Full report

Original page 1

NASA/TM—2005–213609 Safe Affordable Fission Engine- (SAFE-) 100a Heat Exchanger Thermal and Structural Analysis B.E. Steeve Marshall Space Flight Center, Marshall Space Flight Center, Alabama March 2005

Original page 2

The NASA STI Program Office…in Profile Since its founding, NASA has been dedicated to the advancement of aeronautics and space science. The NASA Scientific and Technical Information (STI) Program Office plays a key part in helping NASA maintain this important role. The NASA STI Program Office is operated by Langley Research Center, the lead center for NASA’s scientific and technical information. The NASA STI Program Office provides access to the NASA STI Database, the largest collection of aeronautical and space science STI in the world. The Program Office is also NASA’s institutional mechanism for disseminating the results of its research and development activities. These results are published by NASA in the NASA STI Report Series, which includes the following report types: • TECHNICAL PUBLICATION. Reports of completed research or a major significant phase of research that present the results of NASA programs and include extensive data or theoretical analysis. Includes compilations of significant scientific and technical data and information deemed to be of continuing reference value. NASA’s counterpart of peerreviewed formal professional papers but has less stringent limitations on manuscript length and extent of graphic presentations. • TECHNICAL MEMORANDUM. Scientific and technical findings that are preliminary or of specialized interest, e.g., quick release reports, working papers, and bibliographies that contain minimal annotation. Does not contain extensive analysis. • CONTRACTOR REPORT. Scientific and technical findings by NASA-sponsored contractors and grantees. • CONFERENCE PUBLICATION. Collected papers from scientific and technical conferences, symposia, seminars, or other meetings sponsored or cosponsored by NASA. • SPECIAL PUBLICATION. Scientific, technical, or historical information from NASA programs, projects, and mission, often concerned with subjects having substantial public interest. • TECHNICAL TRANSLATION. English-language translations of foreign scientific and technical material pertinent to NASA’s mission. Specialized services that complement the STI Program Office’s diverse offerings include creating custom thesauri, building customized databases, organizing and publishing research results…even providing videos. For more information about the NASA STI Program Office, see the following: • Access the NASA STI Program Home Page at http://www.sti.nasa.gov • E-mail your question via the Internet to help@sti.nasa.gov • Fax your question to the NASA Access Help Desk at 301–621–0134 • Telephone the NASA Access Help Desk at 301–621–0390 • Write to: NASA Access Help Desk NASA Center for AeroSpace Information 7121 Standard Drive Hanover, MD 21076–1320 301–621–0390

Original page 3

NASA/TM—2005–213609 Safe Affordable Fission Engine- (SAFE-) 100a Heat Exchanger Thermal and Structural Analysis B.E. Steeve Marshall Space Flight Center, Marshall Space Flight Center, Alabama National Aeronautics and Space Administration Marshall Space Flight Center • MSFC, Alabama 35812 March 2005 i

Original page 4

Available from: NASA Center for AeroSpace Information 7121 Standard Drive Hanover, MD 21076–1320 301–621–0390 ii National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 703–487–4650

Original page 5

TABLE OF CONTENTS 1. INTRODUCTION ....................................................................................................................... 1 2. STRUCTURAL DESIGN SUMMARY ....................................................................................... 2 3. DESIGN AND ANALYSIS HISTORY........................................................................................ 3 4. METHODOLOGY ....................................................................................................................... 5 5. TEST CONDITIONS ................................................................................................................... 6 6. DESIGN CRITERIA ................................................................................................................... 7 6.1 Load Controlled Stress (NH–3222) ...................................................................................... 7 6.2 Inelastic Strains (T–1310) .................................................................................................... 8 6.3 Creep-Fatigue Damage (T–1411) ......................................................................................... 8 6.4 Weldments ............................................................................................................................ 9 7. FINITE ELEMENT MODEL ...................................................................................................... 10 8. DIMENSIONS AND GEOMETRY ............................................................................................. 11 9. MATERIAL PROPERTIES ......................................................................................................... 12 10. TEMPERATURE ......................................................................................................................... 13 11. LOADS ........................................................................................................................................ 15 12. POSTPROCESSING ................................................................................................................... 16 12.1 Sleeve-Weld Interface Axial Force and Moment ................................................................ 16 12.2 Section Stress at Critical Locations .................................................................................... 16 12.3 Section Strain at Critical Locations .................................................................................... 16 12.4 Creep-Fatigue Damage at a Point ....................................................................................... 17 13. RESULTS .................................................................................................................................... 19 13.1 Thermal Analysis Results ................................................................................................... 19 13.2 Heat Pipe Core Fixity ......................................................................................................... 20 13.3 Pressure Only ...................................................................................................................... 20 iii

Original page 6

TABLE OF CONTENTS (Continued) 13.4 Thermal Stress Results ........................................................................................................ 22 13.5 Welds ................................................................................................................................... 25 13.6 Braze Strength Effect .......................................................................................................... 26 14. CONCLUSIONS ......................................................................................................................... 27 14.1 Braze Structural Requirement ............................................................................................. 27 APPENDIX A—MATERIAL PROPERTIES ..................................................................................... 29 APPENDIX B—PROOF TEST RESULTS ........................................................................................ 31 REFERENCES .................................................................................................................................... 37 iv

Original page 7

LIST OF FIGURES 1. Solid model of SAFE-100a HX ......................................................................................... 1 2. Cross section of an HX ....................................................................................................... 2 3. Ritz-Toast HX design finite element model ....................................................................... 3 4. Ritz-Toast high bending von Mises stress (psi) ................................................................. 4 5. Octo-Block HX design concept finite element model ........................................................ 4 6. Creep-fatigue damage envelope for 316 SS ....................................................................... 9 7. Finite element model .......................................................................................................... 10 8. Method of estimating linear distribution through wall thickness ....................................... 17 9. Temperature profile (°F) for brazed, 52-Btu/s, nonfailed heat pipe condition ................... 19 10. Temperature profile (°F) for brazed, 52-Btu/s, failed heat pipe condition ......................... 19 11. Stress intensity (psi) due to pressure only—He, 52 Btu/s................................................... 21 12. Maximum strain levels for each test condition .................................................................. 22 13. Center sleeve axial stress versus cover plate thickness—brazed, 62-Btu/s, failed heat pipe ................................................................................................... 23 14. Strain through center sleeve—brazed, 62-Btu/s, failed heat pipe ...................................... 24 15. Maximum creep-fatigue damage ratio ............................................................................... 24 16. Equivalent strain, brazed, 62-Btu/s, failed heat pipe .......................................................... 25 17. SAFE-100a HX proof test setup ......................................................................................... 32 18. Braze cup side .................................................................................................................... 32 19. Side opposite braze cups .................................................................................................... 33 20. Barrel section ...................................................................................................................... 33 v

