNI 'dOiNX 'IOSNX '01 '10001 'NOONWS 'MOHN 'SOHN 'IOHU o6 o 'JOHN 'NOOOHN 'OOHb 'AV030N 'XVWN Iiinwi 'AV3303 'Mc13 OPO 'Id3 'Sc13 'IcI3 'Od3 'MdO 'SdO 'XOdO 'INdJ 'IdO 'NOOdO 5 oL o lmmx 41mx lmx laix lsx lix lax mm 4WOObA 'AM 'in'NOMWIi 'JOMWI.L,'INdN1 'XVWNI '1130U '03WI1 05 0 'XVWII 'SWS 'SMOH 'AN 'OON 'ION 'XVWNd 'OMOd '103SN 4OONN z 6M3NNdN 'ININdN 'ON30N 'SSVN 'N '.LO-IdI 'W3HOI 'NOO-181 'NiSJ I 'IA 4N.LMWa 'IOVAla 41130 4dWVO 'NNISO 4NNISV /.LndNI/ IS113WVN iwrmmx6MMX'IMX'MX'81X'MSX'SX'XVWdX'IdX'HiVdX'ONX'IX'dOINX 2 'IOSNX'SX'X'XGZM'IdMM'XaMM'MM'IdSM'DWSM'SM'TNM'SOM'WOOHA 2 6IN3A'Idn6inlriMi'MigONNISI'NNisigriSi$Si'iNd I.L'OldNi'IdHi 2 lowoouilwooidilxvw 8ilrINI'1130111' li'Oi4XVWi4iOili4M3NIllriii4ii 2 'NOWII'AOWII'03WII'33WIi'3WII'XVWII'1003i'10igi'M3NSI'risi 2 '81'OC3AVI' 3AVI'OZ3AVI'Z3AVi'013AVI'13AV.Llrl Vi'OAV Vi 2 63AV Vi' vilriCv.L6CvilrizvilzvilrilVi'lVi'TANWS'ANWS'NOONWS 2 $13NWS'ONWS40WS4XJWS'CVM3WS'ZVMJWS'W31AWS' VIAWS 2 4ZVSAWS /ibIS3H/ IS113WVN ivsiws'zvNjwslivHjws$iwrjws' vejwslviwslawsg vxvwslzvxvws I 'IVXVWS'IdMVWS'MVWS'.LVWS'IdSVWS'SVWS'NVWS'AVWS43VWS48VWS'SIS I 'SMO 8'XOHNOMOH i$SOHN'WbOHb'OOHH'M3NOHH'IOHH'AOH 8'NOOOHH 2 4OOHH /IHIS3*d/ IS113WVN VOHN4CVOHtl'ZVOHH'TVOHH'A'd' 3H'C3H6Z3N'13H'3 1'10 l'OO 8'N3H4l3H 2 ,vH'iwrmblrimb4mb'IN3Ab4doiib'1091b'NNisbsrisblsbliwrmbb 2 I.LNblmsNb'NNisHblsHNblwoobbIriNb'IW3bb'CW3Hb'ZW3Nb 2 'IW3HbeNbNblinob'NNlsiblssoibiiblib'OiNlb'WON30b'AV33ab 2 4aNOObI.Lo.Li3b'iob'W3Hob'10183b'SOb'3b68b4NNisvb'SSOIVb'b4Md 'AvnVidgnVid'HiSd4HlViSd'WOObd$XVWHd'NOW3Nd'aVHd'OMOd'Od 'H3wnNd'NlWd'SIJNlld'BHNlld'id'HalOHd'N30d'13nAd'WON30d'CANOOd 6ZANOOd'IAN03d4S3NO3d'SONOOd'W3HOd'VOd'3AV Vd$ Vd'iVd 'ZVd$lVd$IOIWXO'NWXO'OWX06WXO'i33SN'aONN'VINNdN'M3NNdN'INIHdN gIdN'dNgCiON'iON'TWNgAV03aN'ON30N'SSVN'N4rixvwlilxvwN'iiN$N 2 2 2 2 2 2 0OU1 0011 G0IN*03803'.' I d-SM*~SM=1 VXVWS z13a;czI3a=8VWS (o*1*3*1±SA) JI Z130.c (MMX+IdMM) =GVWS 0611 0911 IdSM*'1dSM=1VWS OLIr 0911 0511 011 0OM 0111 0111 0011 X13G,1cdMM*Ct1=IdMVWS C0*1*b3'1Si) Al X130*Zl30aO l=1dMVWS X13J'.MM** 1 =MVWS X13OG'1dSMW '=IdSVWS Xl 30*SM** 1=SVWS 00UN* (I 31d* I Ob-030h'03 ) *' I d=OVWS 00bN*cd*c Id=3VWS aObN*cXI30*,03bc ld**l=bVWS X130*cIdMM=XGMM (01*O3*IdSJ) Al XOZO=XOMM 0601 0801 oLoi X13a0czI3a=Xozo 56 0901 MMX+IdMMA=ZI30 (*0oov0ilMMx) 0501 iI IdMM=Z130 0101 56 01 09 (5'o 3*ISJ) 31 tAX+MM=Z1 30 1/ (IdSM-MM) =MSX 0o1 0N01 SX'* 1+SM=IdSM (1-N) /1A=X13J 06 0660 MX*C l+MM=IdMM 0101 0001 0860 oL60 0960 og6o S3fllVA IVIIINI 3 3 MM=SMA 06 010o (o03N*SX) 31 0T6o 0Z60 0160 0060 0680 0=1UVIS I (o0)AdooaH IIV3 3 £8 0± 09 (Otn3*1ldl) 11 £8 01 09 (0~b3*IbV1SI) Al (±ndNI'S)Ov3H 58 18 01 09 3 1=3031 3 666 01 09) (I1b3*031) Al Z8S 58 '18 ((S)A03) Al 3 0880 oLgo 0980 0580 08y0 ( Mot (1)31111 (666=Pua'o68V5')aV3H 18 0=3031 08 3 0180 061S 1 09 V3 18 01 09 (1~3*IHS3'81)) 0080 o6Lo OSLO oLLo o9Lo ogLo 0=1OldN 1=Ibv1I 3 /ooiT '81 (0) ' "'001 '81Z/ XVWdA 'XVWdX( 'XVWA 'XVWX Viva /HOlslZ )V3dHOT/ GVIA viva NO0T 'Ifl1Vb3dW31H~t " OVIO 1 3 3 Vt1T oZL OUT i Od-=NIWd oi=xvw± OILI 3flNIINOO 51t ooLi C't-Loo/'1+±d3/' 1)/CC)8v= (r) NNSIVJ o691 ( t-1d3/ 1+id3/ 1)/Cr)8v= Cr) 8±vd 0291 (*1-Md3/* +((r) IdSV*Sd3) /(r)mv) /(r)mvNisj= (r)msvI zii OL91 CI -Md3/*1+(SOV~cd3) /r)mv) /r)mvcusj= (r)so vi 0991 (r) SSVN*cX130*SJtA*'T=S3V 0S9 1 ZIT 0± 09 C'O3N'SX) JI 0~91 C 1-Sd3/ 1+3d3/ 1)/Cr) sv.tisj= Cr) sovi o001 33sN'i=r 511 00 OZ91 i C -Md3/'Z) /XGMM=MMVJ 0191 C't-1d3/ 1+CXOMM).'Md3) /XOZO) /XaZo=IMVd 0091 3flNIINOO 011 0651 ex*' Cr) SV)AldO).I0Hb= Cr) so oS51 Cr) SSVNcX13G*~ CMM-*.MM- CMY,% C USJ-e 1)3'c* -IdMM) ).cdMM) *MdO).CM0Hb= Cr) MO oLM (r) SSVN3.X130* (SM),SM-IdSM).cldSM)).cSdO),cS0H1=(Cr) so 0951 .L3sN'1=r 011 00 0551 IX*aXI30.'Z130*.cdO*Ic1Hb=10 0'7t mmx~'s*+CMMX-1MX) +18=18 OMS MMX'sS'+CIJ3SN) A130= C±J3SN) A13a 0Z51 MMX-1MX+ I)A13a= CI) A13Q 0151 3flNI±NOO 501 0051 Cr) A130+18=18 06W I r) SSVN).c MMX+1dMM)= Cr) A130 OPT 1 I3SN'i=r 501 00 OLMi '0-18 o9 1 I-N=IWN 05 i1 VIN~dN=dN 0'71 (INIbdN) SSVI=VINIdN oPi 1 3flNIINOD 001 0rM MMX*Z13a*5,+Cr)0vxv=Cr)0vxv (1D3SN'03*r) ji 01'rI CMMX-IMX)*z13G+Cr)0vxv=Cr)0vxv Cvb3*r) ji Cr) SSVN)AVXVWS= (r)0£xv 0 0 y1 offi 1(r) SSVN).cZVXVWS= Cr) zvxv Offl I r) SSVN.c1VXVWS= Cr) tvxv oLM MMXz13*5+(r)ev=(r)9v C33SNw03or) ji 09l(MtAX-IMx)*zi30+ Cr)sv= Cr) sv CI?3or) j i Oscl I r) SSVN~c8VWS= Cr) 8v Orcl 1 r) SSVN*IVWS= Cr) iv OUT Cr) SSVN'*1dMVWS= Cr) 1dMV OM 1 r) SSVN*MVWS= Cr) mv 01C Li r) SSVN*ICdSVWS= Cr) 1dSV ooL t (r) SSVN*SVWS= Cr) SY o6zi ( r) SSVN*OVWS= Cr) Dv o8Zt Cr) SSVN*C.VWS= Cr) jv Uzi (r) SSVN).tVWS= Cr) uv o9Z I .L33SN'i=r 001 00 05Z 1 Z13i 38x=qivxv 0Tzl 1dMM*cIdMM-Z13G*.ZI30=LVXVWS C0'i~b3*iSJ) Al OU1 Zi 30*cMtX=LVXVWS IdSM*IcdSM-MM*cMM=ZVXVWS 0ZZ1 146 WSMG=SMG*DELX/RA RHOO=PO/(RA*TO) F0=.23 CPO=FO*CPOX+(1.-FO)*CPNI FPL=SMG*RHOO*FL UPL=SMG*RHOO*UL APL=FPL+UPL PW=4.*WS+2.*PI*RCO*NROD DE=4.*SMAXA1/PW EL=FL*(1.+2.*SMB)/PI QDENOM=COS (SMB*FL/EL)-COS ((1 .+SMB)*FL/EL) EGEN=0. ECHEM=0. EFUEL=0. ESTR=0. EHOLDR=0. ERAD=0. ELINRS=0. ELINRB=0. ECONCS=0. ECONCB=0. ECONV1=0. ECONV2=0. ECONV3=0. ESTAIR=0. EREMDR=O. EROOM=0. ESINK=0. ELOSS=0. PCONCS=0. PCONCB=0. QSINK=0. QASINK=0. QLSINK=0. QLOSS=0. QOUT=0. DELTO=DELT TIME=0. X (1)=0. DO 120 I=1,NM1 X(+1)=I*DELX Q(I)=(COS((X(I)+SMB*FL)/EL)-COS((X(I+1)+SMB*FL)/EL))/QDENOM 120 CONTINUE DO 130 I=1,NM1 TL(I)=TO+1.E-10 TLTOT(I)=0. QCL (I)=0. QCLTOT(I)=0. 130 CONTINUE DO 140 J=1,NSECT TB(J)=TO+1.E-10 TBTOT(J)=0. 1740 1750 1760 1770 1780 1790 1800 1810 1820 1830 1840 1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2110 2120 2130 2140 2150 2160 2170 2180 2190 2200 2210 2220 2230 2240 2250 05zz OLz OUZ OZLZ OILZ OOLZ 101dl=ldl O=IdN 01=NNISI 013*1=XVWIND 'O=XVWDO 'O=XVWdO Od=WOOldd NNUNOO 591 NNUNO) Z91 *o=(rl)ZT8H *o=(rll)ZVSH *o=(rll)ZVMH 099Z 059Z 09z o0z OZ9Z 019Z OOHN=WKH*d OA=WOOUA oi=wMli 069Z 089Z OL9Z oi=(rli)ivi oi=(ri)zvi oi=(rli)0i oW OW Mz oo=(rll)W3HOb eo=(rll)AV030b *o=(rli)wxo 0T-3*i+oi=(ri) u 01-3'1+01=(ri)si 0T-3'T+01=(rli)mi oi=(rli)13AVi oi=(rll)Z3AVi oi=(rll)0AVI 085Z oL5z 095Z 055Z 05Z ocsz ozsz 015Z 005Z o=(rgi)si *ooS9/*i=(rli)ioN 009Z o6gz 09 z os z Oz oc z OM OOZ oo z offz o9cz oM 09cz 05cz Oiz OUZ ozU OlU OOR o6zz o8zz oLzz o9zz IWN'I=l Z91 00 103SN'i=r 591 oa 3nNIINOO 091 *o=(Pr)INOG *o=(zlr)INDa oo=(ilr)INDO oo=(zlr)ioexos 3nNIINOO 191 0T-3*Z-=(Pr)IN!) 01-3*t=(zlr)IN!D 01-3*1=(Tlr)IND 191 01 00 (0*10*51ndNI)AI 103SN'i=r o91 oa 3nNIINOO 0 1 (r) AV3301 oj=(r)SAVXOA ldV=(r)3A0Vd oi=(r)MOVI oi=(r)qvi ,o=(r)eb *o=(r)iosib so-(Nioisob oo=(r)eob LVT OLZC o9z£ Os?£ Ozc 4,it =flwj i / 1*013',, =Md3 it/ £0013',i =.Ld3 it/ C*013 c f 1 =Sd3 it / £*013', =1d3 it/ P013'i =3d3 it/ £*013 z ',1 =MdJ i / P013'i, =SdO it / £0l3'ii =XOdO i / P013 I 1', =INdO i / C*013',, =1dO i / C*013dT', =NO~dO ,/)ivwboj o16 r O~cdO±IX '±0S)4X '01 'N03>IWS 'MOHbJ 'SOHbI '1OHb z Ozzc'J0Hb 'NOOOHN 'OOHb 'AV330N 'XVWN '±1flwA 'Md3 'Id3 'Sd3 I o1U£ '1d3 '3d3 'MdO 'SdO 'XOdO 'INdO '1dO 'NO~d) (oC6Vr9)3±;liM 061C (103SN'w1='(r)o3wI± '(r)SSVN 'r) (oZ6V19)3±Iw' 021£ C/,,3w1 SSVN r ,/)ivwb0d oI6i OLI £ (o16Vi9) 3IlbM 091£ (C*013',, =MMX it/ £013'1, =IMX 1 051C / £01 3'it =MX it/ £013'i =S.LX i/ £*013', =SX £ OTI£ / £13'il =IX it/ £o13'i, =SX it/ £013'i =MM 11 z Mc1 / £013' =SM it / P013', =W00OIA it / C*013 ' , =±N3A it I 011£ P 0131 1 =infi / £013'i =iNdbit / £013', =XVWb± it + 011£ / £0T3' ±13GU± it/ P013' =N0MW1± / £o13' =AoMWli 6 11 11 11 001C P 013',, =XVWI± it / P013', =SWS it / Po13'i, =SM0b it 8 060£ / 013'i =A i / P£0i =03N i / PO13'ii =lob 11 L 080£ / 013', =XVW~d it/ PO13'i, =OMOd it/ 51i',t =103SN it 9 OLOC / 1 i =CObN it/ 51' =M3N~dN it / S I',,=IN I dN 11/ Si 1 S 11 090£ 11 =0N30N it/ 5 iu =N it/ 5 i0 =101d I i/ S i, =W3H1I 1 050£ Si',,1=N3071131i / £013'i, =bISA it / £013't, =I11 c O17oc / Pov3'XS/£'o13'x8/P0T3',, =b-1MW0 it z Ooc it'1',=~n / £'o13',, =.I130 it / P013'i, =dWVO it 1 010£ / £013'ii =NNIS3 it / Po13d1',, =NiNISV iI/ Lv' 11111)ivwboj 0067 ooo£'lx 'SX 'MM 'SOM 'W00NA '.LN3A 'in 'INdN± 'XVWbLL '±130U.L $NOMWIJ. £ 0661 6A0MWli 'XVWIJ. 'SWS 'SM0Oi 'Jb '03Ni 610b 'XVW~id 'OMOd '±J3SN Z o86z'GObN 'M3NbdN 4±NIbdN 'QN3JN 'N 'L0ldi 'W3HO1 ')iJ019I 'i±SA '1:1 1 oLR 'bI±MW0 '.LOVAIG '113a 'dWVO ')ANISO ')INISV '(1)31111 (0o6V79)3Iim 691 0S6z oM6 iflNI 3±INIM3 016z 0161 oo6z 068z 088Z ±1flwj(A3al)bMdaJ=(A3Ol)AV3A 891 AV330N'I=AO01 891 00 L91 691 0± 09 ±1flwjc(Aoa i) bMeA= (Aoai) AV03ClJ 991 AV030N'I=ADaI 991 00 L91 0± 09 (001*19*ObN) Al 0L81 691 0±i 09(e0±1±1fliwj) ji 0=(I)±ON t=±N3AI (XVW~d3S*W00d) JI =±N3A I 08£z I 1=£±ON (Cob3sNO0181) l o-£±ON (Z~3D01sI0lbo*o3*SX) 31 0=(Z)±ON 0981 058Z 0OR8 0t81 0181 1=(Z)±ON 009Z o6Lz oLLz 09LZ 01=1 dNI 81'T 149 4 E1O.3 / " KMAX= ",15 / " NDECAY=", 5 15 / " RHOC= ",E1O.3 / " RHOCON=",E1O.3 / " RHOF= ", 6 E10.3 / " RHOL= ",E 10.3 " RHOS= ",E1O.3 / " RHOW= ", 7 / TO= E1O.3 / " SMKCON=",E10.3 ",E1O.3 / " XKBOT= ", 8 E1O.3 / " XKTOP= ",E 10.3) WRITE (6,4940) TCOOL I FDECAY"/ 4940 FORMAT(/" WRITE(6,4950) (1, TCOOL(I), FDECAY(I), 1=1,NDECAY) 4950 FORMAT(14, 1P2E12.3) START TIME LOOP 170 DO 174 J=1,NSECT DO 172 I=1,NM1 TAVElO(IJ)=TAVE1(I,J) TAVE20(I ,J)=TAVE2(IJ) TAVE30(I,J)=TAVE3(I,J) 172 CONTINUE TA4AVO(J)=TA4AVE(J) 174 CONTINUE DO 178 J=1,NSECT DO 176 L=1,3 AGNI=AXA1 (J) IF (L.EQ.2) AGNI=AXA2( IF (L.EQ.3) AGNI=AXA3( GNIO1 (JL)=-200.*AGNI DP01 (JL)=-PO GNI02 (J,L)=GN 101 (J,L) DPO2 (J,L)=DPo1 (J,L) GNI03 (J,L)=200.