THERMAL MECHANICAL PRELIMINARY ANALYSIS ELM COIL ALTERNATE DESIGN

  • Slides: 29
Download presentation
THERMAL & MECHANICAL PRELIMINARY ANALYSIS ELM COIL ALTERNATE DESIGN Interim Review July 26 -28,

THERMAL & MECHANICAL PRELIMINARY ANALYSIS ELM COIL ALTERNATE DESIGN Interim Review July 26 -28, 2010 In-Vessel Coil System Interim Review – July 26 -28, 2010 1

Outline • BOUNDARY CONDITIONS • • • NUCLEAR & RESTISTIVE HEAT GENERATION LORENTZ &

Outline • BOUNDARY CONDITIONS • • • NUCLEAR & RESTISTIVE HEAT GENERATION LORENTZ & PRESSURE LOADS RADIATION ; CONDUCTION ; COOLING WATER @ 6 m/sec MAGNESIUM OXIDE to COIL & JACKETS STEADY STATE SANDWICH STRESS RESULTS : – THERMAL + PRESSURE LOAD RESULTS – THERMAL + PRESSURE + LORENTZ LOAD RESULTS • DESIGN IMPROVEMENT STRATEGIES – THERMAL + PRESSURE LOAD RESULTS – THERMAL + PRESSURE + LORENTZ LOAD RESULTS – SUB MODELING ; CORRECTION STRATEGY • CONCLUSIONS / PLAN: In-Vessel Coil System Interim Review – July 26 -28, 2010 2

Nuclear Heat Operating Modes

Nuclear Heat Operating Modes

NUCLEAR HEAT GENERATION (W/M^3) 0. 264 m The Poloidal Leg Applies a Similar Shaped

NUCLEAR HEAT GENERATION (W/M^3) 0. 264 m The Poloidal Leg Applies a Similar Shaped Function The Toroidal Leg Nuclear Heat is Applied Based on a Curve fit of data from University of Wisconsin Team IVC Interim Design Review – 26 -28 July 2010 4

IDEALIZED LOAD DIAGRAMS Load Thermal + Pressure + Lorentz Loading Thermal + Pressure Loading

IDEALIZED LOAD DIAGRAMS Load Thermal + Pressure + Lorentz Loading Thermal + Pressure Loading Time TRANSIENT STEADY STATE APPLIED WATER COOLING TEMPERATURE VS TIME 400 3, 000 sec 30, 000 Pulses Unknown Spectrum 9, 000 sec 350 TEMPERATURE Load 5 hz 300 250 TEMP 1 200 TEMP 2 150 TEMP 3 100 50 Time 0 0 500 1800 2301 3600 TIME (SECONDS) 3601 4100

ELM LORENTZ LOAD VS POSITION 400, 000 RADIAL UNIT LOAD Vs POSITION 300, 000

ELM LORENTZ LOAD VS POSITION 400, 000 RADIAL UNIT LOAD Vs POSITION 300, 000 200, 000 100, 000 SECTOR 5 FE MODEL LOADS Fx Fy Fz ELM_MD_BOT 132, 271 -31, 397 -32, 429 ELM_MD_BLC 130, 406 -8, 635 -41, 265 ELM_MD_LFT 300, 308 -10, 272 7, 491 RADIAL LOAD (N) Critical Quadrant SECTOR 5 0 0. 0000 SECTOR 4 2. 0000 4. 0000 6. 0000 100, 000 10. 0000 SECTOR 12. 00006 SECTOR 2 SECTOR 3 200, 000 (BOT) 300, 000 400, 000 8. 0000 (BLC) (LFT) (TRC)(RHT) POSITION SECTOR #5 UNIT LOADS ARE MORE CRITICAL IN THE LOWER LEFT QUADRANT

SANDWICH DESIGN Section View No Hard Mechanical Attachment for tension Axial Translation Is Allowed

SANDWICH DESIGN Section View No Hard Mechanical Attachment for tension Axial Translation Is Allowed DESIGN CONCEPT ALLOWS THERMAL DISPLACEMENT WITH SUPPORTS TO REACT LORENTZ LOAD IVC Interim Design Review – 26 -28 July 2010 7