Original page 8

LIST OF FIGURES (Continued) 21. Strains and pressure versus time ........................................................................................ 34 22. Finite element model radial strain ...................................................................................... 34 23. Comparison of test and predicted strain ............................................................................. 35 vi

Original page 9

LIST OF TABLES 1. Test conditions .................................................................................................................... 6 2. Braze properties .................................................................................................................. 12 3. Manifold and plenum thermal boundary conditions .......................................................... 13 4. Heat pipe vapor temperatures ............................................................................................. 14 5. Expected test cycle and life usage ...................................................................................... 18 6. Heat pipe axial forces for a fixed core—brazed, 52 Btu/s ................................................. 20 7. Load-controlled stress results ............................................................................................. 21 8. Maximum first principal strains and margins ..................................................................... 23 9. Weld factors of safety ......................................................................................................... 25 10. Effect of braze strength ...................................................................................................... 26 11. 316L SS temperature-dependent material properties ......................................................... 29 12. 316L SS bilinear stress-strain curve data ........................................................................... 29 13. He-equivalent thermal conductivity ................................................................................... 30 14. Radiation gap equivalent thermal conductivity .................................................................. 30 vii

Original page 10

LIST OF ACRONYMS AND SYMBOLS Ag silver ASME American Society of Mechanical Engineers BKIN bilinear kenamatic hardening Cu copper He helium HX heat exchanger LANL Los Alamos National Laboratory MPM Material Properties Manual MSFC Marshall Space Flight Center PVBC Pressure Vessel and Boiler Code SAFE Safe Affordable Fission Engine SS stainless steel TIG tungsten inert gas TM Technical Memorandum viii

Original page 11

NOMENCLATURE C constant (0.24) D damage Ftudesign ultimate tensile strength under design conditions Ftutest ultimate tensile strength under test conditions i time point J sum of principle stresses j cycle type K section factor k time interval Kt (K+1)/2 n number of applied cycles for cycle-type j Nd number of allowable cycles for cycle-type j o time point P total number of strain ranges Pb primary bending stress PL local membrane stress Pm general primary membrane stress q total number of creep intervals R ratio of the weld metal creep rupture strength to the base metal creep-rupture strength S square root of the sum of the square of the principle stresses Sbend bending stress ix

Original page 12

NOMENCLATURE (Continued) Se equivalent stress Seqv von Mises stress Sm time independent allowable stress Sm+b combined stress Smemb membrane stress Smt the lower of Sm and St Sr stress-to-rupture strength St time-dependent allowable stress t thickness Td allowable creep time duration UΖ axial direction UΘ symmetry boundary Δt duration of time interval k εeqv equivalent tensile strain εxi strain in the x direction at time i εyo strain in the y direction at time o εyi strain in the y direction at time i εyo strain in the y direction at time o εzi strain in the z direction at time i γxyi shear strain in the xy direction at time i γyzi shear strain in the yz direction at time i γzxi shear strain in the zx direction at time i ν Poisson’s ratio x

Original page 13

TECHNICAL MEMORANDUM SAFE AFFORDABLE FISSION ENGINE- (SAFE-) 100A HEAT EXCHANGER THERMAL AND STRUCTURAL ANALYSIS 1. INTRODUCTION The Safe Affordable Fission Engine- (SAFE-) 100a experiment is a thermal simulation of an in-space nuclear reactor core. The heat created by the nuclear fission process is simulated by electric heaters placed in the core fuel pins where the uranium would normally be. The heat from the core is transported out of the core via sodium-filled heat pipes that extend out of the core on one side. The heat pipes pass through a heat exchanger (HX) that extracts heat from the pipes and transfers it into a helium- (He-) argon gas mixture. This heated gas would then be used to drive an electric generator. The HX is a welded stainless steel (SS) structure designed by Los Alamos National Laboratory (LANL). The structure is an annular flow design, where the bulk of the heat transfer occurs in the gas flowing though an annular passage around each heat pipe. The HX is designed to fit a core with nineteen heat pipes and sized to operate at the conditions needed for a Brayton cycle power conversion system. For the tests, the HX gas is passed through the test facility gas conditioning system. A full SAFE-100 reactor core consists of 61 heat pipe modules sized to generate 95 Btu/s of thermal power. Due to cost restraints, this experiment setup only uses the inner 19 modules and is called SAFE-100a. This Technical Memorandum (TM) describes the structural analysis and results of the SAFE-100a HX for several test conditions. Sleeves Inlet Outlet Upper Cover Central Jacket Lower Plenum Figure 1. Solid model of SAFE-100a HX. 1

Original page 14

1. STRUCTURAL DESIGN SUMMARY The HX (fig. 2) is fabricated from 316L SS and consists of 19 sleeves, an upper and lower cover, an upper and lower inner plenum, an upper and lower outer plenum, and a central jacket. All of the pieces are welded together by tungsten inert gas (TIG) or electron beam welding. During the manufacturing process, the HX will undergo several intermediate and a final annealing processes. Sleeve-to-Cover-Plate Weld Upper Plenum Sleeve Heat Pipe-to-Sleeve Braze Length Lower Plenum Inlet Central Jacket Outlet Lower Cover With Integral Baraze Cups Figure 2. Cross section of an HX. 2

Original page 15

1. DESIGN AND ANALYSIS HISTORY The original HX design concepts were numerous and varied. The first to be seriously pursued and analyzed was termed the Ritz-Toast design (figs. 3 and 4). This design had individual cross-flow HXs around separate rows of heat pipes. To fit an HX around each row, right next to each other, the design had to be thin. This created a flat pressure vessel that introduced high bending stresses in the flat wall around internal supports and was determined to be unacceptable and without any good design solutions. Y Z Figure 3. Ritz-Toast HX design finite element model. The next design concept investigated was termed the Octo-Block design (fig. 5). This design also had individual HXs around each row of heat pipes but was an annular flow design with an inlet and outlet manifold. Since the pressure walls were cylinders, the pressure stresses were not a problem. However, this design was dropped due to heat transfer, flow, and plumbing issues. The next design that made it to the analysis stage looked similar to the current design. It had the same flow design, but instead of a solid jacket section, it had individual jackets surrounding each sleeve for each heat pipe. For manufacturing ease and improved thermal conduction between the annular flow passages, the jackets were replaced in the current design as a solid piece with through holes to form the annular passages and blind holes to lighten the structure. 3

Original page 16

Y X Z MX 211.737 8,044 15,877 23,710 31,542 39,375 47,207 55,040 62,873 MN 70,705 Figure 4. Ritz-Toast high bending von Mises stress (psi). Y Z Figure 5. Octo-Block HX design concept finite element model. 4