*AGNI DP03(JL)=PO GNI04 (J,L)=GNI03 (JL) DPO4(J,L)=DP03(J,L) 176 CONTINUE 178 CONTINUE 180 DO 550 K=1,KMAX KIT=K PCONV1=0. PCONV2=0. PCONV3=0. PSTAIR=O. IDIREC=-1 START LOOP THROUGH SECTIONS OF POOL 190 CONTINUE INPUTG=0 IFLG=1 761 CONTINUE DO 325 J=1,NSECT TRMAX=8500. IF(IRESRT.NE.1) NPRNTA=40 IF(IRESRT.NE.1) NPRNEW=5 IF(IRESRT.NE.1) NPRINT=+1 START DETERMINATION OF AIR PROPERTIES 3280 3290 3300 3310 3320 3330 3340 3350 3360 3370 3380 3390 3400 3410 3420 3430 3440 3450 3460 3470 3480 3490 3500 3510 3520 3530 3540 3550 3560 3570 3580 3590 3600 3610 3620 3630 3640 3650 3660 3670 3680 3690 3700 3710 3720 3730 3740 3750 o6z OW otz o9z o5z O Z Oczq ozz Olzq ooz o61q xozm;c(rli)iwrMH+ VMV:,c(rll) VMH=INM ZCZ EVMV/ 0(0MM),.W31JWS+ (YOMM-CVMV) ;cCVMAWS) =CVMAWS VMV/(XGMM) dW31H+(XQMM-CVMV)),t(rll)iVMH)=(rll)CVMH ZCZ 01 OD (1*3N*r) ji XGMM+ VMV=CVMV (i33SN*b3or) ji xazm-(r)IdMV=CVMV 5T-3*i=(rll)CVMH (*O*b3*(rll)iVMH) JI WII)CON1 l(rl0HGH Z loxoj 1(r) vxv 1 vmjws l(rll)CVMH ' 3H 'WOOU '(N3VGI)CVOK8 I '(r'Novei) vi l(rli)mi 'HO '19 'IJ 'ONX 'HlVdX 'I)dONdV 11VO MMX*OZ=HO 09lq ((rlI)CGNl l(rll)bOH 'OXOJ l(r)Cvxv z oLiq 'W31JWS 'dWUH 'CH 'WOUd '(N3V91)CVOHb l(rNov9i)Cvi I 091 4(rli)mi 'HO '19 'IJ 'ONX 'HiVdX 'T)dObdV 11VO (I*b3*r) ji 05lq ((rll)CONI l(rll)bGH Z oliq loxoj l(r)Cvxv liwrjws l(ri)iwrMH 'C3b 'WOObd '(N:)VGI)CVMI I ocl (r'Novei)Cvi l(i)twrmi 'Ha '10 '13 'ONX 'HiVdX 'I)dONdV 11VO iwrmtix**Z=HG OQ XGMM=XGZM mmx=iwrmmy Ozlq oll 00147 imx=iwrmmx (i)u=(i)iwrmi 09oq 05oq (i-rli)mi=(i)iwrmi ozz OQ 01 09 XGZG=XGZM o60 080q OLoq O Oq OiOq OzOq olo 000 o6GE oLff o96C OZZ 01 09 (103N*r) ji 03bial*HiVdX+(Z/(33NIGI-1))*IA=ONX X13a*(5*-il)=HiVdX Z/(33Niai+i)+i=amji Z/(03NIaI-I)+I=N3V8I OT8101*1 1+ (Z/ (33b 10 1-1) ) ;cN= I 1+11=11 0=11 ITT 01 09 (I'3N-r)ji I o56C oq6C Tqz ((r) vxv*(r)0XV*WOONd**Z)/ I (l+*b3*03Niai) ji ((r6N)CVI*Vb*ID*19),cdOINX)-(N)CVd=(N) Vd (1-*b3*03HICII) 31 oC15C (11300N) /X1303.c (r) iVXV*d3*WOObd=V3d lNdO*(OXOA-*I)+XOdO*OXOA=dO oz6C (r) oiff vxv* (r) CVXV3..WOObd3.c *Z) / oxo9+(Pr)IND=ID oo6C ME ME USE (d39)SGV=Vd39 INd3*(Vr)IND+XOd3*OXOD=d3D (Pr)IN5*(OXOJ--I)/OXOJ=OXOD OIZ (r)SAVX03=OXOJ ME (r)3AV Vl=(r'i)Cvi ME ME ME OZSE OISE Ooss (r)3AMd=(I)EVd OOZ Olz 01 00 wOObj=OxOj woobi=(r'N)EVl Idn+,o=(N)EVd ooz oi oo (*oi9*(Pr)IND) Al MU oSLC 5 Z 01 00 oLU OLE (*O*.Ll*331JI0I*(Pr) IND) Al 3NnSS3bd WOOH 3AOQV 3bnSS3Nd SI d SN3alOH N33MI39 'i 13NNVHO OST o9S QSS'i 05q OC5 OZ5 ~o6z oo5r o6T17 wooi= Cr'N) ZV± Idl+*0= (N)ZYd oSz o± 09 (,o*± (z'r) 1ND) JI 0± 09 (o±*J1*3l31IaI(z'r) IND) 31 0~6Z 01 09 (003*SX) Al5q 13010H- ONV 3fl13flb±S N33MI39 4Z 13NNVHO (ro)3V)tr X~(ZNV+AOdCN~ (P*3Nr) i CWon~- Cr 'N)0lv)*'d+L4AN0Jd=CAN0~d Osq1 CAN0~d=CN03d (t~b391DI) JI OL r o9'Pr (r) Cvxv*. Cr) CVXV*WOO~d* oq (r)Lvxv.,c(r)CVXV-*WOObd* * ) (,1 0x09= CL'r) ±08x09 v*N.c1*1O.lG - *Z)/( r'N) LVl*Vc I D*IDd~)X)+ z 1fl inc ( (Cr' N)Li.v)/W00O1d) '.cWS+ (N)CVd-(r)3AVqVd=(L'r)da (*o±i9*Lr) 1ND) JI Ozqq I 3flNIiNO3 L'iZ TqZ 01 09 (TWN1II)JI 3flNI1NOO 0tT ((r)Lvxv/(r' I)AVi*r/(xzm).twrws+vmv. Lcvmws)+ z ((r'move1)LCvi- (rIai1 V)*dV r vv r VVWOd I (VU*1I9.cI9) -03bilaI*r' OL3AV±/W00bd cOWSM(NVl)CVd=(0mII)LVd 01 qT oqr o6L'i o8Lci oLL'r 091q OSL'T OqcqCr' I)OL3AV/((r' I)oCAVI- 0=HdIV I*I=HdlV (0OOLC*3D*0U) Al 0*=HdIV 5Q~ Cr' ;)AVI)~V~d+tllviSd= diViSd 3flNI1NO3 ZZ 3flNIIN0o ML C(r) Cvxv~c Cr) LYXV*W00bd) / I CVub*I~19) -33bl i U i C~ 3AVI/W00bd*c9WSM-(COVa I)LVd= (amj )LVd Cr'1) 0L3AV1/(C r'1) 0L3AV1- Cr'1i)L3AV±) sVd+HIV1Sd= IY1Sd xOzm/ CxazM- Cr) 1dMV); Cr'i) LVMH=(r'1) iwMH C(XOZM- Cr) ldMV) * /mmx) / Cxamm'* Cl) tiI vo) = (r'Il) LVMH (r It)i3V*b d*Ob*()1HdIV=C() lI VO oe z/C( r11)C3AVI+ (rI +i) ]AV) =C(r Iam I)LCvi 30Cii*1+ CZ/C031dIUI-t))*N=l IWN't=ll LLR oa Cwoob±' Cr) 3AV1 VI'MMX'X3'±13a'r) 1NVHO liVO 3flNI1NOO LZZ (i-rli)mi= (iii) twfmi 3flNIINOO IIt ZZ 01 09 tb3*r)JI r'ivei)Lv±-(r' I)L3AVI±'*Z=Cr'ami'u)Lvj OzcqCINM+ I oL~r oLi C r'1)0L3AVi/Vd+Vd0D*Z) / IVH+V~d+C(r INova I)Lv±vdD** Z)=(r11)C3AVI ooLq Ci) tiwixazIW Cr' i) wrMH+ Cr' i)m±*Cvmv* r' I) LMH=1VH TST 0805 oLoS 0905 0505 005 OC05 OZ05 0105 0005 o66 086 oL6 o96 in* (((r'N) ZVi*V I) /W0011d) *DWS+ (N)ZVd-(r)3AV Vd=(zlr)dG (,o*iD*(zlr)IND) Al 3nNIiN03 oLz ((r)zvxv/(rll)Z3AViicVAWS+ z (r'Nove i) zvi- (r I ami i) ZVi) *HdIV) * ( (r) zvyv).c (r) ZVXV*W00lJd) / 1 (T8* I D*ID) -018101* (r, 1) Z3AVI/W00bd*5WSM- (NOV 8 1) ZVd= (OM A1) ZVd *O=HdIV lol=HdIV (*000 *3D*ZRI) JI O*Z=HdIV 59Z (rll)OZ3AVI/((r'I)OZ3AVI-(r"I)Z3AVi)*VOd+lllViSd=blViSd (rNovei)zvi-(r'I)Z3AVI**Z=(rlamji)zvi (INM+ I (rll)OZ3AVi/VOd+VdODkc*Z)/(IVH+VOd+(rl)4ovei)ZVI).cVdOD).c'Z)=(rll)Z3AVi Z9Z o56 o /((r)IdSV+(r)mv)).czv ijws+*z/vjws=vjws o6 z/ (r I 0 61 i) bi).c (r) IdSV+ (r) mv) *z/ (r) I dSV+(r)mv) (r' 1) MIH+* Z/IVH=iVH (r- 1) ZVbH+'Z/ I MM= I NM loxoj l(r)zvxv Izvbjws l(r,'I)ZVbH 'Z3b 'WOObd '()40VSI)ZVOHH I oo6 '(r'Novei)zvi l(rli)bi 'HO '19 'IJ "ONX 'HIVdX 'T)dUdV IIVO w6 Z oz6 (zi l(rll)bOH o6P Z9Z 01 09 (0*T*b3*biSJ) JI 08P *z/((r)IdSV*ZVSJWS+(r)mv).tzvmjws)=vjws oLB (r I i) si* (r) IdSV* (r 1 1) ZVSH+(r' i) mi* (r) mv).,. (r 1 1) ZVMH=IVH (r)IdSV;,.(rll)ZVSH+(r)mv*(rll)ZVMH=INM 51-3*i=(rll)ZVMH (*O*b3'(rll)ZVMH) JI ((rll)ZONI 6(rll)bGH Z loxoj g(r)zvxv Izvsjws 11(rll)ZVSH 'Z31d 'WOObd '(NOVSI)ZVOHb I ozsq l(rl)iovs0zvi I(r'i)si 'HO '10 'IJ 'ONX 'HiVdX 'I)dObdV IIVO ((rll)ZONI l(rll)bOH Z loxoj l(r)zvxv Izvmjws g(r,'I)ZVMH 'Z3b "WOOlid '(-AOVSI)ZVOHH I o6Lq l(rl'Aovsi)zvi l(rli)mi 'Ha '19 '11 'ONX 'HlVdX 'I)dObdV IIVO (MM+IdSM)/ZVXVWS=HG 33bl0l*HiVdX+(Z/(03 liai-i))*ij=ONX X13G).c(5*-11)=HlVdX Z/(03bl0l+l)+I=aMAI Z/(33bl0l-l)+I=)13VSI oCLIT 03bl0l*ll+(Z/(O3blGI-1))*N=l IWN'1=11 oLz oa ((r)zvxv*(r)ZVXV*WOOHd**Z)/ I ((ri)zvi).cvd*io*io*ioeNx)-(T)ZVd=(I)ZVd (T+*b3*O3bl0l) Al 098 OSP OP M olsq oosq oSLq oLLq o9L oSLq oqLq ozL oiL ooO o69q ((r)zvxv*(r)ZVXV*WOObd**Z)/ OM oL91 099 1 ((r'N)ZVI*Vb).419*10*dOINX)-(N)ZVd=(N)ZVd (1-*b3*33bl0l) A[ (1130)AVb)/Xl3a*(r)ZVXV;dO*WOOHd=VOd INdO*(0X0A-*0+X0dO*0X0J=d3 059 oxoo+(zlr)IND=ID 01M Mq (dOD)SSV=VdOD INdO*(zlr)IN9+X0dO*0X09=d39 (r)3AVqVd=(I)ZVd 05Z o6Sq (z'lr)IN9*(0X0A-*T)/0X0J=0X09 09Z (r)GAVX0A=0X0J (r)3AVqVl=(rli)zvi OZ917 0191 009q oss oL5q 09Z 01 00 woobj=oxoj 153 C C C C 2 +(XKTOP*GI*GI*RA*TA2(NJ))/(2.*PROOM*AXA2(J)*AXA2(J)) IF (GNI (J,2) .LT.0.) DP(J,2)=PA2(1)-0. 1 -(XKBOT*GI*GI*RA*TA2(1,J))/(2.*PROOM*AXA2(J)*AXA2(J)) GOXBOT(J,2)=GOXO PCONV2-PCONV2+GCP*TA2(N,,J) CHANNEL 1, WITHIN ASSEMBLY 290 IF (GNI (J,1)*IDIREC.LT.0.) GO TO 325 IF (GNI(J,1).GT.o.) GO TO 300 PA1(N)=O.+UPL TA1(NJ)=TR00M FOX(N,J)=FROOM GOX(N)=FOX(NJ)/(1.-FOX(NJ))*GNI(J,1) GI=GOX(N)+GNI (J,1) GO TO 310 300 PA1(1)=PA4AVE (J) TA1(1,J)=TA4AVE(J) FOX(1,J)=FOXAVB(J) GOX(1)=FOX(1,J)/(1.-FOX(1,J))*GNI (J,1) GI=GOX(1)+GNI (J,1) 310 IF (IDIREC.EQ.-1) PA1(N)-PA1(N)-(XKTOP*GI*GI*R A*TA1(N,J)) I /(2.*PROOM*AXA1(J)*AXA1(J)) DO 320 li=1,NM1 I=N*((1-IDIREC)/2)+II*IDIREC IBACK=I+(1-IDIREC)/2 IFWD=I+(1+IDIREC)/2 XPATH=(Il-.5)*DELX XNC=FL*((1-IDIREC)/2)+XPATH*IDIREC GOX(IFWD)=GOX(IBACK)-OXM(I,J)*IDIREC IF (GOX(IFWD)*GOX(IBACK).LE.o.) GOX(IFWD)=0. FOX(IFWDJ)=GOX(IFWD)/(GOX(IFWD)+GNI (J,1)) FOXAV=(FOX(IBACKJ)+FOX(IFWD,J))/2. Gl=.5*(GOX(IBACK)+GOX(IFWD))+GNI (J,1) CALL APROP(2, XPATH, XNC, FL, GI, DE, TS(I,J), TA1(IBACK,J), 1 RHOA1(IBACK), PROOM, RE1, HSA1(1,J), SMFSA1 , AXA1(J), 2 FOX(IBACK,J), HDR(I,J), IND1(I,J)) CALL APROP(2, XPATH, XNC, FL, GI, DE, TR(I,J), TAl(IBACK,J), 1 RHOA1(IBACK), PROOM, REl, HRA1(I,J), SMFRA1 , AXA1(J), 2 FOX(IBACK,J), HDR(I,J), IND1(I,J)) GCP1=GOX(IBACK)*CPOX+GNI (J,1)*CPNI GCP2=GOX(IFWD)*CPOX+GNI (J,1)*CPNI GCP=(GCP1+GCP2)/2. GCPA=ABS(GCP) CP=FOXAV*CPOX+(1.-FOXAV)*CPNI PCA=PROOM*CP*AXA1(J)*DELX/(RA*DELT) IF (HRA1(1,J).EQ.0.) HRA1(I,J)=1.E-15 WK1=HRA1(1,J)*AR(J)+HSA1(1,J)*AS(J) 5090 5100 5110 5120 5130 5140 5150 5160 5170 5180 5190 5200 5210 5220 5230 5240 5250 5260 5270 5280 5290 5300 5310 5340 5350 5360 5370 5380 5390 5400 5410 5420 5430 5440 5450 5460 5470 5480 5490 5500 5510 5520 5530 5540 5550 5560 5570 5580 0809 oLo9 Cr' 1)ZVMH+ ( (r'i)m.- 0909 00o9 0 09 OZ09 0109 0009 o865 oL65 Cr, 1)C3AVi) ( XQZM- Cr) ldMV) * (r' 1)CVMH= Cr' 1)Mb ('i (r, mi)- **Cr, i)si)' *91s'.(r) msvd=ms~ib 5c iwm+ -lm~orim~ 5U 01 09 £c* i) iwrmi1,c!9i s.Acwrvj)A,-x azw.,'t(r,i)1WrMH-= (1))10 C')iob-iwrMb=(I) T OU 0100 (P'3Nr) ji 1wrmuO+((i) iwrmi- Cr' )i3AVi)*xazm*(r i)1wrMH=1wrmb 0665o (~,~i) iwrm1- * cr,'i) mi)-*coi svwrvj=iwrmbb Sz XaMM=xOZM mmvj=iwrvi oS65 0t65 oC65 szc 01 09 xazo=x ozm itMvj=iwrvd 01 6g 9ZC 01 09 (I'3N-r) ji TWN '1=I 0I 00 i33SN'1=r o5C 00 oo65 o685 0885 oL85 S1N3W313 Hfl1Jflis 1VOUt1I3A 01 S3xfl1A 1V3H 0985 0585 0 3 3 ZZ 01 09 (1-*O3*33blaI) 31 3flNI1NOO 5Z oq85 OC8s 3 ozSS 01 SS o6LS OBLS OLL5 Cwoou- Cr'N) lvi) *cINd3) CI 'r)IN9+XdO*(N) X09)+IANO~d=1ANO~d IANO3d=INO~d C1~b*!1dI)31 (i)xo9= (1'r) 109x0 0085 o- ((r) I vxv* (r) I VXVktW00d* - )/((r'1) I v±i I V~I9cI *I 9*cNX) -C(1)1Vd= (I'Ir) da (oo 11(1'r) INO) Al inw(((r'N) I V±i z Vb) /WOObd) *cows+ (Cr) I vxv3., (r)I VXV*CW00d)r* /)C(r'N) tv± V t VIc9I.cl9d01)4X)+(N)1Vd-(r)3AVl Vd=C1'r)da C*oi9,(i'r) 1N9) Al 09L5 05L5 3flNI1N03 0Z oqL5 (Cr) tvxv/(r' i)13AVI*VAWS+ Z oCL5 ((r INva i)I vi- (r'o I i)I VI) *~HdV) * ((r) I vxv*c (r) I VXV*~WO0~d) / I oZL5 (vb).i9).io) -33blak r'I)I3AVI/W00OId'A9WSM- CN3VSI) IVd-(aMdl) l~d (C(r) Ivxv), Cr) IVXV*W00~d*~ 0 ) / I (r1 )1v )cNc1*19.i e )-()IV =()1V (10b3* 3N 1aI) JI 0O=HdIV I=Hd1V (*OOOC30*I3b) Al 01ILS OOLS o695 0*Z=HdIV SIC 0895 (r'I)oI3AV±/(Cr'I)ot3AV±-(r',)Imv±I)*V~d+NVISd=ulViSd (r'rwN)I3AV±=(r'N)IVj. (IWNb3*)Al z/((r' 1I) I3AVI+AVi) =(r 'Nove ) lvi 0995 C11M+ I Cr'I) 013AV1/V~d+Vd39) /(jVH+V~d+AVI*~Vd9) - Cr' I) 3AVi (r'i-i)I3AVl=AVi. (1*3N*)dI Cr) 3AV~Vl=AVJ. ZI o 95 z/()s*(( o005 )s*( or/ Cr) sv* (ivsws-iv~uws) +vdws=viws )IVH-( )NLs( 11 N)+V=V ri Cr) sv3. (Cr'') VSH- Cr')I) VNH)+TMM=INM OZ95 01 9S ZI 0095 0±09 (O'3*USA) Al oz/(C r)Nv~iv~dws+ Cr) sv'ivsdws) =v.ws (r' I i) bi*(r) Nv~c(r I i) I VH+(r I ) s.L*r) sv* (r ) I VSH=iV H o655 0959 0o99 0SS9 OtrS9 0£ S9 0159 005S9 ooS9 09'19 0O=H1VdX O t, 9ztr 01 09 rrr=r rrr=rr sn i 0±0 CP±i*rr-r)) ji Cr)m01je 11vo 9zqi 01 09 r= rr enr 010 Cv19CiDrr-r)) ji 9zq i 010 (0%L1C1+r)G1ND) JI O£'r 01 09 (1:)SN19or) ji r=rr r i+r=r 9zq o=r r 01 9 00~9 060 oS80 OM£ 0909 oS50 otr£ oM£ OM£ o=r 1 3flNI INOO '7Z o l+r) 81No+ (r) 8 1NO) *5S' (r) GA 1ND (C'r) IN9-(z'r) 1N9-C1r) 1N9-Cr)51N9=(t+r)51N9 103SN'i=r 'TZn 00 0o=(i)nfVid 0= CI+1J3SN) 5X09 .0= (T+1D3SN) 81N9 0o=(I)Gx09 0=(T) SIND i01 0 (01i9*ofldNt)JI 3flNI1N03 oL' 01M£ 0o£0 o6z9 SM01.A SSVW 'S31A3dOU M01A 3SV8 0899 tTt 01 09 oLz9 09Z9 OSZ9 otrr9 oZZr9 ol Z9 00Z9 0619 0819 oLt19 0919 05t 9 05C ((r,' I)N-Cr' i) I3AV) * (r) sv* Cr' 1)1VHH+ ((r, o uL- (r'6)Z3AV1) * sbb-((r' I)u-(r'Ii)I3AVI) *(r)tv* (r'Ii)ItYH= (r'1)N Cr' I)sba+(r' Imba=(r' i)mta Cr' I)sb+ (r I) MO= Cr' 0)mb SU£ 01 09 (0O3N*SX) Al bis '*(C(r) sv* Cr' Ii) IVsH+ (r) IdSV*. Cr'II ) ZV SH) -= Cr' Ii) sba msb-sb~3-isI'*(((r i)si-Cr' I) 3AVI)* I I) I VSH+ ((r' I)SI-Cr' 4)Z3AV1) * Cr) IdSV'* Cr' I)ZVSH) = Cr' I)sb Cr) svx. (r' 1y'* Cr' i)u±)* 91s*Cr) s ov j= s~du (iy*~ Cr' Ii) siCxazM- Cr) ldMV) *' r'I) VMH- Cr' i)ma= Cr' I)mOO (Cr i ) mi- r II) C3AVI) *(X0ZM- (r) IdMV) *(r 11)£VMH+ (r i)mb=V OZt9 at 19 0019 0609 Ii)m LM 01 09 C±3SN*3r) ji Ek*rim~i*irj(~sj* t usj*. Cr) mv'S Cr' 4)ZVMH- CXOZM- Cr) ldMV) * (r' 1)£VMH-= Cr' I)0mb twrm~b-msu?