ELEMENT MESH Flexible Mounts To Facilitate Thermal Growth Symmetric Boundary Rigid Boundary UNIFORM HEXAHEDRAL

ELEMENT MESH Flexible Mounts To Facilitate Thermal Growth Symmetric Boundary Rigid Boundary UNIFORM HEXAHEDRAL MESH

STEADY STATE TEMPERATURE ANALYSIS FULL OPERATING CONDITIONS Resistive Heat Generation Nuclear Heat Generation Cooling

STEADY STATE TEMPERATURE ANALYSIS FULL OPERATING CONDITIONS Resistive Heat Generation Nuclear Heat Generation Cooling Water Applied

THERMAL BOUNDARY CONDITIONS Radiation Surfaces with View Factor =1 (dark blue surfaces) ar cle

THERMAL BOUNDARY CONDITIONS Radiation Surfaces with View Factor =1 (dark blue surfaces) ar cle N Nu GE H Unspecified Surface Boundaries are conservatively assumed to be Adiabatic Temp in =105. 7 C Temp out =131. 5 C Conduction into Foundation at 100 C at all foundation interfaces The Copper Coil Temperature Distribution is an Equilibrium of all Combined Effects

RADIATION ASSUMPTIONS All Form / View Factors equal to 1. 0 Incident Radiation is

RADIATION ASSUMPTIONS All Form / View Factors equal to 1. 0 Incident Radiation is very small from 100 C Far Field Emissivity is a Hemispherical Average Across all wavelengths and directions IVC Interim Design Review – 26 -28 July 2010 12

Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation Max Temp

Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation Max Temp = 476 C on Bracket TEMPERATURES ARE REASONABLE and WITHIN OPERATING LIMITS OF MATERIALS

Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation Stainless Steel

Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation Stainless Steel Jackets Max Temperatures ( 472 C = F) are within the limits of Stainless Steel With Cooling Water

Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation Applied Boundary

Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation Applied Boundary is: Temp in =105. 7 C Temp out = 127. 2 C Copper Coil shows net equilibrium of hottest leg 1. 3 degrees higher than the Applied Boundary This trivial difference is explained by 1. ) Initial Gap Contact conditions 2. ) Number of elements through wall thickness 3. ) Graphical Contouring of results 4. ) Net equilibrium balance of all heat transfer effects The Coil Temperatures are Consistent with Hand Calculations and the Net Energy Balance of all Applied Thermal Loads

Steady State Fault Condition with Radiation Cooling Max Temperature Predicted on Surfaces that Exclude

Steady State Fault Condition with Radiation Cooling Max Temperature Predicted on Surfaces that Exclude Radiation Fault Condition (No Water Cool or Resistive Heating) with Far Field Radiation Results in Temperatures that are within Material Capacity (316 SS Melt at 1375 C)

Steady State Fault Condition with Radiation Cooling Conservative Max Copper Temperature= 918 C Melting

Steady State Fault Condition with Radiation Cooling Conservative Max Copper Temperature= 918 C Melting 1, 078 C Fault Condition (No Water Cool or Resistive Heating) with Far Field Radiation Results in Temperatures that are within Material Capacity (Cu. Cr. Zr Melt at 1, 078 C)

STEADY STATE STRESS ANALYSIS THERMAL & DISRUPTION LOADS

STEADY STATE STRESS ANALYSIS THERMAL & DISRUPTION LOADS

Steady State Pressure + Thermal + Lorentz Load Support Reaction Loads RSYS 12 (Newtons)

Steady State Pressure + Thermal + Lorentz Load Support Reaction Loads RSYS 12 (Newtons) FX FY 14038. -0. 16031 E+06 . FZ -19, 761. Poloidal +Z +Z Toroidal +Y RSYS 14 (Newtons) FX FY -36461. -0. 11263 E+06 +Y FZ 13456 Typical Bracket Reaction Loads: FY =36, 036 lbs is away from the Reactor on the Toroidal Bracket FY = 25, 178 lbs is away from the Reactor on the Poloidal Bracket IVC Interim Design Review – 26 -28 July 2010 19