Original page 17

1. METHODOLOGY The HX analysis was performed using a finite element model of the HX along with the heat pipes and a small representation of the core. This model was used to solve for the temperature profile, which was then used in a structural solution. Only the steady state thermal condition was considered. The transient thermal heat up and cool down are assumed to be slow and less severe than the steady state condition. Hand calculations were used to evaluate the welds and pipe stub. 5

Original page 18

1. TEST CONDITIONS There are 12 test conditions proposed for the SAFE-100a experiment with the HX. All 12 conditions will be analyzed. Table 1 lists the 12 conditions. Since the SAFE-100a experiment only has 19 pins (or modules) versus the full suite of 61 pins, it operates at 19/61 percent of full power. Table 1 lists the power level for both configurations. This TM refers to both the 19- and 61-pin power levels interchangeably. Table 1. Test conditions. Power (Btu/s) Temperature (°F) Test Gap Failed Flow Pressure In No. Fill Heat Pipe 19 Pin 61 Pin (Ibm/s) In Out (psi) H1 He no 8.1 26 0.324 909 1,071 200 H2 He no 16.2 52 0.324 621 947 200 H3 He no 19.3 62 0.324 621 947 200 H4 He yes 8.1 26 0.324 909 1,071 200 H5 He yes 16.2 52 0.324 621 947 200 H6 He yes 19.3 62 0.324 621 947 200 B1 braze no 8.1 26 0.324 909 1,071 200 B2 braze no 16.2 52 0.324 745 1,071 200 B3 braze no 19.3 62 0.324 745 1,071 200 B4 braze yes 8.1 26 0.324 909 1,071 200 B5 braze yes 16.2 52 0.324 745 1,071 200 B6 braze yes 19.3 62 0.324 745 1,071 200 The first six tests will be performed without any physical contact between the HX and the heat pipes. The test chamber will be filled with He to provide a medium to help transfer heat to the HX. The final six tests will be performed after brazing the HX to the heat pipes. The braze will provide the thermal path, and the test chamber will be pumped down to a vacuum. Both the He and brazed cases will be performed with and without a failed heat pipe. For the cases without a failed heat pipe, all 19 heat pipes will be filled and work to transport the heat from the core to the HX. For the failed heat pipe cases, the center heat pipe will be replaced with an empty pipe. This pipe will, therefore, only conduct heat through its metal wall. The cases will be performed at three different power levels. In all cases the pressure will remain the same. 6

Original page 19

1. DESIGN CRITERIA The HX operates at temperatures that range from 620 to 1,250 °F. At these temperatures, creepinduced failure of the structure is a possibility. There are two basic failure modes identified for the HX. They are an overpressure-induced failure of the pressure wall and creep-fatigue damage-induced cracking. To address these failure modes from an analysis and design standpoint, the criteria found in “ASME Pressure Vessel and Boiler Code”, section III: subsection NH was partially adopted for use.1Only portions of the code were used in the design and analysis of the HX. A summary of the criteria used is listed here. The code should be referred to for a complete explanation of the criteria. For the purpose of this analysis, the pressure-induced stress is considered to be the primary stress, and the thermal induced stresses are considered to be secondary stresses. The thermal stress is only considered in the strain and creep-fatigue damage criteria. 6.1 Load Controlled Stress (NH–3222) Only the level A and B service limits are considered in this analysis. The design and level C and D limits are not used. The level A and B service limits are Pm ≤ Smt , (1) PL + Pb ≤ K ⋅ Sm , (2) and Pb PL + Kt where Pm = general primary membrane stress PL = local primary membrane stress Pb = primary bending stress Sm = time-independent allowable stress St = time dependent allowable stress Smt = lower of Sm and St K = section factor (1.5 for a rectangular section) Kt = (K+1)/2. ≤ St , (3) 7

Original page 20

6.2 Inelastic Strains (T–1310) The inelastic strains should not exceed the following for the maximum principal strain: • One percent averaged through the thickness. • Two percent due to a linear distribution of strain through the thickness. • Five percent at any point. 6.3 Creep-Fatigue Damage (T–1411) Fatigue is evaluated using a local equivalent strain range and compared to the fatigue allowable given in appendix T of subsection NH:1 2  2 2 ∆εeqv i, = (∆εxi − ∆εyi) ( yi zi) 2 1( + ν)  2 3 + ∆ε − ∆ε 1 2 2 2  22 zi xi (∆γ xyi + ∆γ yzi + ∆γ zxi ) . (4) ++(∆ε − ∆ε ) + 2   The change in the strain components between time i and time o, where time o is at an extreme minimum or maximum, is given by ∆εyi = εyi yo and so forth. The term ν is 0.5. − ε , (5) The combined creep and fatigue damage allowable is given by the relation P q  n   ∆t  ∑   + ∑    Nd  ≤ D , (6)  Td  j=1 j k=1 k where first term = fatigue ratio second term = creep ratio n = number of applied cycles for cycle type j Nd = number of allowable cycles for cycle type j Δt = duration of the time interval k Td = allowable time duration for time interval k. The total damage, D, should not exceed the creep-fatigue damage specified in reference 1 for 316 SS and shown in figure 6. 8

Original page 21

1.2 1 0.8 0.6 Creep Ratio 0.4 0.2 0 0 0.2 0.4 0.6 0.8 1 1.2 Fatigue Ratio Figure 6. Creep-fatigue damage envelope for 316 SS.1 6.4 Weldments For regions within the heat-affected zone of a weld, the strength and life allowables are modified per the following: • Load controlled stress: – The allowable limits for St and Smt are the lower of St or Smt and 0.8×Sr×R. – The stress-to-rupture strength value for a given time and temperature is Sr. The ratio of the weld metal creep rupture strength to the base metal creep rupture strength is R, as found in table I–14.10B of the ASME Pressure Vessel and Boiler Code (PVBC).1 • Inelastic strains: The strain limits for a welded region are one half of the allowable for the base material. • Creep-fatigue damage: – The number of allowable cycles for low cycle fatigue is one half of the number allowed for the base metal. – The allowable time duration is determined from a stress-to-rupture curve by multiplying the base material stress-to-rupture values by the ratio, R. 9