+((r' l)m1-(r' I) 3AVI)*N±SA(r)mv* I 090t 050L 0O=WON3Od 0=H3wflNd oLtr 0± 09 C0±oifDoicNl)31 HfNliN03 TC7 00oL o0oL TofsSThd 'S3l±H3d0~d M013 3SVS OZOL o1oL 000£ 0669 0869 oL69 o969 o569 ot,69 oC69 oz69 ot 69 0069 0689 0999 oL99 0999 0599 0~99 M~9 OZ99 0199 0099 o6L9 o9L9 OLL9 09 £9 05L9 o1TL9 oUL9 ozL9 ot L9 ooL9 0699 0899 oL99 061 0± 09 1+0I101 3flNIINOJ Zi ((i+r) flVid+ Cr) flVid) *5*= (r) AVlid lVidG+ (r)nfVid= Ct+r)nfVid a '' (vxv/ Cr) 3AV TV''z/ (Cr)8v).Otv83ws+ Cr)±v'.'qv±3ws)+ + (r) 'rvi- (I+r) 17Vl) *'HdIV) *c(qTVXV).c 1 VXV*.W00Ud) / CVb*~ I 9k' 19) -=flVidO *0=HdIV I=HdIV (*OQO 3Doq3H) AlI 0 Z=HdIV (r)vi- Cr)3AVqVl*cZ= C14r)qvi~ Cr) 0AV Vi/(C r)0AV1ViCr) 3AV 1V1) cV~d+ I VISd= I ViSd Ci~m+Cr)OAV Vl/ 1 V~d+dO'.NI 9+dJ9*cZ) / iVH+V~d+NlidJ9+Cr) tVi,.dJ9D) Z= Cr) 3AVqIV oz/C(r't) £v±*cCCr)8dJ9-(Clr)Sd39) sev)+Cr't)zvi±tCCz'r)8dJ9(zCI'r) SdIO) sev) +C(r'I) i vi* ( (1 r)Sd39-C I'r)SdO) sGV) )=N I dOD */CClr)iosx09-CPr) 1ND(Clr)±oexo9+Clr) lN)sav+Cz'r)±oex0)- Czr) IN9- CCz'r)±oax09+ C'r) IN9)S8V+CI 'r)±oaxos-CI 'r)1N9-CCi'r)18x09+CI'r) 1N9)Sgv)=NI9 Cr) ai* Cr) ev.-. Cr)VGH+ Cr' I)UJ.' Cr) iv*' Cr) 'ViH=iVH 1TViH=I NIM Cr) 9avc Cr) qVSH+ Cr) iv.,. Cr) 51-Pi=Cr)rV81H C0o3*Cr) VSH) 31 C±13a*VH) /±X)c Cr) 8V'.dJ*W00bd=V~d INd3O' C r)SAVXOJ-' I)+X~dOc Cr) 8AVXO3=d3 *Zi(Zd39+T d39) =d39 lNdO. C+r) 8I N9+Xdc CI+r) 8X09=Zd39 INdJ' Cr) SINS+X0d3*' r)8X05=td3 CBONI lamJ0 Z 'Cr)SAVX03 ' vxv '6vs3ws ' Cr) VSH 6 - 'W00Od ' Cr) 'V0HH T 'Cr)qvi 'Cr)si 'HO '10 '18 '3NX 'HlVdX 'T-)d0~idV 11VO CBONI '8UH Z 'Cr) 8AVXO3 6 vxv 'Tvi.3ws ' Cr) V.H ' 3U 'WOQ~d ' Cr)Wl-l0H I 'Cr)qvi 'Cr'T)Hi 'HO 'ID '18 '3NX "H.LVdX 'T-)d0OIdV 11VO SIX's' =HO t-sII(o0*1vI) A1 0999 0599 0~99 o9 oZ99 ot199 0099 o659 0859 i= a 5T-301=10 COo3I90) Al ((I+r)sxo9+CI+r)aIN9+Cr)sxos+Cr)SIN9)*5'=1D (T-r)A13a*cS'+HlVdX=HlvdX CI'3NC) 31 (r)Al30*5*+HlVdXwHlVdX 133sN'I=r zL~i 00 Ct I )Lvl= (T) 'ivJ *0=3NX 013' I=H.LVdX 9ST 0o5L 05SL OMS oz5L 01 SL 005L 0Etr o -o= (r) 80 o00 = (r) ±08±0b OP~ 0± 09 (0b3*9flndNI0dI VGH- (r) sba ai±o s.%. Cr) e±v j* - Cr) ev. Cr) (c r) (r) eob-ab0+ (C(r) si- Cr) 3AV~i) * (r) av*. (r) tV9H= Cr) aO abb- (C(r 0 bi- Cr) 3AV V) *(r)±iv's(r) V±H=Cr)±08±ib (r) 91) ).c9 i s*C Cr) sivi-abbi o*(r' n) b±- **. (r)d0±0-NN ISO=N I Sb A- Cr) dollO S-*. (r' IWN) u±) *9IS C*r)ANSiV 033sN'T=r o9P 00 (~ I *0=)NIS8-O OPiL 3 os'iL 3 oLtiL o9 iL 3 S±N3W313 1Vnfl±JflUs IV±N0ZIbiH 0± S3xflii ±V3H 3flNI±03 oL'i 3flNIN03 091 ((Cr) 3AV 1TVd- (1 r)do) SeV'XVWdO) 1XVWV=XVWd0 o9tr01 ±09 (vb3*()±ON) Al C'I=i 09qi 00 01Qt~ oC L oz L oiTL oo iL (Ct+r) Vd+ Cr) Vd) 3cS*= Cr) 3AV 1Vd C1+r)nvld+T) qVd=(1+r) Vd £ON=(C)±ON (1*3Nr) ji o6 L o8CL oLM £ ±33SN'i=r oL 00 0= (i)ION 'd WON30d/b3wflNd= (1) 3nlNO o9CL 05CL O'TU oUL 5 (C)13303+(Z) 1A303+(1) tA303+lI3wflNd=b3wflNd oMt 3lI±NO3 0'7 ooU o6zL *0=(l)TAA303 *0=(l)Z.A303 GVT o r 0± 09 ((r)AvnVid-(1'r)da)).()ThA03=(1)t.A303 o8zL oLzL o9zL (1'r) OdO=(i'r) ZOdG O'zL 0±l 09 o ((r) 3Av~d * 1(1'1r) 0dG) AI (1'r) da= (1r)LOda (1'r) IN9=(i'r)i0IND 9C'T MV 0± 09 'Tdo=(1'r)C0d0 ozzL~~8P oizL oozL 061£ ML(1'r) MVi 0±0 091 L ((r)3Av'Td±0(i'r)0dO) Ai OSi £ (i'r) da=(1'r) Z~d0 (1'r) IN9-(1'r) 0N9 0'T I oti ozL £f 6 ottL 001£ o~oL 9C7 0±09o ((r)3A0V'd±I9*(1r)da) 01 0±0 (I1'M0*(1)±ION) i't=1 O0r CI0N=(C)±0N C1*3Nr) ±33SN'i=r OS'T Al Al 00 Ai 00 0O=XVWda oLoL LST 008 0109 0009 'i@013'=XVWJI. H6'C*013'=3V8U9 HO!' *013'=3V Ad0 HOT I I 1130 'W00~d 'OW0~li Oos '31dNI 'OVbADO '3VHdG '11N1 '3W11 '(1)31111 (oZoS'9)31lbiM oCLZ-ldbi=3ldbI. £LM-woao~.=woobl 3flNIJ.N03 086L *LZ-NNlIS=3NIS 066£ K (L)ON IMb~ SC 01 09 (±S3ii±±13W11)dI Al (*0i9*3Wl) (01-XVW1)/3Wu3'sxvw1=ovbjia oL6L o96L oSGL 0*1=ovbU1O XVW IND/XVWD9OVU.A9O 0~6£ oM6 oz6L o16GL oo6L 099 01 09 (*MLoil1XVW1 o99L 595 ldV/XVWdGOVHdO *M-ldb1= (ldN) idA o09/3W 11= (ldN) 1dX T+ldN=IdN S9S 01 09 (o~b3*dI) Al 0OV1NbJdN6IldN) Al 1+dN=dN 095 o£9L at 9£ 3 1fl01Nl~d 098L 08rL oU9L oz9L 009L I 'HOT (CO013 d=3VbA9O 'H01'C*0!3'=3VH~dO 'i*0T3dT'3WlI IV 30N3ANOO 01 0311VA SM01A SSVW HSC/).1VWb0J 0105 ovb9oa 'OVU.~dG '3W11 (0105'9)311W' XVW IN9/XVW9O=3VbA9G ldV/XVWd0=3V 1Ada 3lI1NOJ 055 o6LL OCS (dWVG-*I)*c(1r) IN9G+(1'r) IN9=(1'r) 1N9 oZ5 o9L £3flNI.NOO OLLL 13sN'i=r oC5 00 0%S 01 09 U0o3*3W1) Al 095 01 09 (XVWIN9ktc10011'XVWGOONVIdV'*1001V'XVWdO) Ai 3flNINO3 015 3flNIINO) 005 C((1' r) IN90) Say'xVw90) 1XVWV=XVW90 (((1'r) 1N9)SSV'XVWIN9)IXVWV=XVWIN9 I ((i1'r)Zod0-(1r)C0d0)/ 05LL orLL oCLL oz £ 01 LL ooLL o69L 09£ (I)10N) Al oo5 01o! 09 ( P1=1 005 00 C10N=(C)10N (1*3Nr) ji 133SN't=' oi5 00 0=K()ION 0=Xvwoa 0-=XVWINO 095 01 09 (CLI*9XVW03IUH~*3*S3I)iI o95 01 09 (oI.9*oifdNl)JI 099£ o59L ol9£ oM9 oz9L at 9£ 009£ 3 o65L 3 SM01A SSVW 30 33N3Sb3AN03 095L 3 oLSL 3flINO3 OP 1 o9SL oo=r' b 891 OZ58 0158 QoS 'rimi '(r'I)ZoNI 3flN I NOJ 019 3flNIiNOO 009 'rizvi 'risi I £CLz-(r' I)C3Av1=ri~v± 09119 05119 £Lz- (ri I)M±v.=r Is o Lz-(r' 1) v±v=rIisvi £CLz-(r' 0i~=ribi IWN'1=I 00 79 06Q oB£9 OL£9 0908 0509 Or£8 I 009 00 I 'X i ICVii 'x(L ith 'X'T ,C3AV1,, 'XL 11Mi 1i 'XOTlii 1 xr/)1vwb0j 0805 ,i~tv,1'x(L tilh, 'x'i 113AViii 'XL ,,U.lii 'XOTI (0905'9) 311HM r (oL05'9)3.I~M 019 0± 09 (1 SNw11-r0No011NI~1dN) Al 019 01 09 (£ Z0T*1o10HlJII 3flNI1N03 z6S 0M£ 01 £ 00£M 06z~8 09 z9 (£'i) IND ' OLZS o9zs9 05s9 O'79 0£QR ozze 0tZ9 00zs 0619 0919 TWN'Z=l z65 00 (r'1 1) .ld=101 103SN'i=r 019 00 3flNI1N03 £65 3flNIN03 065 (CoZ138dT'WIX9Z)1VWb0J 0905 I (ri)SAVIND ' (ri)SAVX0J (z'riI 1N9 (TIN1N 'ItlviI'risi 'Ni (o9oS'9)311tfM 6(rO3AV11Vd *£Lz- Cr) 3AV'7v1=rli' £Lr-(rsi)8=risi oS 065 01 09 (£1Z3d'£t'X)1VW8i0J 0505 riu 'Ni (oSoS'9)3iIHM *£Lz-(ri)i±=rii oLS o65 01 09 (C*Z138'£ilx6'Zl3d'i'X)VWNOA005o1o (C'nIr )I NO z £Lz- Cr,1) 3AVvv=r I 7v1 *Lr-(rnii=rii oL1 9 0919 05 19 0181 095 01 09 (lWoio9ri) fi oLS 01 09 (133Soso9i) ji rixvws1=nI o65 00 (133SN' IWN) oxvw=r ixvw o£e1 1 Me oi19 V/ 11 (CAr) I NO,, 0019 0609 'x£ ,g(r)SAVX0A,, 0909 'x£ i1(r)MV11d,, 'x£ 'x£ ,,(r)si 1i IxL ,1r,, 'xiiiuo oLo8 (50013 '=113a H6'i*013 I00OOd ,,(r)3AVq1Vi,, I "xL ,11,, 'xV1/)ivw~oj o0o5 (005'9) 31IPIM £65 01 09 0909 0509 H64C*ot3'=W00LL z H6 6ST o~o6 ozo6 0106 ooo6 0668 0868 oL68 0968 o568 0r68 oC68 OZ68 0168 0068 0688 088 0L88 098 0588 0$88 OC8s oZ88 0188 0088 06UR OUR8 OUR8 09L8 oSL8 O'rL8 OURL OZLS 01 L8 ooL8 0698 oL98 0998 0598 oMT9 0C99 0Z98 0198 0098 o658 0858 oL59 0959 0559 0578 ',,f ',1, (0lW*9§~Wb (JoMWlr*31 I I -N=1 IWN'1=1t o69 00 3WIlaNVwN0MWI.V*103WI1) 31 *0=IW3bIO *O=IdbJ. *0=xvwj o=xvwia *0=101wx0 0=8SiN lid 0O=SbNllid 0O=aVbd *0=bOI0Hd *0=ULSd 0=i3flld * 0=W3H~d * 0=N3Dd I b-0Z3U=lob S3 inionbi.s aNV SllVM HolHl S3ufl1YU3dW31 M3N 31vinlUV 1130.cb IV1Sd+b IV1S3=b IVlS3 ii 3G*AN0~d+CANO03AN0J3 Ii 30*ZAN03d+ZAN03 3=ZAN0 33 113a*T1ANO~d+TANO03=TANO03 113 0+3 W 1=3W 11 3W I.=S3W I. 589 3W11 30NVAGV (o$6)aN1Old llV3 1+J.OidN=L1dN (xvwA '00 "1 'XVWX '10 "1 'Z 'SV1A 'SVIX '3111i. 4 HT '0 '1 '1 'ldN 'idA 'ldX '0 '0)Ndi01d 11VO 0=ldl Al S89 0± 00 CXVWdX±11 oo%9/3wIVaNV*XVWdA*11 * LZ-dL) S89 01L 00 (o0b31dl) Al (M3NldN) SGVI=VlN)JdN (.lNd~l39*XVW) AlI M3NbdN~.LN I dN=IN I~dN (XVW~.U 3D*XVwl) AlI 006 0± 09 (XVWb.w39*XVWI) Al 006 0± 09 (XVWI±L3903W11) Al 089 0=d N (1 0t03 9 '=SSO13 llHWH 61*'01 3'11 =)INIS3 'i0T3',,=HIVlS3 ,~'C03'i, ,, '03' =W00H3 11-4AN03 ,,'C0t3 5 ,,/C*013 -0VU3 ,, lP0t3,,-=I010H3 , '013,, =H1S3 11 '1*013 I =13flJ3 ,,'C*013'ii =W3HO3 11 'C*013dlii, =N303 is//)1VWMI0A 0009 Z 'NII1S3 SS013 ')INIS3 'W00N3 'OIfWU I 'iAN033 'ZAN033 '1AN033 'GON033 'SON033 ISHN113 'SbINI13 '0VHI3 balO1H3 6HIS3 '13n3I3 'W3H33 'N393 (000949)311~M 09T 0£ S6 oz56 0156 0056 Cr) so'.±sj=sso (Cr' i)mb'rimO) NlWV=rlMb C(I- (MMO/113.c(r,0)MbOa)dX3) /C(r, i)mba.*(r, i)m1-*L) =rimb oL 6 o9'6 ZIL 01 00 (03*SX) ~i Cr'I) Mb+b010Hd=bQ10Hd C1- (MM3/1130),,.(r'0)Mba) dX3) it(r Imba/rimb+(r'1m= (rI)OM1 Cr'i)M±=Cr'Usmi (IvO3*DIt)di rimb+ Cr'I) MO-W~=£WT3 o6 6 01~t6 ozr6 o6C6 oM£ oLC6 o9C6 oM£ o~zS T10/ I l-N=l oC 61WN'1=ll OTL 00 (11nwj)sev'~'= OIL oiL 01. 09 1 C(I-A3a1) AV030-A30) AV30) .c( (I-A30 1)10031CAO I)10031) /CC -AO I)100J1-33W II)+ (1-Ak0I) AV03a=j 0109o C01~91-A01)100J1) ji 9oL SOL OIL 01 09 o5C6 0*16 ozC6 at £6 ooC6 o6z6 oLz6 o9z6 oSz6 Oz16 CI) AV330J=J 90L 0± 09 C16190A001) Al *1oL ZOL 01 09 I+C(r) A0330 1=C(r) AV3301 4IOL 01 09 CAV03NOb3A,3QI) 31 'roL 01 09 CCAk30lI1031*31*3Wi) J1 (r)AV33ai,Aoai zoL 1130-3Wu1+(r)O3Wli=33Wll ioL OIL 01 09 Cr),AV030J=J ioL 01L 09) (C01911fiwj) ji CN) 0£ Q6 ozz6 0N03b+0V~dd=aT~d r)doiib-= CN) QN0~ Cr) ±oab.L= Cl) oiz6 ooz6 0616 oM 6 OOL 01 09 oL16 0916 0N0 Cr) SSVN~s(I)b Mwa985Z=tW3bb Cr) SSVN~rc (Z)b1MWG* 0 985Z=ZW3j Cr) SSVN'.' C£)MWo**9S5Z=£3b CdMWI1193WI1N0NMWI1*31*3WI1) 31 OOL *=Wmb *o=iW3b 103SN'1-r QSL o 3flNI1N0O 069 Ci) 1b+stN11d=SbNtld M3N11=ClI 1 016 0£C16 oz 16 0116 0016 10111 o6o6o Z/113a* C3W I Ds'. (01-M3N11) + (1130-3W 11) * (01- (1) 1±) + (1) 1011= C(1) (1.-(13/113a*CI)O)dx3)'.(I)1ba/I1b+CI)11=M3N11 0806 I b+(lI 1-lW3Ni=lW3bb oLo6 C' 1J ± 3a 0*106 i)iba d3)(I 1o)CI - l£)= lI T9T oz66 lll+13ld=13l~d 3flNIiN03 tOL 3flNIINO3 z2rL 'T=(r'ww) SI (ri-i)t3AVI= (r'Iww) u1 IWN'I=WW iziL oa (u'aO o886 r%tU+ibb-W3Hb=IW3Hb oo66 0986 o586 oi86 oU86 oz86 Nw~I (r'I)b+(I+i) ONO~b- C') aN03+(r' I)W3H0b+Cr' 0AV33a0=iUOb Al U(r)mo+*/(r)so+Ho=Ho C5*o b3*HSJ) X130'* (C(r) 3V*003O3HH+ Cr) AV 3,..cA*0HH) =H3 CI 'Iws 'IIAws "n 'jo '(r'I)HI)d0~dJ iivo oUL 'o Cr) £=(r'i)si (5ooo soi * O- ot 86 0086 o6L6 o8L6 oLL6 bv= Cr'I)W3H3b sev) j i r 1in) N3b=Cr' i).L3 1130/((Cr' I)IO-NOH) *c13RL=3?b 59o /o6+rI)±3H=N3H (rH/1a.GX* awxo=Cr' i)wxo z=(r'i)si 031lW1l N0isnliii o9L6 oSL6 oIL6 otL6 oZL6 ooL6 o696 -- z=si 3 ozL 0± 09 (awxo±1i*wxo) Ai Nwx0o= Cr' wxO t=(r' Osi GU.IWII SOI13NI)N OIIOGVVd -- I=SI 3 Nx L3/("1 O-~)*()H*(z1/(0*Z (ob 'NOH 'Cr't)13H '031d '13H '1130 '(r'I)Hu±)w3H3 I1VO otLl 0109 o896 oL96 o996 o596 oIT96 M~6 oz96 ot96 0096 o656 oLS6 o956 o556 (*39G(ri)si) Hi oUL 01 09 (0ti3*W3H3I) Al j*ib()SNO~*00=rlA0o X13a/ ((rI )u±- (rI -i)uI)* v)=(I) 0N03b z(i~sAw)(~ (o/i~sows*ro (0 'i3oWs I)i~ws '33 'o '(r't-i)l~l)d0OIdl 11V3 (0 '3)IWS 'ANWS '33 'jo '(r'I)YLL)d~lidJ 11VO SUL 01 09 (vb3oI) Ai CUL (r'1 1)i.= (r'iS1)s ziL CUL 01 09 (r'I)sb+HiSd=UiSd StL risb+ Cr'1) sb-ZW3=ZW3Nb ZW3H-(r'i)sb=risb (ZW3bb*L!D(ris-(r'Iosb)) ii (Cr' i)sO'risb) tNiwv=risb Z9T 5**( (C*,s3W I 06101 02101 0910ot 05101 0101 I*I d)/AO 3WS.'N0Jd3.'cNOJ0H 1),*XOZOG (1)10ll1i**l) = (1)1 ob (I)13b+S3N0Jd=S3N0~d IWN'1=I OWL 00 *0=93NO~d *0=S3NO~d 002 01 09 (o3V030HOOI) Al 31hJON03 OINI xfl'l 1V3H 3NlW83130 0 101 i130.'8bN I1d+GbN113=SbN 113 I130*S JN I1d+SbN I13=SbN 113 11 30'0Vbd+0Vb3=GVb3 1130 0G10Hd+b010H3=H010H3 11 3G' ISd+N1S3= 1S3 113 0.13fl~d+13lJ3=13l.3 113G).W3H~d+W3H33=W3HJ3 I130*N3Dd+N3D3=N393 3flNI.LNOJ 01101 00101 06001 08001 OLool 09001 "St 19L 01 09 * Z/1130=1130 S3W11=3W11 0=91J I CN0Jd=CANOd I N0~d=TAN0d OV.LSd=)d1I Sd SSL 95L 017001 Cr' i i)sbi= (ri')b Cr' i i)sin~= Cr, ii o r ssi=Cr s Cr 1 )S11= Cr 1 1WN't=II 9SL 00 .L03SN't=r S5L 00 17SL 01 09 (0o10±1*010)AI ( (9'1 ) SbLL- (9'6)) 1) =Iola 1TSL 01 09 CV03C91)SOI)1 TWN'1=I 'TL 00 3flNI1N03 OTo Cr) 80+8bN1I1d=dbN I d 0C001 M3NS1= Cr) 81 Cr)8±=(r)s8±L C10b39511I)A 01 001 0Z/' i13G*~(3W1I" C0-M3N8I)+. 