Microsoft Word - flibe simulant v2.doc

Transcription

1 )XQGDPHQWDO7KHUPRIOXLG([SHULPHQWV 0RGHOLQJ Subtask 3-1 of the US-Monbusho Collaboration Proposal Presented by Neil Morley and Mohamed Abdou Contributions from Fusion Science and Technology, UCLA: M Abdou, K. Gulec, A. Ying, N. Morley, S. Smolentsev, B. Freeze, T. Sketchley, M. Dagher, Contributions from UCLA Faculty M. Abdou, V. Dhir, N. Ghoniem, J. Kim, R. Kelley Other US Institutions: PPPL, ORNL, INEEL, ANL US-Monbusho Collaboration Meeting Makuhari, Japan, February 22-25, 2000

2 7KHUPRIOXLG7DVN2EMHFWLYHV 1. Understand Underlying Science and Phenomena for Flibe Flow and Heat Transfer Issues through: a. Conducting Experiments Using Flibe Simulants b. Modeling and Analysis of fundamental phenomena 2. Compare Experimental and Modeling Results to Provide Guidance and Design Database for both Flibe Blankets and Next Generation Stage of Larger Flibe Experiments. 3. Identifying and assessing new innovative techniques for enhancement of Flibe heat transfer (a major feasibility issue for Flibe blanket designs)

3 0DLQ$UHDVRI&ROODERUDWLYH6FLHQWLILF,QWHUHVW Hydrodynamics at liquid/vacuum interfaces and at solid walls of Flibe simulants flowing on flat and curved plates, and in closed channels and swirl pipes, with and without MHD effects Heat transfer at liquid/vacuum interfaces and at solid walls of Flibe simulants flowing on flat and curved plates, and in closed channels and swirl pipes, with and without MHD effects Sub-areas of Interest for Collaborative Efforts Identification of instrumental and experimental techniques: Radiant heating, laser and ultrasonic surface topology reconstruction, infra-red temperature measurements, laser Doppler anemometety, others. Development and benchmarking of new modeling techniques: MHD/turbulence interactions and turbuelnce/free surface or wall interatctions in k-e, DNS, LES

4 +HDW7UDQVIHU(QKDQFHPHQWIRU))+5DQG/: +LJK+HDW)OX[$SSOLFDWLRQLVFULWLFDO Inlet temperature for Flibe is > 500 C Maximum operating temperature of FFHR structural material is estimated at < 600 C. Maximum operating temperature for LW plasma compatibility is estimated at < 600 C Film temperature rise must be < 50 C to keep operating window open Toroidal field = 10T Flibe property for 34% BeF2 at T=500 C FW heat flux = 0.2 MW/m 2 Divertor heat flux = 1.5 MW/m 2 Material compatibility issues Divertor Flibe Velocity (m/s) Hydraulic diameter: 2.5 cm (can be optimized) Desired film temperature drop (50 K) FW

5 0+'$IIHFWV+HDW7UDQVIHUIRU/RZ(OHFWULFDOO\ &RQGXFWLQJDQG+LJK3UDQGWO1XPEHU)OXLGV q Only a limited amount of data exist on the effects of MHD on heat transfer in a closed channel for high Prandtl fluids q No data available on the effect of MHD on turbulent characteristics and heat transfer of high Prandtl fluids for free surface flows q Therefore, data and models for both free surface and closed channels should be of interest to the US and Japan. Available data shows that the decrease in the heat transfer at a Hartmann wall due to modification of turbulent eddies by the magnetic field is correlated as: Nu/Nu o = N where N is the interaction parameter (based on the hydraulic diameter of the closed duct). Reference: E. Blums et al., Heat and Mass Transfer in MHD Flows, Chapter 6

6 3UREOHPVLQDSSOLFDELOLW\RIGDWD Taken over limited parameter range The above correlation is valid for Re > Re cr and N <0.1; Re cr = 250 Ha (Ha number based on hydraulic diameter) Similarity issues with real fusion needs Data taken at Hartmann wall, not sidewall, in rectangular channel. Entrance effects may still dominate (transition of 3D turbulence seed, instead of generation of MHD turbulence Wrong aspect ratio duct, Proper definition of length scales

7 Explanation: 1HHGIRU&RXSOHG0RGHOLQJ(IIRUW Fusion designs require practical computational models that must utilize input data from more elaborate theory and experiments. ( * Note that theory implies Large Eddy Simulation (LES) and Direct Numerical Simulation (DNS).) Collaboration: 1. Enhance LES and DNS numerical techniques to: understand near-wall and near-surface turbulence structures in magnetic fields, and provide input data for practical computational models. 2. Develop practical computational models using input from experiments, LES and DNS.