Original page 22

1. FINITE ELEMENT MODEL The arrangement of the heat pipe modules in the core repeats every 60°, and each of these sections is symmetric about its axial midplane. The thermal profile is assumed to not vary in the circumferential direction. Therefore, a 30° or 1/12th section with symmetry boundary conditions is sufficient to model the behavior of the entire HX. The finite element model (fig. 7) of the HX, then, is a 1/12th symmetry representation of the HX, heat pipes, and partial core. The model is built and solved using ANSYS 7.0 and is meshed with tetrahedron (SOLID186) and quad brick (SOLID185) elements. Transitional pyramid-shaped elements are used between the interface of the tetrahedron and quad elements. The model geometry was built from scratch using the geometry and dimensions from a solid model created and provided by LANL. HX Heat Pipes Core Figure 7. Finite element model. Symmetry boundary conditions, UΘ=0, are placed on the axial symmetry planes. The axial direction is restrained, UΖ=0, on the bottom of the core and the bottom of the outlet manifold at the outer edges. For the He cases, the heat pipes are coupled to the core in all directions. For the braze conditions, the heat pipes are coupled to the core only in the radial and circumferential directions with the heat pipes free to slide through the core in the axial direction. The thermal and structural loads were applied to the appropriate surfaces of the model as described in section 10. The material properties used in the model are described in section 11. 10

Original page 23

1. DIMENSIONS AND GEOMETRY The dimensions and geometry for the finite element model were taken from a solid model created by LANL. To reduce the element count, small features, such as chamfers and very skinny areas, were not included in the finite element model. Except for the inlet and outlet pipes, all significant structural features are included. Nominal dimensions are used in the finite element model. The orientation of the HX with respect to the core is with the inlet away from the core. This orientation places the hot outlet side of the HX closer to the core to help minimize the radial thermal growth mismatch between the core and the HX. The HX is located a distance of 8 cm away from the core. The heat pipes are all centered within the HX sleeves. For the brazed test conditions, a solid braze from the outer edge of the outlet cover plate to the plane of the solid jacket core on the inlet side is assumed. 11

Original page 24

1. MATERIAL PROPERTIES The HX is constructed from 316L SS. The various pieces are then welded together by either the electron beam or TIG process. The HX will undergo several annealing processes as well as a braze cycle. Therefore, the 316L SS properties are for the annealed condition. The plastic behavior of the HX was treated as a bilinear stress-strain curve in the model. The ANSYS bilinear kinematic hardening (BKIN) material model was used to represent this inelastic behavior. All material properties were taken from the Rocketdyne “Material Properties Manual” (MPM), sections 2211 and 2212, and MIL–HDBK–5F.2,3 Appendix A of this TM lists the properties used in this analysis. For the brazed cases, the HX is brazed to the heat pipes using the silver- (Ag-) based braze alloy Nicusil-8 (BAg-13a). Its composition is Ag 56:copper (Cu) 42:nickel (Ni) 2, and it has a solidus of 1,417 °F and a liquidus of 1,638 °F. There are no readily available properties for this braze at elevated temperatures. An attempt was made to pull together some properties for use in the finite element model based on very limited braze properties and a review of Ag and Cu properties at elevated temperatures. Table 2 lists the properties that were used. These properties are guesses and not verified. One of the guidelines in coming up with these properties was to make the braze soft and weak compared to SS. Table 2. Braze properties. Nicusil-8 Braze Properties Property Units 70 °F 800 °F 1,000 °F 1,200 °F Thermal conductivity Btu/hr-ft-°F 136 112 110 108 W/m-K Coefficient of thermal expansion in/in/°F m/m/K Elastic modulus msi GPa Poisson’s ratio – Yield strength ksi MPa Tangent modulus ksi MPa 12 235 194 190 187 8.4 10 11 12 15.1 18 19.8 21.6 14 12 11 10 97 83 76 69 0.35 0.35 0.35 0.35 25 12 10 8 172 83 69 55 100 80 80 80 689 552 552 552

Original page 25

1. TEMPERATURE Being an HX, the temperature varies throughout the component. These temperature gradients create thermal loads and strains that dominate the structural performance. The temperature profile used in the analysis is generated by applying thermal boundary conditions to the finite element model and then solving for the steady state thermal solution. The thermal conditions for the annular flow between the HX sleeve and jacket were calculated and provided by LANL. The heat transfer coefficient for the sleeve and jacket and the temperature of the heat pipe and coolant were provided as a function of position down the length of each row of annular flow paths. Failed heat pipe conditions were provided for the center failed heat pipe annulus and the adjacent coolant paths. The thermal conditions for the outer two rows are assumed to remain the same for both the nonfailed and failed heat pipe conditions. The thermal conditions for the inner and outer manifold regions of both the inlet and outlet were calculated by the Structures, Mechanics, and Thermal Department (ED25) of Marshall Space Flight Center (MSFC). Table 3 lists the results of these calculations.4 Table 3. Manifold and plenum thermal boundary conditions. Heat Transfer Coefficient (Btu/ft2-hr-°F) Flow Fluid Temperature (°F) Inlet Outlet Rate Inlet Outlet (Ibm/s) Manifold Plenum Manifold Plenum 621 947 0.324 1,102 1,590 1,136 1,374 621 947 0.324 1,261 1,817 1,295 1,573 745 1,071 0.324 1,130 1,670 1,158 1,431 745 1,071 0.324 1,283 1,891 1,312 1,618 909 1,071 0.324 1,130 1,624 1,147 1,528 The lightening holes in the solid jacket are assumed to be adiabatic and have no thermal boundary conditions on any of these surfaces. The exterior of the HX was also assumed to be adiabatic. For the He test conditions, an equivalent thermal conductivity was used for the elements between the heat pipe and the HX sleeves in the thermal runs. For the brazed test conditions, an equivalent thermal conductivity was used for the elements in the vacuum gap between the heat pipe and the HX sleeves in the unbrazed region. These equivalent conductivities are listed in appendix A. The axial temperature profiles of the heat pipes were specified by LANL for both the annular flow region and the length between the HX and core. These temperatures were applied to the inner diameter of the heat pipes. The core was allowed to conduct to a steady state solution based on the temperature specified for the heat pipes. The temperatures for the heat pipes between the core and HX (heat pipe vapor temperature) are given in table 4. 13

Original page 26

Table 4. Heat pipe vapor temperatures. Test Condition Test Gap Failed Power No. Fill Heat Pipe (Btu/s) H1 He no 8.1 H2 He no 16.2 H3 He no 19.3 H4 He yes 8.1 H5 He yes 16.2 H6 He yes 19.3 B1 braze no 8.1 B2 braze no 16.2 B3 braze no 19.3 B4 braze yes 8.1 B5 braze yes 16.2 B6 braze yes 19.3 14 Heat Pipe Vapor Temperature (°F) Ring 2 Ring 2 Center Ring 1 Mid-side Corner 1,156 1,147 1,139 1,131 1,124 1,107 1,090 1,074 1,157 1,137 1,118 1,099 failed 1,170 1,139 1,131 failed 1,153 1,090 1,074 failed 1,176 1,118 1,099 1,097 1,094 1,090 1,105 1,125 1,117 1,110 1,104 1,132 1,124 1,116 1,109 failed 1,135 1,090 1,105 failed 1,201 1,110 1,104 failed 1,210 1,116 1,109