Steady State Pressure + Thermal + Lorentz Load Displacements +Y +Y 0. 0066 m

Steady State Pressure + Thermal + Lorentz Load Displacements +Y +Y 0. 0066 m = 0. 259 in The Displacements are Reasonable for the Specified Boundary Conditions Lorentz Loads Acting Down Toward The Reactor

Steady State Mechanical + Thermal Loads + Lorentz Max Principal Stress Restraint Location Stress

Steady State Mechanical + Thermal Loads + Lorentz Max Principal Stress Restraint Location Stress shows: 1. ) Bending across Restraint Location 2. ) Exterior Jackets in Compression 3. ) Interior Copper Coil in Tension The Stresses are Excessive However they are Manageable With the current strategies in progress

Steady State Mechanical + Thermal Loads von Mises Stress Copper Coil 0. 450 e

Steady State Mechanical + Thermal Loads von Mises Stress Copper Coil 0. 450 e 8 Pa = 6, 526 Psi Copper Coil 0. 185 e 8 Pa = 2, 683 Psi The Max Copper Coil Stress of 6. 5 ksi will be Reduced with Bridge Support

Steady State Mechanical + Thermal Loads + Lorentz von Mises Stress Max Copper Coil.

Steady State Mechanical + Thermal Loads + Lorentz von Mises Stress Max Copper Coil. 184 e 9 Pa = 26, 686 psi Copper Stresses Have Positive Limit Stress Margins and Negative Fatigue Margin Additional Section will be used to Redistribute These stresses

REVISED ANALYSIS With Bridge Support IVC Interim Design Review – 26 -28 July 2010

REVISED ANALYSIS With Bridge Support IVC Interim Design Review – 26 -28 July 2010 24

Updated - Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation

Updated - Steady State Temperatures With Heat Generation ; 6 m/s Water Cooling Radiation Revised Plan July 22, 2010 Inlet Temp 70 C Outlet Temp 120 C Bridge Support to react out Lorentz Loads IVC Interim Design Review – 26 -28 July 2010 25

Sub Modeling Plan Classical Cut Boundary Displacements applied from Global analysis Stress to be

Sub Modeling Plan Classical Cut Boundary Displacements applied from Global analysis Stress to be evaluated for Variable Spring Stiffness and / or applied Preloads Sub Models will be used to test out various strategies in critical areas such as the corners or restraint locations to assure that the best design options are thoroughly investigated

Steady State Displacements Pressure + Thermal +Y Pressure + Thermal + Lorentz +Y The

Steady State Displacements Pressure + Thermal +Y Pressure + Thermal + Lorentz +Y The vertical displacements are reasonable for the specified boundary conditions

Steady State Pressure + Thermal Loads von Mises Stress Copper Coil 0. 37 e

Steady State Pressure + Thermal Loads von Mises Stress Copper Coil 0. 37 e 8 Pa = 5, 366 Psi Bridge Support can be used to Shape and Redistribute Stresses on the Coil Additional Shaping and Stiffness Changes with Sections changes will be used to React out Stresses

Steady State Pressure + Thermal + Lorentz Loads Von Mises Stress Max Copper Coil=

Steady State Pressure + Thermal + Lorentz Loads Von Mises Stress Max Copper Coil= 0. 18 e 8 Pa = 2, 465 psi Bridge Support can be used to Shape and Redistribute Stresses on the Coil Additional Shaping and Stiffness Changes with Sections changes will be used to React out Stresses

CONCLUSIONS / PLAN • The Sandwich Design will be a viable option. Current progress

CONCLUSIONS / PLAN • The Sandwich Design will be a viable option. Current progress shows stress levels can be shaped with design changes. Additional changes to meet fatigue requirements will be completed as required. (Post PDR) • The Christmas Tree and Sandwich Design evaluated for merits in the coming weeks for a down select. (PRE PDR) • Material property testing and MGO Interface boundary determined for accurate results. (PRE PDR) • Revised Toroidal & Poloidal Nuclear Heat Functions Updated with revised coolant temperatures. (PRE PDR) • All three load case scenarios including Transient and Steady State loadings will be completed. (Pre PDR) • Steady State and Transient Load Cases to be completed with Sub-modeling to resolve stress issues. (Post PDR) IVC Interim Design Review – 26 -28 July 2010 30