8 )XVLRQ7KHUPRIOXLG)ORZ)DFLOLWLHVDW8&/$ 0H*$/RRS Liquid Metal Flow Loop Coupled with Large Volume Magnetic Field Facility for open and closed channel MHD experiments Status: Operating since 1992, Upgrade to toroidal magnet in progress 0H62-HW Liquid Metal Flow Discharge System and Vacuum Enclosure Status: Operational - rectangular jet deformation experiments underway )OL+\)OLEH+\GURG\QDPLF6LPXODWLRQ)DFLOLW\ Water/KOH Discharge System and High Field Magnet Status: Design and Construction Other non-fusion UCLA flow facilites include water tunnels, wave tanks, and diagnostic systems for precision turbulence measurements.

9 g r Outlet T control System inlet Filter Fluid In Temp, Fluid Height Momentum Divertor / Dissipater Bulk Velocity Linear Controller Output Sink Reservoir Tank Degasser DAQ Current design specs Butterfly Valve Flow-meter Temp 2 m3 volume Chiller/Heater On/Off Filter up to 100 l/s flowrate up to 10 m/s velocity Actuated On/Off Valve Temp T control from 5 to 50 C CPVC/Lexan/316 SS structure Rotatable Joint Flow Controlling Valve Test section here Vibration Isolating Coupling Pump

10 g r Reservoir Tank Actuated On/Off Valve Vibration Isolating Coupling Test Section Rotatable Joint Flow Controlling Valve Pump Discharge Tank Sink

11 g r Reservoir Tank Degasser Filter Flow Controlling Valve Flange Vibration Isolating Coupling Rotatable Joint Pump

12 )OL+\([SHULPHQWDO*RDOVDQG6FKHGXOH -1 st collaboration year: Loop construction at UCLA and shakedown tests 1 st through mid-2 nd collaboration year Free surface flow on curved backwall as simulation of thin liquid Flibe wall design. Free surface flow on long, inclined plane for benchmark of fully developed free surface turbulence simulations using k-e and DNS. Continued shakedown of various diagnostic techniques: e.g. fast photography, LDA, ultrasonic fluid height, holographic temperature Mid 2 nd through 3 rd collaboration year Exploration of various heat transfer enhancement techniques for open channel flows Testing of heat transfer enhancement techniques for closed channels 4 th 6 th collaboration year (and beyond) MHD/turbulence interactions and effect on heat transfer in closed and open channel tests

13 )OL+\([DPSOH7HVW6HFWLRQV Inclined Plate Heat Transfer Test Curved Plate Heat Transfer Test

14 INCLINED PLANE EXPERIMENTAL TEST SECTION To deliver data on developing and fully-developed turbulence for comparison to DNS algorithm and k-e RANS modeling. Test section width is determined taking into account boundary layer Nozzle Exit width = 2 δ + effective test section δ = 0.16 x 6 / 7 ρu µ 1/ 7 δ = To obtain a fully developed turbulent liquid layer in the test section, the distance required from the nozzle exit is for U = 5 m m / s for h = m width = 0. 4 m at x = 2 L m fully developed Flow Channel Top View width Cylindrical to Rectangular Flow Diffuser L fully developed 60 Dh Replaceable Honeycomb Replaceable Test Section Flow Area Contraction Components m Measurement Region Boundary Layers on the Side Walls Flow Straightener Section Convergent Nozzle Section Replaceable Screen Optically Transparent Back W all Section