Original page 27

1. LOADS The pressure for each load case is 200 psi (1.38 MPa) and is applied to all of the interior surfaces. There are no other loads applied to the model. 15

Original page 28

1. POSTPROCESSING Several postprocessing techniques are employed to extract relevant data from each of the finite element solutions. Sections 12.1–12.4 describe each of these methods. 12.1 Sleeve-Weld Interface Axial Force and Moment The axial force and moment about the global circumferential axis is determined using an ANSYS macro. This macro selects the nodes on the sleeve-to-upper-cover interface and then sums the force at each node in the axial direction to obtain the total axial force. The contribution of each nodal force to the moment about the global circumferential axis is also summed based on the force and the distance from the axis. This macro also determines the total axial force at the upper cover-to-outer-manifold interface. The results are then sent to an output file. 12.2 Section Stress at Critical Locations The ASME PVBC design criteria are based on membrane and bending stress intensity across a section. To get these stresses, an ANSYS macro is used to compute the membrane and bending stress through several critical locations. These critical sections are the peak stress intensity location on the outside of each sleeve, the peak stress intensity location on the inside and outside of both the inlet and outlet covers, the peak stress intensity location on the outside of the side wall of the outer manifold, and the peak stress intensity location on the inside of the through holes of the jacket. A path is created at each critical location from the peak stress intensity surface node to a node closest to a point on the opposite surface of the wall normal to the wall surface. The section stress command within ANSYS is used to compute the membrane and bending component stress for each of the paths. The membrane and bending stress intensity is then computed from the section component membrane stresses. The membrane and bending stress components are combined according to /1.25 per the ASME PVBC criteria found in NH–3223.1These combined compo- Sm+b=Smemb+Sbend nent stresses are then used to compute the membrane plus bending stress intensity. 12.3 Section Strain at Critical Locations The allowable strains for any section are listed in the ASME PVBC, appendix T–1310.1They are the average, linearized surface, and maximum point strain based on the maximum positive principal strain. To obtain these values, an ANSYS macro is used. This macro creates a path at critical locations as described in section 12.2. The critical locations evaluated are the peak first principal strain location of the outside of each sleeve and the peak first principal strain location of the inside and outside of the inlet and outlet covers. 16

Original page 29

To calculate the average strain, the component strains are determined along each path. The component strains at each path point are then averaged and used to determine the average first principal strain. The strain at the surface, based on a linearized strain profile, is estimated using the first principal strain profile along the path. This is done according to figure 8. 1.25 1.2 1.15 1.1 Data 1.05 1 Actual 0.95 Slope Equivalent Linear 0.9 0 0.02 0.04 0.06 Distance Slope, m, of Linearized Line Surface Value Due to Equivalent Linear Distribution Line of Slope, m, Passes Through Average Value Average Value 0.08 0.1 Figure 8. Method of estimating linear distribution through wall thickness. 12.4 Creep-Fatigue Damage at a Point The strain and stress at critical locations are extracted by a macro for use with the creep-fatigue damage criteria. The macro finds the maximum strain (total equivalent strain), stress (von Mises stress), and temperature location of each sleeve, inlet cover, and outlet cover. The total equivalent strain at each of these points is extracted to compare with the low cycle fatigue allowable. This analysis uses the total equivalent strain at the steady state thermal condition as the total strain range to enter the low cycle fatigue curve. An equivalent stress at each point is calculated according to Se=Seqv×exp(C(J/S–1)) per the ASME PVBC in paragraph T–1411.1This stress is then used to find the allowable creep life from the stress-to-rupture curve. In order to calculate a creep-fatigue damage ratio, an expected test usage was used (table 5). Using the strain and temperature at each critical point, an allowable number of cycles was determined from the low cycle fatigue curve per figure T–1420–1B of ASME PVBC.1Points between values on the curve were linearly interpolated. A fatigue damage ratio was then determined for each test condition. 17

Original page 30

Table 5. Expected test cycle and life usage. Test Gap Failed No. Fill Heat Pipe H1 He no H2 He no H3 He no H4 He yes H5 He yes H6 He yes B1 braze no B2 braze no B3 braze no B4 braze yes B5 braze yes B6 braze yes Power No. Total Test (Btu/s) Cycles Time (hr) 8.1 14 40 16.2 3 12 19.3 3 12 8.1 5 16 16.2 3 12 19.3 3 12 8.1 14 40 16.2 3 12 19.3 3 12 8.1 5 16 16.2 3 12 19.3 3 12 The equivalent stress, Se, and temperature at each critical point were used to enter the stress-torupture table and determine an allowable creep life for each test condition, a life ratio was calculated, and the sum of all the fatigue ratios and life ratios for each component were computed.6These sums were then plotted against the allowable creep-fatigue damage curve. 18

Original page 31

1. RESULTS 13.1 Thermal Analysis Results The steady state thermal solution was solved for each of the twelve test conditions using the thermal boundary conditions given for each case as described above. Figures 9 and 10 show the temperature profile for a brazed, 52-Btu/s, nonfailed heat pipe and failed heat pipe condition. 748.017 MN 790.197 832.377 874.557 916.737 958.917 1,001 1,043 1,085 MX 1,128 Figure 9. Temperature profile (°F) for brazed, 52-Btu/s, nonfailed heat pipe condition. MN 748.0287 798.905 849.782 900.66 951.537 1,002 1,053 1,104 1,155 1,206 MX Figure 10. Temperature profile (°F) for brazed, 52-Btu/s, failed heat pipe condition. 19

Original page 32

13.2 Heat Pipe Core Fixity One of the boundary conditions that was found to have a significant effect on the HX results was the assumed fixity of the heat pipes in the core. The heat pipes and core were initially one solid piece with either all or a portion of the core fixed in the axial direction. This essentially forces the axial displacement of the heat pipes to be zero at the core. This strongly effects the displacements at the HX outlet cover plate by not allowing the outlet cover plate to move axially due to the differences in thermal growth for each heat pipe/sleeve. The only way this boundary condition can occur in the current design of the core is for each heat pipe module to be kept from slipping due to friction and the compressive forces that develop in the core due to the radial restraints of the core. It seems unlikely that the core would restrain a heat pipe module from slipping due to the differences in thermal growth. The magnitude of the forces that develop in the heat pipes with the heat pipes fixed axially were checked and are listed in table 6 for a brazed, 52-Btu/s condition. Assuming a friction factor of 0.5 between the SS heat pipe modules, a load of 1,200 lb would be required to prevent the module from slipping. Table 6. Heat pipe axial forces for a fixed core—brazed, 52 Btu/s. Heat Pipe Axial Load (lb) Center 537.6 Ring 1 –600.2 Ring 2 mid 275.2 Ring 2 corner 235.2 Releasing the heat pipes in the axial direction creates a more severe condition for the HX because that allows the outlet cover plate to displace according to the thermal growth of each heat pipe/sleeve. Since this is a more severe condition, the analysis of the HX includes the release of the heat pipes in the axial direction at the core. If testing or analysis of the core shows that the core prevents the modules from slipping, then this boundary condition can be revisited. 13.3 Pressure Only In order to remove the thermal induced stress and strain, the pressure-only cases were run with the coefficient of thermal expansion set to zero . Since pressure in this case is the only load, the results for each test condition are about the same. The only difference is due to the change in the material properties (Young’s modulus and Poison’s ratio) due to the temperature differences between the test conditions. The stress levels for the pressure-only cases were low enough to stay below the SS yield point, so the cases were run elastically. Figure 11 is a plot of the stress intensity for the He 52-Btu/s condition. The maximum stress locations occur in the cover plates at the largest unsupported span. The peak stress is indicated at a point in the jacket wall. This peak is at one node and appears to be an artificial artifact of the model. 20