1130-3W 11) (oCa- Cr) 81) ) + Cr) 1018±- Cr) 10181 00001 13a*( r)800) dX3)*( r) 00/ Cr) 0+Cr) 81=M3N81 3flNIINOO 0666 0866 oL66 0966 05L O1L C(r61'I"1L'dbI)TXVWV=1du.l (103SN03or) Ai C(r' I i) u11xvwi) i xvwv-xvw± CCb/i~b) s8v'xvw10)ixvwv=xvw1O Cr' I)WXO+io1wxo=IO1WXO Cr' DW3Ho0+W3H3d=W3H~d Cr'I) AV300+N30d=N39d 01766 OC66 C9T 1130'3WIl 16665 ININd 3 OTLOT 0 o0oi 08901 o0oi 09901 05901 0 901 OC901 OZ901 01901 00901 06501 08501 0 00LOT 3,dnsoio in*Wd0H l*9WS=Idn I=IN3AI (XVAld*3D'WO&8d) l WOONI*Vb*WNOHb=WOONd 508 01 09 (T'b3*IN3AI) Al M3NdO=WOObdO M3NOHN=W fOHb M3NJ=WOONJ (M3NdO*WOOIdA).cM3NOHld)/(1130),tWOOllb+WOOHI).tWOONdO*WOObA cWNOHH)=WoobI il3O*WOObb+WOOb3=WOON3 SSOIVb-IN3Ab-ANISVb-OlNlb=WOONb lNdO .t(M3NA-'T)+XOdO),,M3NA=M3NdO 1 (WOOIdA).cM3NOHb) / (1130%.c ( (OJ).cOOK8-WOONJ* OLSOT 09501 05501 0501 0 501 OZ501 OTSOT 00501 o6qoi OPOT Mol 05 01 0 01 WNOHN)*IN3A+inowNo*wooNj+smoN*ioiwxo)-WOObA*WldOHII*WOONJ)=M3NJ 1 WOONA/(il3(1* ((OOHld-WNOHN)*IN3A+inowNg+smoN),tioiwxo)-WOOIJA).cWIIOHN)=M3NOHb I113G*SSOlb+SS013=SS013 IN3Ab+NNISlb+SSOIVb=SSOlb WOONI*WOONdo*inowN9=ssoivb (WOObi*WOONdO)/(iN3Ab-NNIsvb-OiNib)=inow'do (T*b3*IN3AI) Al *o=inowbo (01*OdO*OOHH-WOONi*WOObdO*WNOHN)*iN3A=IN3Ab SMON*( ANOOd+ZANOOd+IAN00d)=OiNib Z08 OL 01 NNIS3/113G*NNISb+NNISI=NNISI 1130*NNISb+NNIS3=NNIS3 NNISIb-SMON*NNISNb+NNISVb=NNISb (NNISI-WOONI)).cNNISV*WNH=NNISVb OMI OMI (( **01- **NNISI)*DIS*L*+(OI-NNISI)*inOH)*NNISV=NNisib OMI 0001 (XONI $WNaH 'SZOO 'NNISV 4XJWS sinOH '3b 'Od 'XOHN '01 'NNISI 4013*1 601-3*1 '013*1 6013*1 '013*1 $i)dONdV IIVO offoi OKOT (XGNI 'WNOH 'WOONA 6NNISV 'XJWS 'WbH 43b 'WOObd 6WOONi 'NNISi '013*1 601-3*1 oLM 09M 05col 01COT OCCOT OZSOT 01M 00COT NMI OLZOT 09ZOT 05ZOT 'XOHb IIVO '013*1 4013*1 '013*1 4i)dONdV ZOS Ol 09 (*Oeb3*NNISV*NNIS3) JI INdO*(WOONA-*I)+XOdOkcWOOIdA=WOObdO 009 S311b3dObd NIV WOOU 3NIWH3i3O 3 113a*G3NO3d+83N003=S3NO03 113a*S3NO3d+S3N003=SON033 3nNIINOO OLL il 3a* (r) sob+ (r) ioisob= (r) ioisob o6zoi i 113a/ ( (r) iolsob5***((S**3WI1.*Id)/NOONWS*NOOdO*NOOOHb)*(r)av*(r)ioiei*" )=(r)e:)b (r)G3b+93NO3d=S3NO3d 103SN't=r oLL oa 3nNIINOO 09L il3a* 0) iob+ Mioilob= 0) io.Liob 1130/ ( (010110b- O Zol OCZ01 OZZOT t9T 165 5999 FORMAT(3X,"TIME =",2X,F12.2,3X,"DELT =",E1O.4) 805 EREMDR=EGEN+ECHEM-EFUEL-ESTR-EHOLDR-ERAD-ELINRS-ELINRB 1 -ECONCS-ECONCB-ECONV1-ECONV2-ECONV3-ESTAIR PREMDR=PGEN+PCHEM-PFUEL-PSTR-PHOLDR-PRAD-PLINRS-PLINRB 1 -PCONCS-PCONCB-PCONV1-PCONV2-PCONV3-PSTAIR IF (NP.EQ.O) WRITE(6,6010) PGEN, PCHEM, PFUEL, PSTR, PHOLDR, 1 PRAD, PLINRS, PLINRB, PCONCS, PCONCB, PCONV1, PCONV2, PCONV3, PSTAIR, PREMDR, QROOM, QSINK, QLOSS 2 PFUEL= PCHEM= ",E10.3, " 6010 FORMAT(/" PGEN= ",1PE10.3, " PRAD= PHOLDR=",E1O.3, " PSTR= ",E10.3, " 1 E1O.3, " 2 E1O.3/ " PLINRS=",E1O.3, " PLINRB=",E1O.3, " PCONCS=", PCONV2=" PCONV1=",E1O.3, PCONCB=",E1O.3, E1O.3, " 3 PREMDR=", PSTAIR=",EIO.3, " E1O.3/ " PCONV3=",E1O.3, " 4 QLOSS= ", QSINK= ",E10.3, " QROM= ",E1O.3, " E10.3, " 5 6 EIO.3) GO TO 170 900 IF (IPL.EQ.0) GO TO 910 TITLE, XLAB, CALL PLOTPR(O, 0, XPL, YPL, NPL, 1, 1, 0, 1H C YLAB, 2, 1, 0., XMAX, 1, 0., YMAX) C NPLOT=NPLOT+1 C CALL PLOTND(940) 910 IF (NCEND.EQ.0) GO TO 81 10720 10730 10740 10750 10760 10770 10780 10790 10800 10810 10820 10830 10840 10850 10860 10870 10880 10890 10900 10910 10920 999 CONTINUE C 999 IF (NPLOT.GT.0) CALL EXTFLM(O) STOP END 10930 10940 10950 i 01 09 (iNOJ'I1) JI ol900UdV 7101 09 ((I) iJ'1*1) JI Z 009ObdV 065OObdV ) /30) =d (*9/* 1) ** (0139' 1/bd,'c 19) *((1X 095008dV OZ 01 00 (ZOb3'IGNI) 11 OL50OoUV S313130 1N3W31V1S DNIM0110J N013VA NOUOHN03 095OObdV 05500~dV NOli13ANOJ iv~fl1VN 1N3-fl8wfl 'S31V-ld 1311VNVd NO N0I133OO 0O1TOOdV 0C500dV Z=IGNI OZ00bdV .c091 I 0=1 fNO 01 (3NX/30)* C C0.c*Xb9) 01 S00bdV OOSOO~dV V NO N01133ANOO iv~fl1VN IN31flefl 31V~d IVOI1NM 061T00bdV 021I00dV 0z 01 09 OV 0obcdV 1=10N1 091T00bcV (ONX/3O)*~(5z0o't'X~9)*oc9C~0=1fNOI 0S1100bdV 0VT00bcdV IVOI1NM V NO N01103ANOO iv~fl1VN NVNIWVI 31V~d OMJ00cIV 0Z1T00dV 01 01 09 (6+i319DXN!9) Al O01T0dv O04T00dV ''(x/N)19=9 o6CoobdV (Z.c*3Nlwx) / (vi-mi)SV.'(C**1X)*(Z'.c',c3N0HU)*~Vfl8,.cWS=1I39 09C?00~dV 3NNV91/*1-VI39 OLCOObdV (9S*6oi+NtlV81)/IS.1'*u4c3NV81).d-39'I 0=3Nflwx 09CO0bdV (3N)JV81cVt) /W00bd=3N0Hb 05C00bdV (Vl-mI)*WcO -mi=3NNV9I 04 100MV Oz 01 09 (0*11*01) if 0ff0ObdV o=tnlNO 0ZC00~dV 01 LOQ~dV (OScV0HN) /lwx=Aia 00C00~dV Nd/lwX,.d 3=WS 06Z00~dV Moq*rLZ*=d3 08Z00~dV OLZoo~dV 05Z00UdV 0qZ00dV 0Q00bdV 0OZ00OUV 01 ZOO~dV 00Z00~dV o61 0OHdV oLTOOudV 09T~OOdV '58Z 'Sty* 'S 1* 'SoP (8S*6oI+NVsL) l 3 3 3 3 3 3 3 3 3 3 3 3 PL4=3s 3a/H1VdX*c3N=X3N (v'~fwx) /3Q.c (9) SIEV=3b /5*1**IVg±&y-39 t o=nwx (V1cVN)/W00OId=V0HH Z/ (VI+MI) -11val oo96=ows HiI VWV/L+3q7T1-S=vH /W6'98* 'oLo* '6L 'SWo 6UL 6W 4z5- 6-o I '* 'LI WH 'HJ 'AN viva (L)WH '(L)HJ O5TOdV NOISN3WI0 3 0ZT0OOdV OTT00~dV QOOO~d 3 3 07100OdV OCt0OOdV S3I1lbd0~d )iIV (aNI 'UH $X0J 'V 'JWS 'H 43NI 'WOO~d 'V0HN 'Vi 'MI 63a '9 '1X '3NX 'HiVdX 6aI~d0IV MuounI~lfs I 99T oL600bidV 096O~dV 0S600bdV 01760O0UV 0C600bdV oz600NdV 01 6OObdV 00600bdV MOM8od 02800Od oLSoo~dV 0980o0M 05800Md 0r800Od MO8ooM orSo~bdV3 01 80o0M 0080o0M oELoo~dV oRLooMd 08 01 09 1=CGN I W4W IX/3GObd3bOrCr0+''L=nNa S31V~d 131IVbVd N33M13S M01J O33b0J NVNIWVI 3 3 3 05 01 09 (*OOOC03N) Al 09 01 09 (vOV~(i)sev1) Ai o r 3 Z=ZaNI (0*.tX3b)/zSo*=zAws (HiVdX/3a) *c(9'0**bcd) c(840*ccX~b) .96Z0'=rfNO 0 31Y~d V .LSVd M01J G33bO0I1N31lSbIl 3 3 0 01 09 t=raNI (50*.'X3b)/r99o=JWS (HiVdx/3a)*~(tC*o*3'ud) k (S00*c'*X3) '.Zr=ZnNa OLL00OM o9Loo~dV OSLOOMid o rL~odv oULoo~dV 31V~d V 1SVd M01J 033N0A NVNIWVI 3 0C 01 09 (5+3*S*1D9x3b) Al SZ 001 01 09 OZLoo~dV 1flNa=fNG 01 LooMd SZ 01 09 (C"Nai) ji or OOLooMd o690oodV O NOud oL90O~dV o990oodV oS900udV 01 T90V Al aH/iinWH~1 fNO=1 fNO 1T=AIGH (*I±9DAiaH) Al (lX/JNX)*ffo-jLL+W=AIaH *i11flWH (*ij.9'1lnWH) Al 0o=11lWH (01V11~inWH) Al (I -1)WH- (1)WH)c(I - ) -(1)1 )/(I--1)1 -) + (I-) WH=1flWH 'TI ri oC9oo0M 01 09 I+1=1l or9008dV L9T 168 C C C TURBULENT FORCED FLOW BETWEEN PARALLEL PLATES 50 DNU3=.023*(RE**.8)*(PR**.4) SMF3=.00140+.125/(RE**.32) IND3=2 GO TO 80 C 60 IF (RE.GE.3000.) GO TO 70 C C C LAMINAR FORCED FLOW THROUGH AN ARRAY OF TUBES DNU3=8. SMF3=25./RE IND3=1 GO TO 80 C C C TURBULENT FORCED FLOW THROUGH AN ARRAY OF TUBES 70 DNU3=.023*(RE**.8)*(PR**.4) SMF3=.00140+.125/(RE**.32) IND3=2 C 80 DNU=DNU1 IND=3*INDI-2 IF (DNU1.GT.DNU2) GO TO 90 DNU=DNU2 IND=3*IND2-1 IF (DNU2.GT.DNU3) GO TO 100 90 IF (DNU1.GT.DNU3) GO TO 100 DNU=DNU3 IND=3*IND3 100 SMF=AMAX1(SMF2,SMF3) H=SMK*DNU/DE HD=DIF*DNU/DE RETURN END APR00980 APR00990 APRO1000 APRO1010 APRO1020 APRO1030 APRO1040 APRO1050 APRO1060 APRO1070 APRO1080 APRO1090 APR01100 APRO1110 APR01120 APR01130 APR01140 APRO1150 APR01160 APRO1170 APR01180 APRO1190 APRO1200 APRO1210 APRO1220 APRO1230 APR01240 APRO1250 APRO1260 APRO1270 APRO1280 APRO1290 APRO1300 APRO1310 APRO1320 APRO1330 APR01340 OMOIAS 011001:18 0000-1:18 offoolls offoolAs oLcoolis 09001AS oscool is O Cooljs oicoolis ozcoolje olcoolis o0coolis 06ZOOIJS o9zoolis OLZOOIAS o9zool is OSZOOIAG Ozoolis oQ00lis ozzoolis ol z6ol is oozoolis 06100138 081001AS oLloolis 091001 Aa 05100118 O Tooljs ocloolls 0zloolle 011001 As 001001:18 aN3 NiinlH 3nNIiNO3 XOdO*(ir)IOGXOD+lNdO),c(ir)IND=(ir)8d3D (ir)IND*((r)SAVXOA-oi)/(r)SAVXOA=(ir)ioaxoo 0 01 09 (00011*01r)IND) Al C11=1 os 00 (i+r)SIND* ('0'39' (i+r)SIND) Al ((r)SAVXOJ-*I)/(r)SAVXOJ=(i+r)exo!) (r)SIND*((r)GAVXOJ-*i)/(r)SAVXOA=(r)exoo (*0031*(r)SIND) Al (NIXOD+NIIND)/NIXOD=(r)SAVXOJ (i+r)8XOD-NIXOD=NIXO9 o+r)SIND-NIIND=NIIND O 01 00 (*0*3D0(t+r)SINS) Al (r)SXOD+NIXOD=NIXOD (r)SIND+NIIND.=NIIND OZ 01 00 (*0*31*(r)SIND) Al 3nNIINOO (-ilr)lO8XO9-NIXO9=NIxoD (i'r)IN9-NIIN9=NIIND 01 01 00 (00*30*(16r)IND) 31 05 0 OC OZ Pi=i oi oa OWT=XOdO OWT=INd3 *O=NIXOD *O=NIIND IND 49d39 4GAVXOJ /MOIA/ NOWWOO 10SX09 'SXOD 49IND (Pioaxoo "(L)SXOD NOISN3WICI (P9)IND "(P9)8d3D '(9)SAVXOA NOISN3Wia (L)GIND SMOIJ SSVW 3SV9 (r)MOIAS Minoan 69T OT031-3 01 '003HO 001T03H-3 o6 0O3HO 09CO03HO OLCO03HO o90oo3HO 05CO03HO 070031-3 OU003H-3 OZC003HO at £003HO 00C003H43 06Z003HO OSZ003HO OLZO03HO ON3 N~fn13N 3flNIN03 0~ 1130/ (lON-N3OI)*~Hl3G03 JI ION=NON (ION*39*NO) 3lNI1NO3 OZ (M)l~bS=N3Oi OZ 01 09 (*031*M) 31 13ld)13l+ (NZ0HtlNZ0HN) /(1130G*)31VN) =M ION=NOH (31! (O1/C3+O) -) dX3*ct3=)I31Vb 'T+38*L=Hl13o C+359=NZ0HN 01 o09 1T1=zo 53W5O=10 t1Y31) Ai 01 01 09 (o6 8389*?=13 05Z003H3 04T003H-3 O0Z003HO 'oL99z=.o 09ZO03HO 0tZ003HO 0OZ0031$3 061 0031-3 091003H43 oL1003H43 091003H43 O0510031-13 0110031-13 0T03HO 0Z1003HO 011003143 001003H43 01 01 09 A~1~3) i 04 o36i 9+31 r9=43 C+3809=ZO '+3L*T-10 NOISS3NdX3 1VNIWON 3 3 3 3 0T 01 00 (1313*39*13H) Al I3N-OHI3 ob 3 3 3 NOII33UN 3aix0 wflIN03IZ A0 S01I3NIN (30 'NON '13OI 'ONI $ION '1130 '31) W3HO MuNotifsfl 171 SUBROUTINE FPROP(T, CF, CC, SMKF, SMKC, ICALL) C C C FUEL AND CLAD PROPERTIES IF (ICALL.EQ.0) GO TO 20 IF (T.GE.3200.) GO TO 4 THETA=535.285 EO=37.6946 CAPK1=19.1450 CAPK2=7.84733E-4 CAPK3=5.64373E+6 R=1.9865E-3 THETT=THETA/T EXPT=EXP(THETT) CF=(CAPK1*THETT*THETT*EXPT/((EXPT-1.)**2)+2.*CAPK2*T+((CAPK3 1 *EO)/(R*T*T))*EXP(-EO/(R*T)))*4.184/270.13 GO TO 6 4 CF=51.*4.184/270.13 6 CONTINUE IF (T.GT.1223.) GO TO 10 CC=(7.1E-2+1.7E-5*T-0.89E+3/(T*T))*4.184 RETURN 10 CC=0.087*4.184 RETURN 20 SMKF=0.030 SMKC=0.30 RETURN END FPROO100 FPROO110 FPROO120 FPROO130 FPROO140 FPROO150 FPROO160 FPROO170 FPROO180 FPROO190 FPROO200 FPROO210 FPROO220 FPROO230 FPROO240 FPROO250 FPROO260 FPROO270 FPROO280 FPR00290 FPROO300 FPROO310 FPROO320 FPR00330 FPR00340 FPR00350 FPROO36O FPR00370 0N3 woon=(riN)fli (N)V39/rNC3Ai*()O-N~vwOO= (r'i-N) 3AVI O 1-'Z'OZ=N OZ 00 (i-Pr) vwwv9= (r'oz)3AV.L NOiiflOS 3.LfdWOJ 0 bOI03A 3 (rr)VjA/(ot-rr) vwwvo).c(rr) v-(rr) a) =(rr) vwwv (i -Pr) VAS8/ (i -Pr) o* (PP) v-(Pr) 9= (Pr) V.13E i'tz'c=rr 51 oa 51 0 VWWVD I2VAS.3 3.LdWOO Z)a= (I ) a W00oL+ (I MTVi+ (Z) 0= (z) 0 0'0=(IZ) 3 0*0=(Z) V 3flNIINQ3 (I-W)IHdIVlcZWV9)/(r, t-w) C3AV.L+ 01 ((i-w) iwrmL+(rPI'I-W) MI).c~WV= (W)0 IZ'Z=W 01 00 Z**(MM/Z130) =mWV ((Z13a~cZ130)/*t) ci13a=ZWVD Zl~azl30/*T*11/MM=MMV ((MM MM)T+ 3flNII.N03 3AVL-3968*7=(N) tHdlV L6tz-(r'N)AVI~c-3t+(Pi+() ozlm5 00 TMd1V' 3AVI' IWPmiL'M /0018/ NOWWO3 (W00OLL"TAVi'MMX'Z13a'J.130'P)INVHO Muo3n~ilfs 3 ZL 173 APPENDIX C PARTICLE INPUT, OUTPUT AND PROGRAM LISTING CONTENTS page Section C. 1 PARTICLE Input and Output 174 C.2 PARTICLE Program Listing 177 C-1 Sample Input Listing 175 C-2 Sample Output Listing 176 Table 174 PARTICLE INPUT, OUTPUT AND PROGRAM LISTING C.1 PARTICLE INPUT AND OUTPUT The particle program is an interactive code programmed in FORTRAN IV. Input is entered via a free format READ statement, or through data statements. Input variables are given below in order of input. Input Variable Name Definition TR Temperature of zirconium particle Nominal Value at its point of origin (*C) TAVEl Temperature of airflow into which particle is spalled at its origin (*C) TROOM Temperature of air in the spent 10. fuel building above the fuel pool (*C) GNI Nitrogen mass flow rate in the assembly where the particle originates (mg/s) AXAl Cross-sectional area of holder from which the particle originates (cm2) 579. 