15 CURVED WALL EXPERIMENTAL TEST SECTION Designed to achieve similar hydrodynamic conditions of the thin liquid flibe wall base case Observe the gross behavior of flow: i.e. attachment, wave trains, flow depth near sidewall - High Speed Camera 1000 frame/s, 512*256 pixel - Strobe with variable frequency Measure flow rate and fluid depth for comparison to numerical models - Pressure sensors, flow meter and thermocouples - Ultrasonic and laser height measurement technique

16 Velocity fluctuation diagnostics 2-D LDA (Laser Doppler Velocimetry): System Spectra Physics Model 165 Argon-Ion laser: nm green, 488 nm blue Isomet 1201E-1 Acoustic-Optic Modulators (f center :40 MHz) Isomet 221A-2 digital drivers (frequencies 40 Mhz and 43 MHz) Hamamatsu R928 PMT (converts light into voltage versus time signal) A Gage Compusscope 265 A/D card (130 MHz) Labview Development Suite Flow field visualization and velocity distribution measurement using O 2 bubble (compatibility with KOH not yet determined) anode: 25 µm platinum wire, cathode: tungsten wire 200 mm downstream.

17 April 2001 Both UCLA FliHy and Tohoku loops operational before start of collaboration April 2002 April 2003 April 2004 April 2005 April 2006 April 2007 Test section design and construction (US and Japan) Hydrodynamics and Heat Transfer Experiments (UCLA and Tohoku) Fli-Hy MHD Experiments procure magnets test magnets Upgrade (UCLA) Instrumentation and Experimental Techniques Development (US and Japan) Pre-design of Large Flibe Loop Practical Computational Modeling (US) LES Basic Modeling Advanced Modeling Including MHD Effects DNS Modeling Advances (Japan)

18 8QLTXHWKHUPRIOXLGFDSDELOLWLHVDW8&/$LQ FRRSHUDWLRQZLWKRWKHU86-DSDQLQVWLWXWLRQV PDNHVIRUDVXFFHVVIXOFROODERUDWLRQ Instrumentation Laser doppler velocimetry Micrometer & ultrasonic flow depth probes Bubble/dye flow visualization Holographic temperature profiling Laboratory Facilities Multiple flow loops Multiple magnets Low voltage high current power supplies (from PPPL and MIT) Special materials handling capabilities (Be, Flibe) High bay space and high load crane Vacuum systems Computational Tools DNS/LES/MHD codes Free Surface Codes Parallel computing clusters UCLA Fusion Science and Technology Group experience Magnet design and construction Thermofluid/MHD experimentation MHD/free surface modeling Interested UCLA faculty with worldwide reputations - Nasr Ghoniem: Fusion materials - John Kim: DNS and MHD - Robert Kelly: Free surface flow - Vijay Dhir: Fluid heat transfer

19 0DMRU([LVWLQJ0DJQHW )DFLOLWLHVDW8&/$ MicroTor R = 40 cm, a = 10 cm 80 cm 24 Tara Field Coils for New MeGA- Loop

20 3ODQVIRUWKH0H*$/RRS Upgrade magnet set to moderate field solenoid or torus (20 additional coils and high current power supply acquired) Investigate effect of field gradients and orientation on free surface liquid metal flows (feasibility of passive flow MFE liquid wall concepts) Investigate effect of short pulsed toroidal and poloidal fields on free surface liquid metal flow (liquid motion in present large plasma test devices, e.g. NSTX) Investigate potential for in-situ control of free surface flows with applied electric currents (feasibility of actively controlled MFE liquid wall concepts)

21 )/(;)OH[LEOH*HRPHWU\0DJQHWLF)LHOG)DFLOLW\ Basic design, Iron Core B > 1.8 T at 40 cm gap Adjustable gap size and shape using interchangeable fluxhorns Easy access to working volume Adjustable orientation to gravity Close coupling to the MeGA-Loop, and Fli-Hy Magnet gap If higher fields are needed for relevant heat transfer tests, then B > 4T possible with high-current air-core solenoid (US-Monbusho) Design of small, low cost 4T test coil is underway in collaboration with PPPL. Basic Design: 2 T FLEX concept with coils on side and iron yoke