Original page 33

Y X Z 6.479 1,002 1,997 MX 2,992 3,987 4,982 5,977 6,972 MN 7,967 8,962 Figure 11. Stress intensity (psi) due to pressure only—He, 52 Btu/s. The criteria for assessing the pressure-induced stresses are based on the membrane and bending stress through a section. Table 7 lists the membrane and bending stresses for the criteria through the wall thickness at the point of maximum stress intensity for the jacket and cover plates for each of the 52-Btu/s test conditions. Table 7. Load-controlled stress results. Temperature Pm P +P L b L b t t Test (°F) (psi) (psi) Jacket P +P /K Margin of Safety S Life (psi) Smt Sm (hr) H2 846 5,319 8,960 8,224 1.54 1.26 100,000 H5 861 5,322 8,967 8,230 1.54 1.26 100,000 B2 1,002 5,287 8,905 8,173 1.55 1.27 100,000 B5 1,014 5,290 8,913 8,180 1.5 1.26 100,000 Upper Cover H2 862 3,018 8,212 7,151 3.47 1.47 100,000 H5 863 3,020 8,223 7,160 3.47 1.46 100,000 B2 801 3,331 9,083 7,886 3.05 1.23 100,000 B5 801 3,332 9,091 7,892 3.05 1.23 100,000 Lower Cover H2 1,039 2,434 7,404 6,387 4.23 1.67 100,000 H5 1,044 2,444 7,429 6,409 4.16 1.65 100,000 B2 1,107 3,369 8,433 7,377 1.72 1.23 100,000 B5 1,107 3,369 8,434 7,378 1.72 1.23 100,000 21

Original page 34

The results listed in table 7 show that the HX has sufficient margin against pressure induced burst and stress-rupture failure. 13.3.1 Proof Test The HX was proof tested before installation in the SAFE-100a experiment. The proof factor is 1.25, and an environmental correction factor was applied to account for the difference between the material strength at room temperature and the elevated test temperature. The environmental correction factor is 70/43=1.63 (316L strength at 70 °F/strength at 1,200 °F). The required proof pressure is 200 psi×1.25×1.63=407 psi minimum with the maximum proof pressure as 430 psi (420±10 psi). The proof-pressure case was run at 430 psi. The minimum margin of safety occurs in the upper cover and is 0.82. Appendix B includes the results of the actual proof test. 13.4 Thermal Stress Results The temperature distribution within the HX develops internal thermal stresses and strains throughout the part. The failed heat pipe conditions are the most severe because the cold failed heat pipe is surrounded by hot active heat pipes. The stresses and strains for this condition are concentrated primarily around the sleeve-to-cover-plate joint. Figure 12 plots the maximum equivalent strain for the highest four strained parts for each test condition. 0.04 Center Sleeve 0.035 Ring 1 Sleeve 0.3 Inlet Cover Outlet Cover 0.025 0.2 0.015 Equivalent Strain 0.01 0.005 0 H1 H2 H3 H4 H5 H6 B1 B2 B3 B4 B5 B6 Test Number Figure 12. Maximum strain levels for each test condition. 13.4.1 Cover Plate Thickness The cover plate thickness was found to have a significant effect on the structural behavior of the HX, and this effect is most pronounced for the failed heat pipe conditions. This is because the difference in axial thermal growth between the cool failed heat pipe/sleeve and hot adjacent heat pipes/sleeves 22

Original page 35

must be accommodated by the bending of the cover plates over a relatively small distance. Figure 13 plots the net section axial stress in the center sleeve versus cover thickness. A thickness of 0.1 in was chosen to keep the axial stress below the yield stress at the maximum center sleeve temperature. 30,000 25,000 20,000 15,000 10,000 Axial Stress (psi) 5,000 0 0.05 0.1 Cover Plate Thickness (in) Yield Stress at 800 °F 0.15 0.2 0.25 Figure 13. Center sleeve axial stress versus cover plate thickness—brazed, 62-Btu/s, failed heat pipe. 13.4.2 Inelastic Strain The strain through the wall thickness of each component was evaluated using the automated process described in section 12.3. It is assumed that the peak strains occur in the heat-affected zone of the sleeve-to-cover-plate weld. Therefore, the weld allowables are used to compute the margins of safety. The minimum margins of safety against the ASME strain criteria are listed in table 8. The surface strain due to an equivalent strain distribution in the center sleeve exceeds the criteria by ≈10 percent in the brazed, 62-Btu/s test condition only. All other conditions have positive margins. Table 8. Maximum first principal strains and margins. Component Test εavg εlinear Center sleeve B6 0.0084 0.0211 Ring 1 sleeve H6 0.0085 0.0113 Upper cover H6 0.0048 0.0099 Lower cover B6 0.0009 0.0023 εmax MSavg MSlinear MSmax 0.0257 0.2 –0.1 0.9 0.013 0.2 0.8 2.8 0.0119 0 0 1.1 0.0027 4.6 3.3 8.3 Since the surface strain is estimated using the method shown in figure 8, the strain through the center sleeve that produced the negative margin was reevaluated using an actual linear curve fit. Figure 14 shows that a linear curve fit produces a surface strain of 1.96 percent, which corresponds to a margin of +0.02. 23