175 PROOM Room pressure in fuel pool building 1.0315e5 (Pa) UL Unheated upper length of fuel assem- 0. bly (cm) 5. The number of particle sizes to be N analyzed The input listed in Table C-1 was employed in the analysis of Chapter 3 for the first four variables described above; the other variables appear in data statements. A portion of the output appears in Table C-2. The out- put is self explanatory, with the following exception: W = total airflow exiting the channel (nitrogen and oxygen), and is given in units of mg/s. TABLE C-l Sample Input Listing TR TAVEl TROOM GNI 1390 1370 283 14800 176 TABLE C-2 Sample Output Listing TR= 1390.0 K W=.1820E+05 MG/S MAXIMUM PARTICLE SIZEw PARTICLE DIAMETER= TIME,SEC 0. 1000E-O 0. 2000E -01 0. 3000E -01 0.4000E-01 0. 5000E -01 0.6000E-01 0.7000E-01 0.8000E-01 0.9000E-01 0. 1000E+00 0. 1100E+00 0. 1200E+00 0. 1300E+00 0. 1400E+00 0. 1500E+00 0. 1600E+00 0.2150E+00 0.3000E+00 0.3900E+00 0.4800E+00 0.5700E+00 0.6600E+00 0.7500E+00 0.8400E+00 0.9300E+00 0. 1020E+01 0. 11 10E+01 TAVEI= 1370.0 K 0.1313E-01 CM 0.1665E-02 TEMPERATURE, 0.1333E+04 0.1273E+04 0.1215E+04 0.1160E+04 0.1108E+04 0.1059E+O4 0.1012E+04 0.9684E+03 0.9271E+03 0.8881E+03 0.8513E+03 0.8166E+03 0.7838E+03 0.7528E+03 0.7236E+03 0.6961E+03 0.5706E+03 0.4433E+03 0.3666E+03 0.3255E+03 0.3043E+03 0.2936E+03 0.2883E+03 0.2857E+03 0.2844E+03 0.2837E+03 0.2834E+03 CM TROOM= 283.0 MINIMUM PARTICLE SIZE= PARTICLE ABS. VELOCITY= HEAT GEN., 0.2727E-05 0.1297E-05 0.6713E-06 0.3653E-06 0.2508E-06 0.1736E-06 0.1206E-06 0.8371E-07 0.5799E-07 0.4003E-07 0.2750E-07 0.1879E-07 0.1277E-07 0.8623E-08 0.5790E-08 0.3866E-08 0.386iE-09 0.1022E-10 0.3350E-12 0.2678E-13 0.5400E-14 0.2148E-14 0.1288E-14 0.9678E-15 0.8187E-15 0.7370E-15 0.6856E-15 J DISTANCE, CM 0.1171E+01 0.2287E+01 0.3352E+01 0.4371E+01 0.5348E+01 0.6286E+01 0.7187E+01 0.8054E+01 0.8891E+01 0.9698E+0i 0.1048E+02 0.1123E+02 0.1196E+02 0.1267E+02 0.1336E+02 0.1403E+02 0.1739E+02 0.2179E+02 0.2568E+02 0.2901E+02 0.3192E+02 0.3450E+02 0.3683E+02 0.3894E+02 0.4087E+02 0.4266E+02 0.4432E+02 0.OOOOE+00 CM 0.1200E+03 HEIGHT, CM 0.1123E+01 0.2102E+01 0.2948E+01 0.3669E+01 0.4274E+01 0.4767E+01 0.5154E+01 0.5439E+01 0.5622E+01 0.5707E+01 0.5694E+01 0.5639E+01 0.5583E+01 0.5528E+01 0.5472E+01 0.5417E+01 0.5112E+01 0.4640E+01 0.4141E+01 0.3643E+01 0.3144E+01 0.2645E+01 0.2146E+01 0.1647E+01 0.1148E+01 0.6488E+00 0.1499E+00 CM/S ILLS 0 C(086*0HH0*'O0 DA/ll=ew i o49 tTI- =Z3 83WoS=1 0 *oL99z- =Z3 9399*1=10 o9L9i- =zo 3flNIiNO3 C) / (SOH )c.c*~p9A) =XVWQ OZ 01 00 ('005*13)iI 01 snwv/xvwa*DA*M H= 3 ((L o** (0HbG*' 086) *C51 0) / (DAIi" 0**Snwv*c6Z 0.*cS0HN) ) =XVWO SnfwW/xvw0a.cA*~S0HH=3 N C(o86*0HIIO) / (9A).sfwv* 81) )iids=xvwa (1VXV.'S0HN) /01 ND=DA (swi i i)nfwv=jnwv 0 ZI (13AVi+i) =swI1i (13Av1)nfwv=Sfwv (w00o±) nwv=NJnwv S0HII- *0059=OHUO (13AV13c Z+30L8*z) /W00Oid=S0H8 (W00NI* Z+3oL8 z)/W00~d=N0HN /0qT*0'O 4T*O/ZOZd3'Zd3 VIVO (I L j'xi ,=wootu, xi'xi I't Lj'xi ',,=I 3AV111 IND*CZ*1=0IND e £L +woo.=woou *CLZ+T 3AV1=t 3AV1 lND'WO01't3AVI'Nl 'aV31I (8S *o60+vi) / (5* I *Vi*i -39'l' '0) = (vi) nwv (i **c (mnwv/iinwv) 3.0 *0Ud*c (99 *0**03N).90 00+ 5*~o3Nkcq0)+ *Z))c a/JIVO =(dmwliw"3~liv)I '2 (Z0Zd3*WvV 1*+ZdD~ (wv- **a.c'7 £0q7)) / '2 (((WOO~II-Mi) *±H~c9Tqt + ( **WOONI-q*Mi) *ci -399 Z 1)*a~a- I2 ±~aWa*C~OO Z = (WOONI.'O'IH'mi Iiawa ' WV) iLi (wv+4c*aVqoiQ/ (i ** ( (oi' (OMI/ Z ) d Xhct13) ib s (01) I± S / ( ( (oM±/ Z3) dX3 *13)iabOs) * ( ( (OMI/ Z ) dX3 q2 -2 '2 ONIISII WV'd9Odd arID.lUVd zoo LLT ,r O=DVI 3 1 0 SSVW NI 30NVHO 31ndWOO I (IWO 'iH913H,,'XV ,iWO '30NVISIO,,'XZ Cz '*N30 iV3H,,'XZ',,N '311niVb3dW3i,,'X5',,03S'3WII,,'X9'/)IVW 80A (CZ'90)31lbM R lNlbd (iS/WO,,'XZ' *013'XZ',,=,kilOO13A 'S8V 3101A Vdi, '2 IX5111wollxz' *013'XZ611=N3i3WVIO 3131AVd,,'X5'/)iVA10J zz OZ'a (ZZ'90) 311 M OZ60'ZZ lNlbd z=rr 500=Aao 5*o=xao bi=oMi ( iL*olmnwvlsnwv'03b'O'IdIVO)OIH=IH snwv/o*oz* o*l*SOHb=03b OZ=OA IA-DA=OZ OL (SOHb/OHbO*O* *696z)lNbS=iA OL 01 09 (*005*il*3b)Jl 09 snwv/O*IA*SOHH=3b (C o*).csnwv*6z*O ,c*SOH'd)/(IL*O).,.),t(#o863,cOHbO)).c 51*0* 1*1**G)=iA 09 01 09 (*Z*11*31d)Jl snwv/G*iA*SOHN=RJ (snwv**9i) / (oo86*OHbG*a*O) =.LA 1130+3WI.L=3WIl W11=113a (113a*il*Wli)Al ((Idi/ZO)dX3)P*OZkclO)/Z).c).cG*9300 *I=WlI 500*0=il3G C-30*1=0 (C-30*1*il*0)J1 (o (r) ivoi j*aa) ***oi =a 3nNIINOO GMqT-=z3 83qO*5=13 LC oi oo (o6zqi*ii*u)ji LC 5C OL99z-=zo 8389*q=13 5C oi oo 009LCI-=zo *oqC6=i3 O'O=OA oo86=to O'O=OX 000*0=3WIl N4i=r oS oa SNOuvinolV3 31311HVd N1038 (,,W3,,'X' *0T3'XZ',,=3ZIS 313111IVd wnwlNIWi SX519 ,W3,,"X" *0T3'XZ",,=3ZIS 31311HVd wnwixvw,,,IX541///)IVWHOA 0 3 '2 TZ Niwa4lxvwa (TZ'90)311bM NIWG'XVWa'IZ INIbd (C-30*1/XVWG)01001*((N)ivoij/66*)=ao L*08ZZ/(SWII*(bi/ZO)dX3*13)iubS=Niwa 8LT 0 #0=1 0 0 *O=AOO (OA *il *,k) J I iOZ-=AA 091 01 09 (iOZ-*ID*AA)JI 3nNIiNOO 051 (80HVIMP.0 0*696z) idbs=ioz i LLS (C **bnwv*6z **bOHb) /I L Okc* (o 051 01 09 (*005*11*RI)JI 0O=iOZ o86).ciWd) 2.0 1 *I;c*.G*C5T 051 01 00 (*Z*11*311)31 bnwv/AA*O*UOHb=3b NnwWW 5.*IOHb*G*Q=IOZ 'dMl- (C3.c*a*9CZ5 *0) / (TWV+C**O;M'CO 0 =iOHH AiI3013A IVNIWb3i 03A31HOV SVH 3101INVd JI MlWbU30 AAO+OZ=AA AO+OA=A 0 *9/ KAI+ZAI* * +Ikl) =,kAa AN* * +l AN) =Aa KAN+z o *9/ (80HId'ID'O',kGO'ZAI**Z+IAI-iO.LA'CWV)XAJ*il3Q= AI (ZAI**Z+TAI-OZ),*il3O4AN (bOH8'19'0'AGO'O*Z/TAI+iO-LA'ZWV)XAA*il3G=ZAI (0* Z/Ikl+OZ) Ac1130=ZkN (b0HId'T9'O',kGO'iO-LA' TWV) XAhc1130=TAI OZ*1130=IAN (AA*AA+XA3.cXA)I'dbS=lOiA 30NViSIO A UndWOO qwv=cwv Nv=zwv NV=TWV 59 oi 09 (o4olq*((Twv)sev-Nv))jl 0 0 o6 oi ou (T*b3*DVlJl)AI 3nNIINOO 0*041awa O*O=Ziowa o*o=iiawa qwv=cwv qwv=zwv NV=TWV q8 01 09 (Z*3N*TDVIAI)AI (a'Z3'T3'OMi'il3G+3WII)ziawa=Ciawa (G'ZO'13'OMI'OOZ/.LI30+3WII)ziawa=ziawo (alzo,113"OMI"3WII)ZIOWO=TIOWa C**a*zo 6u=Nv (a'ZO"10'OMI'1130+3WII)ZOWO=CWV (a"zo4iolomi"*Z/il3O+3WIi)ZOWG=ZWV (G'Z34T0'OMi"3WII)ZOWO=TWV r*z-oz=rr (oooii*omi)ji r*z-61=rr (*oo5*ivomi)ji r*z-Ci=rr (oooVivomi)ji 05 01 09 (T9b3*9VIJI*ONV*Z*b3*TSVIJI)JI W11=113a (i*b3*OVIJI)J1 0001/((I"0-((OMI)DOIV/*I))**001)dX3=WlI OqT lmiNlbdN 0=19VIAl 6LT 180 160 C 80 C 90 CONTINUE IF(Y.LE.O.O) IFLAG=1 IF(IFLAG.EQ.1) PRINT 111,TIME,TWO,QOXD,XO,YO IF(IFLAG.EQ.0) GO TO 80 VY=0.0 Y=0.0 VX=0.0 X=XO HT=O.O DELT=TIM GO TO 140 COMPUTE X DISTANCE KX1=DELT*VO LX1=DELT*FYX(AM1,VTOTCDX,DO.,RHOR) KX2=DELT*(VO+LX1/2.0) LX2=DELT*FYX(AM2,VTOT+LX1/2.0,CDXD,O.,RHOR) KX3=DELT*(VO-LX1+2.*LX2) LX3=DELT*FYX(AM3,VTOT-LX1+2.*LX2,CDX,D,O.,RHOR) DX=(KX1+4.*KX2+KX3)/6.0 DVX=(LX1+4.*LX2+LX3)/6.0 X=XO+DX VX=VO+DVX COMPUTE TEMPERATURE CHANGE KT1=DELT*FT(AM1,DMDT1,TWO,HTD,TROOM) KT2=DELT*FT(AM2,DMDT2,TWO+KT1/2.0,HT,D,TROOM) KT3=DELT*FT(AM3,DMDT3,TWO-KT1+2.*KT2,HT,D,TROOM) DT=(KT1+4.*KT2+KT3)/6.0 TW=TWO+DT C C COMPUT DRAG COEFF FOR NEXT TIME STEP VTOT=SQRT(VX*VX+VY*VY) RETOT=RHOR*VTOT*D/AMU(TW) CD=0.44 IF(RETOT.GT.500.) GO TO 93 CD=18.5/(RETOT**0.8) IF(RETOT.GT.2.0) GO TO 93 CD=24./RETOT 93 C C C 116 100 CONTINUE COMPUTE HEAT TRANSFER COEFF IF(IFLAG.EQ.1) GO TO 116 AMUW=AMU(TW) HT=HTC(CAIR,D,RETOT,AMUS,AMUW,0.714) DETERMINE NEW OXIDATION PARAMETERS CONTINUE C1=9340. C2= -13760. IF(TW.LT.1193.) GO TO 100 C1=4.68E8 C2= -26670. IF(TW.LT.1493.) GO TO 110 C1=5.04E8 OS 01 09 ON3 dOIS 3nNIINOO 01 ol 09 05 SS r*z-ii=rr (T*b3*9VlAl)AI MI=OMI X=OX A=OA XA=OA AA=OZ 113G+3WIi=3WIl OS 01 00 (0-0-31-A)AI (0*0*b3oA*GNV*O*Z*31*(WOO81-Mi))AI 3 T+iNI 8dN=INI IdN OZI O=iNiHdN 3nNIINOO 11 55 oi oo (rrWIN18dWAI ((*0T3'XS)9)IVW80A III A6X'aXOb'MI43WIi (111'90)UIM A'X'OXOb'MI'3WIl'ITl AINd ZT1 11 oi 09 aXOb'Ml'3WIl (111'90)31lbM axob'MI'M14111 lNlbd ZIT 01 09 (O*b3*9VIAI)AI on oi oo (rr*3N*INIHdN)AI d3lS 3WIl IVNIA ONV SON033S O*Z A SIMUNI iv inOiNlbd 50*0=1130 (0*0031*axob)Ai 0 Z=IDVIAI (0*0*31*GXOb)Al (Twv4wv)*98z* i=axob 3nNIiNO3 Oll OOC9 1- =zo T8T 182 APPENDIX D THE VAPOR PROGRAM INPUT, OUTPUT AND PROGRAM LISTING CONTENTS page Section D. 1 VAPOR Model Equations 183 D. 2 VAPOR Input 186 D. 3 VAPOR Output 189 D. 4 VAPOR Program Listing 191 D-1 Sample Input Listing 188 D-2 Sample Output Listing 190 Radial Conduction Model Control Volumes Used in VAPOR 185 Table Figure D. 1 183 THE VAPOR PROGRAM INPUT, OUTPUT AND PROGRAM LISTING D.l VAPOR MODEL EQUATIONS The VAPOR code was developed to assess the potential for generation of zirconium gas within the inner portions of a piece of Zircaloy-2 clad undergoing rapid oxidation. The postulated mechanism is that the chemical oxidation energy deposited in the unreacted clad may exceed the rate at which the oxidized outer thickness (which has a much lower thermal conductivity) can either conduct away the heat or ablate, with resultant vaporization (Cl). The one-dimensional radial conduction model employed in the code accounts for a fuel, gap and clad cross-section. A finite difference grid for the model equations is shown in Fig. D.l. The finite difference equations for transient conduction are derived from energy balances performed over the control volumes depicted in Fig. D.l. For example, the energy balance for the center fuel/heater node may be written as: -kAdT q"'V + pcV dT = 0 D.1 A finite difference representation of this equation gives: 184 2T ~T' 2 k (2rAr) 1 J+ 2 q"' Ar = 1 _l_~__l D. 2 T aAt Ar/2 which is finally reduced to: 4 T2 - 2T, + Ar qm T, - T, = R -D.3 a At where: q"' = decay heat/input power volumetric generation rate, (W/cm3 a = thermal diffusivity of fuel/heater node, (cm2/S) k = thermal conductivity of fuel/heater node, (W/cm K) (K) T = temperature, Ar = radial distance, (cm) and the superscript "lo" refers to old time temperature values; new temperatures are not superscripted. The remain- ing equations corresponding to Fig. D.1, though more complex, are derived in a similar manner and may be inferred from the program listing. The chemical oxidation energy deposited in the clad during any time period is calculated explicitly, using the isothermal parabolic reaction law. The VAPOR program was not used for quantitative analysis in this report due to lack of modelling of axial conduction effects, mass transfer through the oxide layer, and 185 FUEL GAP CLAD I N N N N N N N N N N 0 0 0 0 0 0 0 >2 N -HAr3 K- KAr -A] 1> FUEL (HEATER) CENTERLINE Figure D.l / ADVANCING OXIDATION FRONT Radial Conduction Model Control Volumes used in VAPOR. 