22 Direct Experience with Flibe 6DOW6HH6DZ Heat transfer experiments in open and closed channels Molten Salt flows from supply reservoir to collection reservoir through thermofluid test section of interest System rotates with See-Saw action reversing supply and collection reservoirs (no pumps or pump seals, no pipe connections) Many different tests possible, including MHD in FLEX (see inset picture) Flibe safety: Completely enclosed system Procedures need to be established in conjunction with INEEL safety experiments and local UCLA EH&S Office Heater Test article rotates Film Flow Reservoir Test article rotates

23 Advantages: )HDWXUHVRI)OLEH6LPXODWLRQZLWK $TXHRXV.2+6ROXWLRQ Low cost for working liquid Low operating temperature Wide material compatibility and low material cost Large flow volumes and flow rates possible for free surface flow tests Transparent medium for optical flow measurements Scaling favors reduced size and flow velocity tests Relatively high electromagnetic parameters for simulation of MHD/turbulence interactions Concerns: Some health hazard and corrosive characteristics, but good materials and safety procedures have already been identified. High vapor pressure at elevated temperatures

24 HYDRODYNAMIC SIMILARITY CONDITIONS For Re and Fr Number Equality U U exp base ρ = ρ base exp µ µ exp base 1/ 3 L L exp base ρ = ρ base exp µ µ exp base 2 / 3 For Re and We Number Equality U exp = U base µ µ base exp ρ ρ exp base L exp L base µ = µ exp base 2 ρ ρ base exp σ σ base exp Re and Froude number similarity should be taken into account. * The effect of back wall curvature on the hydrodynamic characteristics of the flow is taken into account by modifying the Froude number using acceleration due to centrifugal force Fr 2 U = gl Fr 2 U c = = ach Similarity condition for the modified Froude number is geometric, and independent of thermophysical properties of the operating fluid. R h 2 U a c = R

25 WATER AND AQUEOUS KOH ARE GOOD FLIBE SIMULANTS Candidate operation fluids for experimental simulation study C p k el Pr Flibe E-07 1 Water 5 C E-07 2 Water 25 C E-07 3 Water 50 C E-07 4 KOH 35% wt 5 C E-07 5 KOH 43% wt 5 C E-07 6 KOH 35% wt, 50C E-07 Hydrodynamic scaling of Candidate Fluids for Cliff operating fluid SCALING (Re+Fr) U base /U exp L base /L exp Note: KOH cases give a closer match to We number as well

26 Key MHD Physical Properties & Scaling Parameters of Interest Properties Flibe KOH+Water HTS 2 Working Temperature C Density ρ (kg/m 3 ) Electrical Conductivity σ (1/Ωm) Dynamics Viscosity µ(kg/ms) Important Factors for Heat Transfer and MHD Effect Considerations Prandtl Number C p µ/k Hartmann Factor (σ/µ) 1/ Interaction Factor 1 (σ/ρ) Notes 1. All Flibe designs are not fully laminarized. The interaction number indicates the amount of turbulent modification and heat transfer degradation. 2. Based on data for 53 % KNO 3, 40 % NaNO 2, 7 % NaNO 3, melting point =142 C (Electrical conductivity data is at 400 C, and should be considerably lower at 200 C) KOH solution at elevated temperatures has high electrical conductivity for MHD turbulence interaction studies. (However, it is uncertainty whether the vapor pressure would create difficulties for free surface heat transfer experiments.)

27 ([SHULPHQWDO6SDFHRIWKH 6HH6DZ)/(;µ7HVW )DFLOLW\XVLQJ'LIIHUHQW6LPXODQW0DWHULDOV Average fall velocity = m/s over a vertical fall distance of 75 cm Film thickness = 1-10 cm Red:Flibe/Purple:KOH(15%)+water/ Green:HTS Open Symbols= 4 T Solid Symbols = 2 T Re/Ha

28 8&/$([SHULPHQWVZLWK.2+ZLOOFRPSOHPHQW +76H[SHULPHQWVLQ-DSDQ Data from alternate simulant for comparison Enables high flowrate experiments Enables non-invasive optical measurements to quantify turbulence characteristics Flexible, low temperature facility for fast turnaround on multiple test sections High field magnet expertise and power supplies (developed for Liquid Metal MHD experiments at UCLA) can be used for Flibe simulant MHD/turbulence experiments