Original page 36

0.03 FEM Results 0.025 Estimate Linear FEM Results 0.02 y=0.3883x+0.0052 0.015 0.01 First Principle Strain 0.005 0 0 0.01 Path Distance (in) 0.02 0.03 0.04 Figure 14. Strain through center sleeve—brazed, 62-Btu/s, failed heat pipe. 13.4.3 Creep-Fatigue The creep-fatigue results were compiled using the automated routine described in section 12.4. The maximum calculated damage ratio for the four most severely loaded components is shown in figure 15. The center sleeve has the highest damage ratio at 1, primarily due to the magnitude of the strains that occur in the failed heat pipe conditions. These results assume that the peak strains all occur within the sleeve-to-cover-plate weld heat-affected zone, and therefore, the weld criteria is used to calculate the damage. If the nonweld criterion is used, then the damage fraction for the center sleeve drops to 0.5. Figure 16 shows the strain levels for the worst case and location of the maximum strain in the fillet of the center sleeve. 1.2 1 0.8 0.6 Creep Ratio 0.4 0.2 0 0 0.2 0.4 Fatigue Ratio Creep-Fatigue Curve Center Sleeve Ring 1 Sleeve Inlet Cover Outlet Cover 0.6 0.8 1 1.2 Figure 15. Maximum creep-fatigue damage ratio. 24

Original page 37

MX MX 0.413×10–5 MN 0.413×10–5 0.003939 0.227874 0.11809 0.015744 0.019678 0.023613 0.027548 0.031483 0.035418 0.003939 0.227874 0.11809 0.015744 0.019678 0.023613 0.027548 0.031483 0.035418 Figure 16. Equivalent strain, brazed, 62-Btu/s, failed heat pipe. 13.5 Welds The welds in the HX were analyzed by hand. The loads for each weld were either extracted from the finite element model or calculated. Full penetration welds are assumed. The weld strength allowable used is 80 percent of the parent 316L material strength. Table 9 lists the calculated factors of safety for each weld. Table 9. Weld factors of safety. Weld Sleeve to cover plate Factor of Safety 4.4 Cover plate to inner plenum 19.1 Jacket to inner plenum 29.2 Inner plenum to outer ring high Pipe stub to plenum 3.5 The pipe stub-to-plenum weld factor of safety in table 9 is for the pressure load only on the external structural skip weld. Although there should not be a significant moment on the pipe stub since the coolant piping includes flex lines, the structural, skip-weld possesses less moment carrying capability than the pipe itself. If the internal seal weld is included, then the welds have more moment carrying capability than the pipe. 25

Original page 38

13.6 Braze Strength Effect The effect of the heat pipe-to-braze strength on the behavior of the HX was evaluated. Two additional brazed, 62-Btu/s, failed heat pipe cases were run. In one, the braze material properties were set to be 316L SS, and the other case, the braze was completely removed. The thermal profile for each case was the same. The results were then compared with the standard case using the Nicusil-8 braze properties listed in table 2. Table 10 lists the results and the differences for some of the peak stress/strain locations. Table 10. Effect of braze strength. Nicusil–8 SS 316L Missing Braze, 62-Btu/s, Failed Heat Pipe Results Analysis Result Value Value % D Value % D Center sleeve axial force (lb) 999 1,001 0.2 965 –3.4 Center sleeve maximum eqvivalent strain 0.0337 0.0339 0.6 0.0276 –18.1 Center sleeve maximum equivalent stress (psi) 29,340 29,419 0.3 27037 –7.8 Table 10 shows that there is little difference between the SS braze and the Nicusil-8 braze. The missing braze case is actually a less severe case. Since the heat pipes are hotter and expand more, removing the braze does not introduce as much thermal stress and strain into the HX. 26

Original page 39

1. CONCLUSIONS The HX was evaluated for its expected pressure and thermal loads. Sufficient margin was found against the pressure load, even at the elevated temperatures. The thermal loads were found to be more severe, especially for the failed heat pipe conditions. Regions of plastic deformation occur due to the failed heat pipe loads around the sleeve-to-cover-plate welds. Running multiple tests with a failed heat pipe condition causes low cycle fatigue to be the primary failure mode of concern. The calculated damage ratio is at the maximum allowable of 1 for the proposed test series. The sleeve-to-cover-plate welds are, therefore, the most critical location in the HX design. It is important to develop the welding/inspection process of these welds to ensure high quality, void-free welds. 14.1 Braze Structural Requirement Currently the maximum axial load that develops in the sleeves is 999 lb, which occurs in the failed heat pipe. The minimum sleeve area is 0.0748 in2. This results in an axial stress of 13,356 psi. The yield strength of 316L at 800 °F is 15 ksi. Based on these numbers and the results shown above, the braze is not essential to maintaining the structural integrity of the HX. 27

Original page 40

APPENDIX A—MATERIAL PROPERTIES Table 11. 316L SS temperature-dependent material properties. Thermal Specific Thermal Ultimate Young’s Yield Temperature Conductivity Density Heat Expansion Modulus Poisson’s Strength Strength (°F) (Btu/hr-ft-°F) (Ibm/ft3) (Btu/Ibm-°F) (10 /R) (psi×10 ) Ratio (ksi) (ksi) –400 5.43 504.4 0.063 –300 5.93 502.8 0.075 –200 6.43 501.3 0.085 –100 6.91 499.7 0.095 0 7.38 498.2 0.103 100 7.84 496.6 0.110 200 8.29 495.1 0.116 300 8.74 493.5 0.121 400 9.18 492 0.125 500 9.62 490.4 0.128 600 10.05 488.9 0.131 700 10.47 487.3 0.133 800 10.9 485.8 0.134 900 11.32 484.2 0.136 1,000 11.74 482.7 0.137 1,100 12.15 481.1 0.139 1,200 12.57 479.6 0.141 1,300 12.98 478.1 0.143 1,400 13.39 476.5 0.146 1,500 13.81 475 0.149 1,600 14.22 473.4 0.153 –6 6 5 29.2 0.285 50 184 5.83 29.1 0.287 41 150 6.55 28.9 0.29 36 120 7.19 28.6 0.294 32 95 7.75 28.2 0.298 27 77 8.23 27.8 0.302 24 69 8.64 27.2 0.307 22 66 8.98 26.7 0.311 22 63 9.27 26 0.315 21 61 9.51 25.4 0.319 19 59 9.71 24.7 0.322 17 58 9.87 24 0.326 16 58 10.01 23.4 0.329 15 57 10.12 22.7 0.331 15 55 10.22 22 0.334 15 53 10.31 21.4 0.336 14 49 10.4 20.7 0.338 13 43 10.49 20.1 0.34 11 34 10.6 19.5 0.343 9 26 10.72 18.9 0.346 6 20 10.88 18.4 0.349 4 15 Table 12. 316L SS bilinear stress-strain curve data. Temperature (°F) 70 200 600 800 1,200 1,400 Elastic modulus (ksi) 28 27 25 23 21 20 Yield strength (ksi) 31 26 19 18 16 15 Tangent modulus (ksi) 400 350 350 350 350 350 29