186 thermophysical property data. It was beyond the scope of this project to formulate mechanistic models for clad oxidation. Should more data on non-isothermal zirconium oxidation become available, it is recommended that the heat transfer equations described in this program be joined to the one-dimensional radial mass-transfer equations described in reference Bl. The mechanistic equations described in that report, which neglect energy release via oxidation reaction, yielded predictions in good agreement with experimental data. D.2 VAPOR INPUT The version of the VAPOR code in this appendix was used in attempts to simulate the experimental data presented in Chapter 5. Input to the code is performed using free format READ statements and DATA statements. Input variables are defined below in order of input. Input Variable Name Definition timax Termination time for experimental simulation, (sec) nprntl Number of time intervals between printouts during assembly heat-up nprnt2 Number of time intervals between printouts during oxidation reaction Nominal Value 187 tburn Time at which oxygen is introduced - into the test assembly, (sec) xcl NOT USED 0 xc2 NOT USED 0 delt Computational Time Step, (sec) tempr Temperature of assembly at time= 0, 283 (K) troom Temperature to which assembly ini- 283 tially loses energy, (K) h Convective heat transfer coefficient, (W/cm2 K) n Number of fuel volumes 5 m Number of clad volumes 10 qf Resistance heating power of heaters, - (W/cm3) Additional input values related to thermophysical properties of the assembly have been given mnemonic variable names and may be inferred from the program listing. A sample input for the experimental assembly configuration is given in Table D.l. Results for this case, at time = 360 seconds (since the introduction of oxygen) are shown in Table D.2. 188 TABLE D.l VAPOR Input timax = 3800. Values input via DATA Statements are: nprntl = 60. = 0.002 nprnt2 = 1. h tburn = 3600. qf = 20. delt = 60. 189 D.3 VAPOR OUTPUT The following is a list of output variables, as they are in order in the VAPOR output of Table D.2. Output Variable Name Definition time Elapsed time since introduction of oxygen, (sec) rin Outer radius of the unreacted Zircaloy clad, (cm) qf heater input linear power, (W/cm) qtot Total energy released by the oxidation reaction, (W) first column Radial node number: heater nodes 1- n (n+ 1) - (n+ 3) (n + 3) - (n + m + 3) gap nodes clad nodes second column Radial distance of node from center, (cm) third column Temperature at nodal locations, fourth column Energy released by oxidation reaction during time step, (W) (K) 190 TABLE D-2 Sample Output Listing time= qf-w/cm=20.00 120.0 rin=0.544589 qtot= 0.000000 1 2243.71 2 2243.57 0.070444 2243.02 0.140889 3 2242.14 4 0.211333 2240.94 0.281778 5 2240.19 6 0.317000 7 1594.73 0.400500 0.484000 8 1171.99 0.OOOOE+00 9 0.487263 1171.98 0.0000E+00 0 0.493789 1171.96 0.OOOOE+00 1 0.500316 1171.94 0.OOOOE+00 2 0.506842 1171.93 0.0000E+00 0.513368 1171.91 3 0.0000E+00 4 0.519895 1171.89 0.OOOOE+00 0.526421 1171.88 5 0.0000E+00 6 0.532947 1171.86 0.OOOOE+00 0.539474 1171.84 7 8 0.546000 1171.83 0.6287E+01 0.3772E+03 0 0 0 0 0 0 0 0 0 0 191 D.4 VAPOR PROGRAM LISTING dimension aa(20) ,bb(20) ,cc(20),dd(20),to(20) ,t(20),r (20) dimension condg(3),alphg(3),condc(20),alphc(20),qchem(20),iflg(20) print," input: timax, nprnti, nprnt2, tburn, xci, xc2, delt" read, timax,nprnti,nprnt2,tburn,xci,xc2,delt nprnt=nprnt1 iflg2-0 tempr=283. troom=283. h=0.002 iflg3=0 time=0. qtot=0. n=5 m-10 nn=0 condf=0. 16297 alphf=0.042513 do 4 i=1,3 alphg (i)=0. 1554 4 condg(i)=2.13e-4 do 6 1=1,20 condc (1)=0.30 alphc (1)=0.7118 if1g (1)=0 6 qchem(1)=0.0 npi=n+1 np2=n+2 np3=n+3 np4=n+4 np5=n+5 nmp1=n+m+3 do 5 j=1,nmpi t(j)=tempr 5 to(j)=tempr tai r=troom sig=5.67e-12 eps=0.25 rci=0.484 rco=0.546 rcl=rco-rci rf=0.317 qf=20. r in=rco rct=0. drf=rf/ (n-0.5) drc=(rco-rci)/(m-0.5) drg=rc i-rf r (1)=0.0 do 2 i=2,n 2 r(i)=r(i-1)+drf r (npl)=r (n)+drf/2. 192 r(np2)=r(npl)+drg/2. r(nP3)=r(np2)+drg/2. r(nP4)=r(nP3)+drc/2. do 3 i=nP5,nmpl 3 r(i)=r(i-l)+drc al=4./drf**2 bl=qf/condf cl=l./(alphf*delt) a2=1./drf**2 a3p=(drf/drg)*(4.*rf+drg)/(4.*rf+drf) c3=2.*al e3=8./drg**2 ag=l./drg**2 bg=l./(2.*(rf+drg/2.)*drg) a5P=(drg/drc)*(4.*rci+drc)/(4.*rci+drg) c5=e3 e5=8./drc**2 a6=1./drc**2 b6=1./(2.*(rci+drc/2.)*drc) alp=(rco-drc/2.)/drc cl=h*rco d]P=0.5*(rco*drc-drc**2/4.) elp=(rco*drc-drc**2/4.)/(delt*2.) if(time.eq.0.) go to 37 60 nn=nn+l 61 a3=a3p*condf/condg(l) d3=1./(alphg(l)*delt) f3=a3*cl g3=a3*c3 cg=l./(2.*alphg(2)*delt) a5=a5p*condg(3)/condc(l) b5=d3 d5=1./(alphc(l)*delt) f5=a5*b5 g5=a5*c5 c6=qchem(2)/(condc(2)*r(np4)*drc*6.28) d6=1./(alphc(2)*delt) al=alp*condc(m) bl=sig*eps*rco*(troom**2+t(nmpl)**2)*(troom+t(nmpl)) cl=h*rco dl=qchem(m)/6.28 el=elp*condc(m)/alphc(m) bb(l)=l. zz=al+cl cc(l)=-al/zz dd (1) = (cl *to (1)+bl) /zz do 10 i=2,n-1 zz=2.*a2+cl aa(i)=-(a2*(I.-O-5/float(i)))/zz 193 bb (i)=1.0 cc (i)=-(a2*(1.+0.5/float(i)))/zz 10 dd(i)=(cl*to(i)+bl)/zz zz=a2*(3.+0.5/float(n))+c1 aa(n)=-(a2*(1.-0.5/float(n)))/zz bb(n)=1.0 cc(n)=-(2.*a2*(1.+0.5/float(n)))/zz dd (n)=(cl*to(n)+b1)/zz zz=f3+g3+d3+e3 aa(npl)=-g3/zz bb(npl)=1.0 cc(npl)=-e3/zz dd (npl)=((f3+d3)*to(np1))/zz zz=2.*ag+cg aa(np2)=-(ag-bg)/zz bb(np2)=1.0 cc(np2)=-(ag+bg)/zz dd (np2)=(cg*to(np2))/zz zz=f5+g5+d5+e5 aa(np3)=-g5/zz bb(np3)=1.0 cc(np3)=-e5/zz dd (np3) = ((f 5+d5) *to (np3) )/zz zz=3.*a6-b6+d6 aa(np4)-(2.*(a6-b6))/zz bb(np4)=1.0 cc(np4)=-(a6+b6)/zz dd (np4)=(d6*to(np4)+c6)/zz do 20 j=np5,nmpl-1 i-j-np3 c6=qchem(i)/(condc(i)*r (j)*drc*6.28) d6=1./(alphc(i)*delt) zz=2.*a6+d6 aa(j)=-(a6-1./(2.*(rci+(float(i)+0.5)*drc)*drc))/zz bb (j) =1.0 cc(j)=-(a6+1./(2.*(rci+(float(i)+0.5)*drc)*drc))/zz 20 dd(j)=(d6*to(j)+c6)/zz zz=al+bl+cl+el aa(nmpl)=-al/zz bb(nmpl)=1.0 dd(nmpl)=(el*to(m)+bl*troom+cl*tair+dl)/zz call tridag(l,nmpl,aa,bb,cc,dd,t) go to 62 if(iflg3.eq.1) go to 62 if(iflg2.ne.1) go to 62 iflg3=1 go to 61 62 continue time=time+delt 194 if(nn.ne.nprnt) go to 37 print," " print 28, time,rin,qf,qtot 28 format(lx,"time=",f7.1,3x,"rin=",f8.6,3x,"qf-w/cm=",f5.2,3x,"qtot=",3xel0.4) print," " do 30 j=1,np3 print 32, j,r(j),(t(j)-273.) 32 format(3x,i5,5x,f8.6,3x,flO.2) 30 continue nn=0 do 34 j=np4,nmpl print 36, j,r(j),(t(j)-273.),qchem(j-np3),iflg(j-np3) 36 format(3x,i5,5x,f8.6,3x,flO.2,3x,elO.4,3x,i4) 34 continue 37 continue if(time.gt.timax) go to 70 do 50 j=1,nmpl 50 to(j)=t(j) rhozr=6500. do 80 i=1,3 alphg(i)=4.896e-7*t(n+i)+1.325e-3*t(n+i)-.2197 80 condg(i)=alphg(i)*0.38783/t(n+i) j=nmp1+1 90 continue if(t(nmpl).gt.292.) trom=0.90*t(nmpl) if(t(nmpl).gt.292.) tar=0.96*t(nmpl) if(t(nmpl).gt.1300.) trom=0.89*t(nmpl) if(t(nmpl).gt.1560.) trom=0.88*t(nmpl) if(troom.lt.trom) troom=trom if(tair.lt.tar) tair=tar if(time.lt.tburn.and.iflg2.ne.1) go to 120 if(iflg2.eq.0) nn=0 if(iflg2.eq.0) time=delt iflg2=1 if(t(nmpl).gt.1323.) eps=0.25+4.5e-4*(t(nmpl)-1323.) nprnt=nprnt2 j=j-1 qchem(j-np3)=0. xc1=9340. xc2=13760. if(t(j).lt.1193.) go to 100 xcl=4.68e8 xc2=26670. if(t(j).lt.1428.) go to 100 xc1=5.04e5 xc2=14630. 100 continue if(iflg(j-np3).eq.1) go to 90 if(j.1t.np4) go to 120 rct=rco-rin ratek=xcl*exp(-xc2/t(i)) w=(ratek*delt)/(rhozr*rhozr)+rct*rct rcn=sqrt(w) 195 rinn=rco-rcn if(rinn.gt.(r(j)-drc/2.)) go to 110 rinn=r (i)-drc/2. iflg(j-np3)=1 110 qchem(j-np3)=(rin**2-rinn**2)*7.8e4*3.1416/delt rin=rinn if(iflg(j-np3).eq.1) go to 90 120 continue xnu-4.2 tfilm-(tair+t(nmpl))/2. h=xnu*(1.Oe-4+5.913e-7*tfilm)/(2.*rco) do 130 1=1,m ts=0.001*(1.8*t(np3+1)-32.) condc(1)=((((.20635*ts-.5567)*ts+.6748)*ts-.13153)*ts+.68244)*.20768 if(iflg(1).eq.1) condc(1)=0.80*condc(1) ccl=4.184*(7.le-2+1.75e-5*t(np3+1)-0.89e3/t(np3+1)**2) if(t(np3+1).gt.1223.) ccl=0.364 alphc(1)=condc(1)/(6.5*ccl) if(iflg(l).ne.1) go to 130 tcon=2.667e-3*t(np3+1)-3.12507 condc(1)=1.531e-3+6.027e-4*tcon+1.0434e-4*tcon**2 go to 133 ccl=((3.764e-9*t(np3+1)-1.667e-5)*t(np3+1)+0.02457)*t(np3+1)-0.81896 133 continue alphc(1)=condc(1)/(6.5*ccl) 130 continue do 140 k=1,m 140 qtot=qtot+qchem(k)*delt go to 60 70 continue end subroutine tridag(if,1,a,b,c,d,v) dimension a(1),b(1),c(1),d(1),beta(35),gam(35),v(35) beta(if)=b(if) gam(if)=d(if)/beta(if) ifpl=if+1 do 1 i=ifpl,1 *c (i-1) /beta (i-1) beta (i) =b (i) -a (i) *gam(i-1) )/beta (i) 1 gam(i)=(d (i) -a (i) v (1)=gam(1) last=1-if do 2 k=1,Iast i=1-k 2 v(i)=gam(i)-c(i)*v(i+1)/beta(i) return end 196 APPENDIX E CLAD INPUT, OUTPUT AND PROGRAM LISTING CONTENTS page S ection E.1 CLAD Input 197 E.2 CLAD Output 202 E.3 CLAD Program Listing 205 E-1 Sample Input Listing 200 E-2 Sample Output Listing 201 Table 197 CLAD INPUT, OUTPUT AND PROGRAM LISTING E.1 CLAD INPUT The CLAD code, used to analyze the experimental test assembly, is based on the stand alone SFUEL code which was developed to analyze open frame spent fuel holders in NUREG/CR0649. Input is entered under the heading $INPUT and is defined below: Input Variable Name Description DELH Energy released by chemical Nominal Value 7.8e4 oxidation reaction, (J/cm3Zr) DELT Computational Time Step FL Active fuel (heater) length, 33. (cm) FRAD* Radiative view factor (a dif- 0.0 ferent formulation of radiative heat transfer is employed in this code) G Nitrogen mass flow rate (gm/s) HWOUT* Initial value of heat transfer coefficient from exterior of assembly, (W/cm2 K) 0.0 198 IH20* Flag indicating presence of water 0.0 in the storage assembly N Number of axial node interfaces NPRINT Number of time intervals between 11 - printout NROD Number of rods in assembly NTIME Number of time steps for program 9 - run POWO Total power input to heater 1.35 array, (KW) RC Clad outer radius, (cm) 0.546 RCI Clad inner radius, 0.484 RF Radius of heater element, (cm) 0.317 SMB* Axial power profile parameter 0.0 TIMEO* Decay heat parameter 4.0e6 TO Initial temperature of assembly, (cm) 283. (*K) TSAO Ambient