Original page 41

Table 13. He-equivalent thermal conductivity. Temperature (°F) 40 90 140 190 240 290 340 440 540 640 740 840 940 1,040 1,140 1,240 1,340 1,440 1,540 1,640 1,740 1,840 1,940 2,040 2,140 2,240 2,340 2,440 2,540 2,640 2,740 2,840 2,940 3,040 Thermal Conductivity (Btu/hr-ft-°F) 0.085 0.091 0.096 0.102 0.107 0.112 0.117 0.126 0.136 0.145 0.153 0.162 0.17 0.178 0.186 0.194 0.202 0.209 0.217 0.224 0.231 0.238 0.245 0.252 0.258 0.265 0.272 0.278 0.285 0.291 0.297 0.303 0.31 0.316 Table 14. Radiation gap equivalent thermal conductivity. Temperature (°F) 1,163 1,253 1,343 1,433 1,523 30 Thermal Conductivity (Btu/hr-ft-°F) 0.00303 0.00356 0.00415 0.00481 0.00553

Original page 42

APPENDIX B—PROOF TEST RESULTS The SAFE-100a HX was hydrostatically proof tested to accept the hardware for use in the SAFE- 100a pressurized coolant loop system. The target proof pressure was to be held for a minimum of 10 min at a level of 420±10 psi, which was determined as follows: • Design pressure = 200 psi. • Temperature correction factor = Ftutest/Ftudesign=70/43=1.63. – Ultimate strength (316L SS) at design temperature (1,200 °F), Ftudesign= 43 ksi. – Ultimate strength (316L SS) at proof test temperature (70 °F), Ftutest=70 ksi. • Proof test factor =1.25. • Proof pressure = 200×1.63×1.25 = 407 psi (minimum). The test was performed at MSFC by Propulsion Research Center (TD40) personnel in building 4655 on February 20, 2004. The HX was instrumented with a calibrated pressure transducer and 10 strain gages. Figure 17 shows the test setup. The numbering and locations of the strain gages are shown in figures 18–20. Figure 21 plots the pressure and strain versus time for the test. The pressure was maintained between 410 and 417 psi for a total of 634 s, meeting the targeted pressure and hold time. The HX successfully completed the proof test with no leakage. A finite element model of the HX was built to analyze the design for the expected test conditions. The proof test was also evaluated with this model. Figure 22 shows the model and the predicted radial strain for a maximum proof pressure of 430 psi. A comparison of the measured strain with the predicted strain at a pressure of 417 psi is shown in figure 23. Overall, with the exception of gages S2001, S2101, and S3003, the measured and predicted strains agree fairly well. The measured strain at these two locations is significantly greater than the predicted level. The strain gradient in these locations is steep and, therefore, sensitive to positional variation. The model uses nominal wall thickness while the as-built wall thickness is unknown. Since the strain in these two locations has a significant bending component, it is sensitive to the wall thickness to the order of t–2. Therefore, the difference at these locations is not unexpected. 31

Original page 43

Hydro Inlet SAFE-100a HX Pressure Transducer Strain Gauge Hookup Figure 17. SAFE-100a HX proof test setup. S2102 S2101 Figure 18. Braze cup side. 32

Original page 44

S1001 S2002 S1002 S2001 Figure 19. Side opposite braze cups. S3002 S3003 S4002 S4003 Figure 20. Barrel section. 33

Original page 45

600 500 Target Proof Pressure 420�10 psi ) 400 –6 10 � 300 200 Pressure (psi) Microstrain (in/in 100 0 –100 0 500 1,000 Elapsed Time (s) S1001 S1002 634 s S2001 S2101 S2002 S2102 S3002 S3003 S4002 S4003 Pressure 1,500 2,000 2,500 Figure 21. Strains and pressure versus time. –0.530×10–3 –0.387×10–3 –0.245×10–3 –0.102×10–3 –0.403×10–4 –0.183×10–3 –0.325×10–3 –0.468×10–3 –0.610×10–3 –0.753×10–3 X Y Figure 22. Finite element model radial strain. 34

Original page 46

600 500 ) –6 400 10 × 300 200 Microstrain (in/in100 0 –100 Test Predicted S1001 S1002 S2001 S2101 S2002 S2102 S3002 S3003 S4002 S4003 Gage Number Figure 23. Comparison of test and predicted strain. 35

Original page 47

REFERENCES 1. ASME Pressure Vessel and Boiler Code, ASME, New York, section III: subsection NH, pp. 26–123, 2001. 2. Materials Properties Manual, Rockwell International, Rocketdyne Division, secs. 2211 and 2212, 1986. 3. “Metallic Materials and Elements for Aerospace Vehicle Structures,” MIL–HDBK–5F, Government Printing Office, Washington DC, pp. 2-20–2-218, November 1990. 4. “Heat-Transfer Coefficient Calculation for a SAFE-100 Monolithic Brazed Heat Exchanger,” Memorandum ED25–02–25, Marshall Space Flight Center, November 12, 2002. 37

Original page 48

Form Approved REPORT DOCUMENTATION PAGE OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operation and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503 1. AGENCY USE ONLY (Leave Blank) 2. REPORT DATE March 2005 4. TITLE AND SUBTITLE 3. REPORT TYPE AND DATES COVERED Technical Memorandum 5. FUNDING NUMBERS Safe Affordable Fission Engine- (SAFE-) 100a Heat Exchanger Thermal and Structural Analysis 6. AUTHORS B.E. Steeve 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) George C. Marshall Space Flight Center Marshall Space Flight Center, AL 35812 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) National Aeronautics and Space Administration Washington, DC 20546–0001 11. SUPPLEMENTARY NOTES 8. PERFORMING ORGANIZATION REPORT NUMBER M–1131 10. SPONSORING/MONITORING AGENCY REPO NUMBER NASA/TM—2005–213609 Prepared by the Structures, Mechanics, and Thermal Department, Engineering Directorate 12a. DISTRIBUTION/AVAILABILITY STATEMENT Unclassified-Unlimited Subject Category 39 Availability: NASA CASI 301–621–0390 13. ABSTRACT (Maximum 200 words) 12b. DISTRIBUTION CODE A potential fission power system for in-space missions is a heat pipe-cooled reactor coupled to a Brayton cycle. In this system, a heat exchanger (HX) transfers the heat of the reactor core to the Brayton gas. The Safe Affordable Fission Engine- (SAFE-) 100a is a test program designed to thermally and hydraulically simulate a 95 Btu/s prototypic heat pipe-cooled reactor using electrical resistance heaters on the ground. This Technical Memorandum documents the thermal and structural assessment of the HX used in the SAFE-100a program. 14. SUBJECT TERMS heat exchanger, stress analysis, thermal analysis 17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION OF REPORT OF THIS PAGE Unclassified Unclassified NSN 7540-01-280-5500 38 15. NUMBER OF PAGES 48 16. PRICE CODE 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT OF ABSTRACT Unclassified Unlimited Standard Form 298 (Rev. 2-89) Prescribed by ANSI Std. 239-18 298-102