temperature outside the 283. canister, (*K) W Width of the zirconia duct in 5.08 which the rod array is placed, (cm) XCS* Distance between clad and struc- 0.0 ture, (cm) XOX* Initial oxide thickness, (cm) 0.0 199 XS Thickness of zirconia liner, 0.254 (cm) XW Radial average thickness of 5.84 alumina fluff, (cm) Note that the asterisked (*) quantities are input in the present version only to satisfy the namelist read statement; otherwise, these variables are not used in the program. A sample input listing for the experimental simula- tion is given in Table E-l. 200 TABLE E-1 Sample Input Listing $INPUT RF = 0.317 RC = 0.546 w = 5.08 xS = 0.254 FL = HWOUT = 0.0 Powo = TIMEO = 4.0e6 TSAO = SMB = 0.025 FRAD = 0.0 TO =-283. RCI = 0.484 xox = 0.0 xCS = 0.0 NROD = 9. N = 11. NPRINT = 6. NTIME = 360. IH20 = 0. DELT = 10. xw = 5.84 G = 12.0 38.1 1.35 283. TIME = 0.i0OE+02 PEXIT - JT 36 DELT TCENT 0.33706E+03 0.43768E+03 0.52304E+03 0.58551E+03 0.62817E+03 0.65504E+03 0.66863E+03 0.66422E+03 0.60642E+03 0.32986E+03 RIN(I) 1 2 3 4 5 6 7 8 9 10 EGEN ORAD PGEN POGEN 0.54600E+00 0.54600E+00 0.54600E+00 0.54600E+00 0.54600E+00 0.54600E+00 0.54600E+00 0.54600E+00 0.54600E+00 0.54600E+00 * 0.000E+00 G - 0.252E+00 PHI - 0.000E+00 KIT a 0.100E+02 TCAVG 0.33679E+03 0.43735E+03 0.52262E+03 0.58504E+03 0.62767E+03 0.65453E+03 0.66813E+03 0.66377E+03 0.60622E+03 0.32986E+03 OF(I) 0.13500E+03 0.13500E+03 0.13500E+03 0.13500E+03 0.13500E+03 0.13500E+03 0.13500E+03 0.13500E+03 0.13500E+03 0.13500E+03 TAVE( I) 0.12851E+03 0.31371E+03 0.42830E+03 0.51054E+03 0.56841E+03 0.60695E+03 0.63010E+03 0.63812E+03 0.61626E+03 0.49024E+03 QCHEM(I) 0.OOOOOE+00 0.00000E+00 0.00000E+00 0.00000E+00 0.OOOOOE+00 0.00000E+00 0.00000E+00 0.00000E+00 0.OOOOOE+00 0.00000E+00 TSIN(I) 0.13170E+03 0.30691E+03 0.41443E+03 0.48974E+03 0.54122E+03 0.57440E+03 0.59339E+03 0.59798E+03 0.57179E+03 0.44325E+03 OCOND(I) -0.38587E+01 -0.26516E+02 -0.22487E+02 -0.16458E+02 -0.11241E+02 -0.70829E+01 -0.35876E+01 0.11488E+01 0.15178E+02 0.72876E+02 TSOUT( I) 0.10130E+03 0.22635E+03 0.29522E+03 0.33863E+03 0.36540E+03 0.38098E+03 0.38887E+03 0.38912E+03 0.37051E+03 0.28353E+03 OT(I) 0.14596E+03 0.10975E+03 0.95152E+02 0.82205E+02 0.70012E+02 0.59349E+02 0.49604E+02 0.35272E+02 -0.74332E+01 -0.15678E+03 TWIN( I) RHO(I) TWOUT(I) 0.10130E+03 0.22635E+03 0.29522E+03 0.33863E+03 0.36540E+03 0.38098E+03 0.38887E+03 0.38912E+03 0.37051E+03 0.28353E+03 O.10000E+O2 0.12867E+02 0.13709E+02 0.14192E+02 0.14472E+02 0.14631E+02 0.14719E+02 0.14729E+02 0.14479E+02 0.13076E+02 0.12471E-02 0.67867E-03 0.54012E-03 0.47107E-03 0.43149E-03 0.40803E-03 0.39434E-03 0.38730E-03 0.38739E-03 0.40682E-03 EXI(I) EX2(I) FOX(I) 0.29943E+01 0.88247E+01 0.14818E+02 0.20831E+02 0.26287E+02 0.30716E+02 0.33876E+02 0.35480E+02 0.34250E+02 0.25763E+02 0.00000E+00 0.20750E+00 0.24016E+00 0.25245E+00 0.25433E+00 0.25112E+00 0.24549E+00 0.23686E+00 0.21417E+00 0.13022E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.23000E+00 0.486E+06 ECHEM = 0.000E+00 EFUEL 0.303E+06 ECONV * 0.172E+05 DELE 0.201E+05 0.766E+05 OSTR 0.687E+05 ECHOX 0.000E+00 0.135E+04 PCHEM 0.0O0E+00 PCONV 0.107E+03 PRAD * 0.307E+03 POUT * 0.234E+03 DELP = 0.15E+02 POCHEM 0.000E+00 POCONV 0.302E+01 0.594E+01 PDOUT a 0.549E+00 PDRAD - 0. 702E+03 I5(I) 0 0 0 0 0 0 0 0 0 0 202 E.2 CLAD OUTPUT A sample output for the experiment simulation is shown in Table E-2, corresponding to a step in the transient heat shown in Fig. 5.4. The output corresponds to the pre-oxi- dation conditions of the experiment, for an elapsed time of 360 seconds. As described below, the output variable "TIME" gives time from the introduction of oxygen, while the total elapsed time for the experiment is given by: TOTAL ELAPSED TIME = JT * DELT (seconds) The output is defined below in order of appearance in Table E-2. Output Variable Name Description TIME Time from introduction of airflow (sec); during pre-oxidation heat-up, this equals DELT PEXIT Not used in this program version G Total mass flow rate PHI Not used in this program version (02+ N2), (gm/s) 203 KIT Number of successive substitution iterations performed to properly account for radiative heat transfer JT Number of time intervals elapsed DELT Computational Time Step I Axial nodal location, where I= 1 is the base of the assembly and I= 10 is TCENT the top Temperature of center rod in the array at axial location I, (*C) TCAVG Average temperature of all rods at location I, (*C) TAVE (I) Average airflow temperature at location I, (*C) TSIN (I) Temperature of air-side of zirconia liner, (*C) TSOUT (I) Temperature of alumina fluff side of zirconia liner, (*C), (also equal to TWOUT(I)) RHO (I) Density of gas flow at location I, (gm/cm ) RIN (I) Outer radius of unreacted clad, (cm) QF(I) Heater input power in the rod array at axial location I, (W) QCHEM(I) Energy released by chemical oxidation, (W) QCOND(I) Energy conducted into heater/clad at location I, (W) QT (I) Total energy exchange between rod array, airflow and zirconia liner, (W) 204 EXl(I) Energy absorbed by zirconia liner, (W) EX2(I) Energy removed by natural convection from the exterior of the canister, (W) FOX Oxygen mass fraction at axial location I IS(I) Flag indicating regime of oxidation kinetics 1 - kinetics rate-limited 2 - diffusion rate-limited EGEN PGEN PDGEN Energy, power and heat flux input to system 2 via heaters, (J), (W), and (W/cm ) respectively ECHEM PCHEM PDCHEM Energy, power and heat flux via oxidation 2 reaction, (J), (W), and (W/cm ), respectively EFUEL Energy stored in the heater and clad, (J) ECONV PCONV PDCONV Energy, power and heat flux input to airflow and convected out of the assembly, (J), and (W/cm2), DELE DELP (W), respectively Energy and power imbalances-present version does not account for convective loss from outside of canister, (J), QRAD PRAD PDRAD respectively Energy, power and heat flux out of the ends of the rod assembly by radiation, (J), (W/cm2), QSTR POUT PDOUT (W), (W), respectively Energy, power and heat flux to zirconia liner 2 and insulating canister, (J), (W), (W/cm2) respectively ECHOX Energy stored in oxidized clad layers, (J) (113a',c Z) / (*i/ (ZO* -Zda *LW)=dSJ Z-da/ CZ/Zba-LWI =5V (Zba-%.9b Z)/P1 =98 (~0~~Z) P' I=58 Z~~) /I' i= ~~ait ~ (Za'cbc /t0~C1 / (0+W 1=V ~ ±130!' t=dZO (±130* Z)/(I (TG0' ItM) +t U* I b) =d 10 *l/IbO+lb=dtJ lbG+Tb=9b '1 /M= IN~ S*m~b S.i/sx=tba *'OZi l3W1 O=W3HON ClI I , 11H1 1I X8 / Z II lVlIZ 39 / W13wIN,, XL 11 INI dN,i X9 AiN1 C XII 11 AMN 1 X8 ,1SOX,, X6 ,,xox, x6 ,iJiobi x6 ,1130,1 X8 Ali XQI Z ,,avbj 11 xS / 711301 / 11ews,, x6 110VSI 1i X9 ,03W1.I 11 xL A1 MUd 1 , Xg I ,a1flMH,, X8 O,, X01 g,SX i XOI Mll XIi ,OH i0 Ji XOi)IVWNO 91 1 11 imWoi infdNi 3Hi SI SIHJ. QIHI '3WIiN'1NIlJdN'N'0ObiN'SOX'XOXI 'I'13'J.a'0V'8WJ'OYS'OS3WII'OMOdnOMH'13'SXM'3OI'AJ (9i'9) 31Ibm (indNI '5) 0V3N H13a ' OV3bI UA'MX'9'1130'OZHI '3WIiLN'.LNI~dN'N'aObIN'SJX'XOX'1 Ii 'O.L'GVbU'8WS'OVSI'O3WIL'OMOdlflOMH'1I 'SX'M'JU' AU /.LldN I/i.SI13WVN 13SVOS 11VO H13a'Il3al'b'I~b4Xl3'~V NOWWOO 1 T1i/Id VILVO / Ci ) ~~iNQISN3WIQ (Ti)1X3'Cii)iX3 NOISN3WI0 '(Ti)N31' (1i) 3AVI'" (ti.fl (ii)NIMI' Ciu) sob" (it)Az (Ii)OHNU(1)1 (ii) iO.nosi' Cii)NIS1' Cu) sbs (tii' CII) 0N03bi '(1)01' ii)W3H?~ C~)~ Ci~d'Ci)±'(113± (l~xNOISN3WIO CL) ao' L)033' L) se' L) CH'L) 010± NOISN3Wia C.ndino'±ndN0)13fljs WVM0~d vv'(hi'L)M3N±' DNI1lsirI WVHDOdd 3 vqVTD Id-=IdV in*OOHb*DWS-Id=Id IJ*OMJ*9WS-=Id (Oi,%cVb)/Od=OOHb *06=ows OMOd*'0001=OMOd 91 99Z'O=MOdVO SU*O=MOdV 91 01 00 (9+3* *11*03WIi)JI OWO=MOdVO iSZ*O=MOdV ((*I-Sd3/*1)3.tSV/(GObNY/*I-*I)).'.bV+3d3/*I)/5*0=S3d3 M*M=Gvbv GObN*3b'%e3b*Id-M*M=V OV+JV=IV HIM* ( IOb* IOb-O J*Ob) * Id=OV GObN*Jb*J I*Id=JV GObN*XI30*0b*Id**Z=HV SO=OSO SOS*SOHb*XI30*SX*M**q=SO X13a*M** *biSJ=SV (I -N) (1) X** Z-1 A) =X130 OZ-3*1=(I)X 5*9=OOHb Lz*C=AOHb (91*8z/LL*0+00*ZC/Czo)*L+3q IC*8=V I 9+35MO'T=Od ZI-3L9*5=ois 01-3*1+01=031 W'ZT39 / I llmxll XOT lox,, XOT oxoNws,, xL ,mNws,, xg ,sNws,, )(8 ,Hisj,, x8)ivwboA Li MX'DX'XONWS'MNWS'SNWS'HiSA (LI49)31lbM CZ500=mos C0Z'O=SOHH q8001=s3s 96o*O=MOHb (01-(Sd3/*I)+(XOd3/*I))/*I=S3d3 (*0*10*XOX) Al (*I-(Sd3/*T)+(3d3/0T))/500=S3d3 ((OT-3d3/"1)3.c(*I-GObNX)/*T+3d3/*I)/5*0=iN30d3 (aOHN)ivoij=aOHNX 59*=XOd3 Z*O=Sd3 0O=3d3 qiLeo=ox Z9000=XONWS U100=MNWS 6zioo=sNws O*I=HISA 9*i5=in 5=M3NHdN OZ+3*T=INd3I *9=M3NMS *000c=MS31 OZ+3*1=XVWOI 90Z 3flNIINOO 5Z 0=(I) WHOb 0= (i)wxo £Zeo=(I)xoj o00=(I)W3H3b o0o=(i)mb ooo=(i) sb oo=(i)ibL 0*0=(I) 0N0Jb Ml~ 0OHbJ= (I1) 0*0=(I)d oi=(i)inomi o±=(I) NIMi O±=(i) .LfOSJ Oi=(I) NNIS.L O±=(I) NISI 01= (I) I oi=Ci'r)aioi ot L'i=r 01 00 001.=(I)NXOIi 0J.L=(I) NINJJ. 03=(I) JNOI Oli= (I)N1U 0=(1)113W 01= () 011 001=(I) NOI 00.1= (1)31 X13a+(I) X= (1+1) X (xox'qoos9/oGT1) tVwV=0)iOH TWN'I=I 5Z 00 I -N=TWN * I+MOdV-=IMOdV VHi/X13~0*0d*DWS=DWSM T rz3'UM0d 31AV)VUH~TI' HI)IYWH0A ZZ (* IM (ZZ69)llI1IM OMOd/1 A*S=IM ( ZT3i i,1Dws, x6 gAM,~g x6 o,~3 x6 / *TI301 / ,3, XOi C ,,Mdii X01 il1dit XOI i,00H~i X8 Ai X11 013t XOI OM~dii XS 1 1SV,i X01 Z 1 1 1OT3I01 , Xt1Xl3a,, xS i Xu T IVX01 / 11OV,i XOT ,Vii XIT ,i.V,X0I/ DWS'3)IX63)IX'3aMd'ld00HHI I IVW80:1 3718VI V dfl13S 31-1 SI SIHI 0*0=3NX 00O=ONX Mdl/V*c 'T] 0 G0H~N*OH*~I d*s +M* 'T=Md LOZ 208 C C TC (N)=0. TCN (N)=0. TCX (N)=O. TCXN (N)=0. TSIN (N)=O. TSINN(N)=o. TSOUT(N)=0. TWIN (N)=O. TWOUT(N)=0. RHO (N) =0. QF (N)=O. QT(N)=O. QS(N)=O. QW(N)=O. QCHEM(N)=0. FOX(N)=0.23 EGEN=0.0 ECHEM=0.0 CHOX=O. EFUEL=0.0 ESTR=0.0 ECONV=0.0 QRAD=0.0 QSTR=O.0 TIME=O.0 DELTO=DELT NP=O F=1.0 G=G*0.021 START TIME LOOP DO 70 J=1,NTIME JT=J TIME=TIME+DELT DO 30 I=1,NM1 T (I)=T I0(1) TC (I)=TCN (I) TCX (I)=TCXN(I) 30 TSIN(I)=TSINN(I) START DETERMINATION OF AIR PROPERTIES GNI=0.77*G GOX(1)=0.23*G GI=G DO 38 I=1,NM1 KIT=O 11=1 FOX(1+1)=GOX(1+1)/(GoX(1+1)+GNI) Gl=.5*(GOX(I)+GOX(1+1))+GNI 45 KIT=KIT+1 CALL APROP(GI,DE,TSIN(I),T(I),P(I),RHO(I),CP,RE(I),HSIN,SMF,A 1 ,FOX(I),CPOX,CPNI,RA,HDR(I)) CALL APROP(GI,DETCX(I),T(I),P(I),RHO(I),CP,RE(I),HR,SMF,A 1 ,FOX(l),CPOX,CPNI,RA,HDR(I)) IF (HSIN.EQ.0.) HSIN=1.E-20 WK1=RHO(I)*CPNI*A*DELX/DELT +(1) N3±i( 1) NOI) CD- M~ vv (ZG-ZV) - (Z)vv 1) I)1O.Lc IO+(1) NODL'I S+(I)3AVic IV= (1)00 o3-=(I) 33 GHdIV/dS3=53 8>Iws/L~i1flOMH=L9 (il3O*.8HdIV) / i=£ (1130.c'VHd1V) / t=£ dCV*c(SNWS/VNWS) =Cv VHdl V/dZO=ZO dlG*.SOS*cS0Hbi=l0 dlO*.VNWS=t 3 1 ) NIS.LI (1I)N3.L) I bhS3d3*c9 I S*I d/ * frt1 tINISH=IV MNWS=8)IWS SNWS=V)4WS (M3ScM0Hb)/M)IWS=SHdIV (S3S*.SOHb) /SNWS=VHdIV -3686+C-3Z9 t=M)IWS ii1*L1* t -3£9I 'Z+i1 LOO=MOs (00£P±9*(I)NIM)dI Vro+(I) NIMI*c'-356*i=moS 1** 3 0*1iTo +i7 -3LZ9+-3lSI=SNWS 96812o- (I)NIMI~c (LSZO*O+(i) NIMIb.*(S-3L991- (I)NlMIAC-39L£))=S3S OZ-3*i=ifOMH (*O~3.lOtH) Al (±nOMH'OVSI'(i)imM' (I)X)ANO3M 11VO 3bufli.f1s aNV IIVM HoflObHl S3~flJV~dW3.L M3N 31LVin3lV3 3 3flNI.LNOO M£ (N)I*~ (I Nd3*~ I ND+X~d~.c (N)X09) =ZN03 £ 01 09 (IWN'l.VII) Al 3fliNO3 ££ (I).*~(INdO.cIN9+X~d3O (I)X09) =NOJ3 CC 01 09 (1 Ho, ) AI (I)1= (1)0 I11 HiivN3dW3.L HIV M3N U1vinlflV )M+4m+NM+l NtM=SNM INd3S9=T)M Xl 30AcV'tH=0)M Xl30acSV*cN ISH=ZNM 3 6OZ 31* (3NWS*3V+JNWS),'V) /(C1) - (1-1)01* Xl 301 (NWS*3V+JNWSJV) ) =(1) ONOJo 0c, =i)WS L6z9i 'o=jiws P1 01 09 (POb3xt)l1 I=XI IWN't=t 8P 00 0=11v31 31 010 HilM Xfl1 IV3H NOul20NO3 3 31 11V031i 3NlNINO3 Si Mb= (I) X3 (I) (1) sb= C1) 1 X3 (1) S30+(ONISi- (I)3AVi))vNISHv.SV=(I) sb (1) SiO+((I) 3AV- (1)X3i) bHvbV= (I)A (ONI Sl-(1) X31)vS3Hv-(O0bNX/'* I-'* )v.bV=(1) sob ISl+Zv.( I)X31) * (ON ISi+ (I)X31) * IS*S3d3=S3H (I)M Idv'Z= (OVSi- () ifoMi)10MHX130Ll (iI'rr) M3N1= (I'rr)01O± 171 L't=rr 'i 00 (I'L)M3N1=(I)1noMi (I)NIMI= (I)iloSi (I'C)M3N1= (I)NIMi 3flNIINOi r' Si 01 00 (ZOO'9'(ONISi-(I)NISi)SV)I1 Ol1 019O (501q9i)JI (I ')M3NI=(I)NISi (rr)Wbdl=CI 'rr)M3NI £1 L'i=rr it 00 (I)NISI=ONISi (EvvON 3 00 (I 'L)10i1v5+0VSi''Le= (L) 0'0= L)33 5v-= WL vv (I'9) 010n13= (9)00 (9G+%V)-=(9)33 17i+'TV~ = (9) 68 (91-1)V) -= (9) VV 0S8+%7v M= CS0 3 00 = )( (I'E) 010i+ ZS-7)-= CM vv (I '17)* 010a+C17= (1)00 C3-= C0)30 t3+toGt9+t3= Ct) 1130/ *0=(I)W3H3O *0=(I)WXO ill oo59/oti=(i±, 6li DNI1V3H IVOIW3HJ ON 3 05t 01 0D ONC8V= (1) W3H3b 1() 1J1I-NO I)'.' (H130) =Ob u8V/1130.c0WX04 ((*9* )/Z 16) +(1I±J=NOJU GWxO=(I)wxO Z= (I)S I 0311W11 N0IsfliOi - z=si 6l i 01 0 (awxo±1ii*wxo)i 3 11 30/ (NN I Id- (I1) NIN)*H130=0b £iTI 01 09 (o031'NWXO) JI NWX0=(a)wxO T==() S I Q311WII S313NIN 3I109VHVd - T=SI (1)X0J.c (I1) 0HHI~ (1)HOH~cuJV=GWX0 ~bN3=(I)1U V, 1130/ ( (1)I-t NOH) 3.4H130=1 Ob INOH-OH=(I) 3NIHJ u8V/O0uNX*c1130,.cWXO*9S8V1.0+ (I).L=t NOu OwxO=(I) IWXO L'i 01 09 (awx0.L1-Nwx0)ni INNIH=(I)3NIH (aWXO*11*NWXO)31 0=(i)lWxO C*OJ(I*Nxod)1 )1wx0=(I) TWX0 (1)X0J* (1)MIu.* (1)HH*IV=aWX0 0=tDb (o031xwx)4WX1 113/ IN a -11)3a/N IH (I)oNx/II)*HIv=iob (1a)XOd'INN I H(1) O N'lb'T NON '(1)Id' (1)1NOI) W3H3 11V3 6qi 01 09 (1b3*oZHIvuO'O~3*W3H3N) JI Pi 0109D (50ooo0o1 *(I t- (1)N11) s~V)A (I)NTN3±=(a) tNO1 X13aOV*30 HHA~cO O =3 Miot=jo lf9*o=t33 (*i.LTID*a)NO)JI IWN'tIl o5 oa (I-N) 3±*50*ONWS cV+ W - V(N) 0N03b (N.c0o- * (1) 31) *OTV* Sd3*.9I S-= (I)GN03b saoH Ao SON3 WObJ xnii 1V3H N011VIaVN 31Vifl1vo 3flNIINO3 O 3 TTZ sv/lfl~d=iflOGd 11 30'.W3H~d+W3H3 3=W3HJ3 IV/W3H~d=W3HOad (i)SO+±id=±flOd ((I) W3H~b- (I)WHb) + (I)W3H30+W3H~d=W3H~d IWN'I=I £S 00 0o=LfOd O=W3H~d 5'(9)'rZ0 / (+38L',tZZ 16) )+XOH3=XO0H3 113'.' (N)X09- (1)X09) 11 3a,,c vMd+0Vll=0Vib (MV~. Z) /GT~d=0V~0d ()0GN030- (N)G0O3=0V~d 11 30*.ANO3d+ANO3 3=ANO033 (SV+ V) /ANO3d=ANO3Od I NO33-ZN003=AN03d 1130'.N3Dd+N3O3=N3O3 WJVN30d=N3 D0d J*0M0d=N30d swflS ADOi33 31Viflo1vo 3flNIINOJ 05 (INO+ (1+1) X09) / (1+1) X09= (1+1) X0J 0=UT+I)X09 (o0'31*(I+I)XOD)Hl 1I) twxo-( 1)wxo- (I) xOo=(1+t) X09 q*Czo= (I)xoo 0 0+d.LS3=U.LS3 (I) NIS.L=(I) NNISJ. 3bfl.Vb3dW31 3bfl13flis M3N 31vifl1vo 0 031=(I)NxOI (1)N31= (1)NI NO (00±11iI)Nx3I) Al 05Z 01 09 (o031*xOX) 31 (I) N3i= (I) NX3J. 1 *(1) NI N3) A 3flNIINOJ Z5 ( (1)ND (I)NOI=NNOI Z5 0± 00 ((I)W3HJO±1i*I)WH3O)AI (1'() 01L- **. (1) 1 N31L) *'00MN/MV' I ((0)lb- (1+1) 0N03b- (1)aN03b+( I) W3H3b+(I) J0) * (0/1130)+(1) 01=(1) N3J. 3flNIINO3 P1T 05Z 0± 09 (I)INOI=(I)NINOI 0 (1)01= (1)NO1 P'T 0± 09 (-0±1(I()3QNbZ)JI PIr 01 09 (±13W±11-(1)0±)iI T==() 113W (113W±LD* (1)I) 31 a0N.cX13'.c d'.'(Z**10 I3 ''TI' CIN3 3nN IiNO3 oL 3flNI INOO oLt 3flNI iNOO I=W3H-3N (oz±9±~r)Ai 000=3W11 (I3JHN) Al 9=iNI~dN (Ovioor)Ai zL 0±0o (XVW3Ir39*(I)N3I) Al IWN'I=I oLT oa 3fliNO3 69 0=dN (C.Z13,,= iflOCd 1 1 'ZI 0 = 0d Jt '~Ih=N390d 1'xi)i.vwnA 691 ' *Mot= ANO00d it'tMii= W3HJ13d ifl0d'CIVbCd'ANQJGd'W3H30d'N30d (69T'9) 3.IbM d-1304inl~d'GVbd'AN0Jd'W3H~d'N39d (S9'9) 31IJM GVbb I'XI)I.VWb0J ( Z3, XOH33 i'C'Z3,,= IdLSO II~T3, £9 X0H-3'Lusb'0V CL9'9) 3itbM 313a'ANO03'-I3nU3'W3H33'N393 (IWN'T=I (99'9)3i;bM '(I) SI' (I) X0AI ((505138Z HI)!! 11(I)0HH,, x6 , 1(l)inOMI,, XL ii(I)NIM.Lo Xg 11(i)inosi,, xLt (IWN't=I'(I)0HbJ'(I)30MI I iVWbI0A 319VIbVA ifidiflO 31-11 SI SIHJ. (io,,=() LvwoA 9 (ioZT3,= 113 AV.() 3AVlb 11Lag- (I) t~(I, XJ±=(I) JX1L S50 Wn'tINO 69 0± 0 (NI=NL)3N) I *Uz-(~inoi=(dNodN ~±SO0V~OAN0J-U±S-13lZJ-W3HJN 9=l3130 idW3inos =d1 30s (lNSI()3S ifl~d-0V 11 214 SUBROUTINE CHEM(TCRCT,RCTTQC,RIN,RINNFOX) COMMON AR,DELX,RCI,RC,DELT,DELH QC=0. RCL=RC-RCI IF (RCT.GE.RCL) GO TO 40 C1=9340 C2=13760. C3=0.0 IF(TC.LT.1193.) GO TO 10 C1=4.68E8 C2=29000. IF(TC.LT.20000.) GO TO 10 CI=5.04E5 C2=15630. 10 CONTINUE RHOZR=6.5E+3 RCT=RC-R IN RATEK=C1*EXP(-C2/TC) W=(RATEK*DELT)/(RHOZR*RHOZR)*1. + RCT*RCT RCN=SQRT (W) IF(RCN.GT.RCL) QC=O. IF(RCN.GT.RCL) GO TO 41 RINN=RC-RCN RCTT=RIN-RINN+1./6500. IF(RINN.GT.RCI) GO TO 40 41 CONTINUE RINN=RCI RIN=RCI RCTT=RCT QC=O.0 40 CONTINUE RETURN END 215 SUBROUTINE APROP(GDETC,T,P,RHOCPREH,SMF,A 1 ,FOXCPOXCPNIRA,HD) AMOX=32.00 AMNI=28.16 AMAIR=1./(FOX/AMOX+(1.-FOX)/AMNI) RU=8.3144E+7 RA=RU/AMAIR P0=1.01325E+6 TBAR=(TC+T)/2. RHO-PO/(RA*T) CPOX=0.27*4.184 CPNI=0.27*4.184 CP=FOX*CPOX+(1.-FOX)*CPNI XMU=0.146E-4*(TBAR**1.5)/(TBAR+109.58) RE= (G*DE) / (XMU*A) PR=0.714 SMK=CP*XMU/PR SC=0.748 DIF=XMU/(RHO*SC) IF(RE.GE.3000.) GO TO 10 XNU=10.02 SMF=25./(RE+1.E-5) GO TO 15 10 XNU=.023*(RE**.8)*(PR**.4) SMF=.00140+.125*(RE**(-.32)) 15 H=SMK*XNU/DE HD=DIF*XNU/(DE*2.) RETURN END 216 SUBROUTINE WCONV(XTWTOHW) SIG=5.67E-12 FWR=O. EPW=0.4 EPO=0.7 EPWO=1./((1./EPW)+(1./EPO)-1.) SMG=980. P0=1.01325E+6 RA=2.768E+6 C C 10 20 1 2 TBAR=TW-0.38*(TW-TO) BETA=1./TBAR RHO=PO/(RA*TBAR) XMU=0.146E-4*(TBAR**1.5)/(TBAR+109.58) IF(X.LT.1.E-10) GO TO 20 GRX=SMG*BETA*(RHO**2.)*(X**3.)*(TW-TO)/(XMU**2.) IF(GRX.GT.1.E+9) GO TO 10 LAMINAR BOUNDARY LAYER ON WALL IF(GRX.LT.O.O) GO TO 20 XNU=0.360*(GRX**0.25) GO TO 20 TURBULENT BOUNDARY LAYER ON WALL XNU=0.1160*(GRX**0.333333) CONTINUE CP=0.28*4.184 PR=0.714 SMK=CP*XMU/PR HW=SMK*XNU/X HWR=FWR*SIG*EPWO*(TW+TO)*(TW*TW+TO*TO) HW=HW+HWR RETURN END SUBROUTINE TRIDAG(A,B,C,D,V) DIMENSION A(1),B(1),C(1) ,D(1),V(7),BETA(10),GAM(10) BETA (1) =B (1) GAM(1)=D(1)/B(1) DO 1 1=2,7 BETA (I)=B (I)-A (I)*C (I-1)/BETA (I-1) GAM(I)=(D (I)-A (I)*GAM(I-1) ) /BETA (I) V (7) =GAM (7) DO 2 K=1,6 1=7-K V(I)=GAM(I) -C(I) *V(1+1)/BETA (I) RETURN END 217 REFERENCES B-1 Benjamin, A. S., et al., "Spent Fuel Heatup Following Loss of Water During Storage," NUREG/CR-0649, (March 1979). B-2 Biederman, R. R., Ballinger, R. G., and Dobson, W. G., "A Study of Zircaloy-4-Steam Oxidation Reaction Kinetics," EPRI NP-225, (September 1976). B-3 Best, F. R., personal communication, Massachusetts Institute of Technology, August 15, 1982. B-4 Baumeister, T., and Marks, L. S., Mechanical Engineer's Handbook, McGraw Hill Book Company, (1958). C-i Cook, B. A., and Hobbins, R. R., "Fuel and Cladding Structures Formed During Severe High Temperature Transients," Proceedings, Am. Nucl. Soc. Topical Meeting, Sun Valley, Idaho, (August 2 - 6, 1981). C-2 Chilton, T. H., and Colburn, A. P., 26, 1183 (1934). E-1 Edwards, D. K., Denny, V. E., and Mills, A. F., Transfer Processes, Holt, Rhinehart and Winston, (1973). G-1 Gear, W. C., Numerical Initial Value Problems in Ordinary Differential Equations, Prentice Hall, Inc., (1971). H-1 Hayes, E. T., and Roberson, A. H., "Some Effects of Heating Zirconium in Air, Oxygen and Nitrogen," J. Electrochem. Soc., 96, 142, (1949). H-2 Holman, J. P., (1976). H-3 Hartman, T., Nagy, J., and Jacobson, M., "Explosive Characteristics of Titanium, Zirconium, Thorium and Uranium," Bureau of Mines, Investigation 4835, (December, 1951). H-4 Hamming, R. W., Numerical Methods for Scientists and Engineers, McGraw-Hill Book Company, (1962). K-i Kendall, L. F., "Reaction Kinetics of Zirconium and Zircaloy-2 in Dry Air at Elevated Temperatures," Hanford Atomic Products Operation, Wash., Contract No. W-31-109-Eng.-52, (September 1955). Ind. Eng. Chem., Heat Transfer, McGraw-Hill Book Company, 218 K-2 Kanury, A. M., Introduction to Combustion Phenomena, Gordon and Breach, (1977). K-3 Kayes, W. M., Trans. ASME, 77, 1265, (1955). L-1 Lestikow, S., et al., "Study on High Temperature Steam Oxidation of Zircaloy-4 Cladding Tubes," Nuclear Safety Project Second Semiannual Report, 1975, KfK-2262, Karlsruhe, 233, (1976). L-2 Lustman, B., and Kerze, F., The Metallurgy of Zirconium, National Nuclear Energy Series, McGraw-Hill Book Company, (1956). M-1 Miller, G. L., M-2 Mallett, M. W., and Albrecht, W. M., "High Temperature Oxidation of Two Zirconium-Tin Alloys," J. Electrochem. Soc., 102, 407, (1955). N-l Norris, R. H., et al., Heat Transfer Data Book, General Electric Co., Schenectady, N.Y., (1978 update). P-1 Perry, J. H., Chemical Engineer's Handbook, McGraw-Hill Book Company, (1952). Q-1 Quill, L. L., The Chemistry and Metallury of Miscellaneous Metals, New York, (1950). R-1 Rohsenow, W. M., and Choi, H., Heat, Mass and Momentum Transfer, Prentice Hall, Inc., (1961). S-1 Stalker, K. T., personal communication, Sandia National Laboratories, letter dated July 6, 1982. S-2 Sparrow, E. M., and Loeffler, A. L., AIChe Journal, 5:325, (1959). S-3 Smith, G. D., Numerical Solution of Partial Differential Equations: Finite Difference Methods, Clarendon Press, (1978). S-4 Schlichting, H., Boundary Layer Theory, 4th ed., McGrawHill Bood Company, (1960). S-5 Sparrow, E. M., and Loeffler, A. L., "Heat Transfer to Longitudinal Laminar Flow Between Cylinders," J. Heat Transfer, Trans. ASME, 83:415, (1961). Zirconium, Academic Press, Inc., (1957). 219 S-6 Siegel, R., and Norris, R. H., "Tests of Free Convection in a Partially Enclosed Space Between Two Heated Vertical Plates," Trans. ASME, 79, 663, (1957). S-7 Skinner, G. B., Edwards, J. W., and Johnstone, H. L., "The Vapor Pressure of Inorganic Substances, V. Zirconium between 1949 and 2045K," J. Amer. Chem. Soc., 73, 174, (1951). W-l Whitaker, S., "Forced Convection Heat-Transfer Correlations for Flow in Pipes, Past Flat Plates, Single Cylinders, Single Spheres and Flow in Packed Beds and Tube Bundles," AIChe J., 18, (1962). W-2 White, J. H., reported in AEC Fuels and Materials Development Program, Progess Report No. 67, GEMP-67, General Electric Co., 151, (1967). W-3 Willoughby, R. A., ed., Stiff Differential Systems, Plenum Press, Inc., (1973).