flow boiling and condensation experiment (fbce) for · pdf fileasgsr 2015 boiling and...

www.mudawar.com

Boiling and Two-Phase Flow Laboratory (BTPFL)ASGSR 2015 November 2015

Flow Boiling and Condensation Experiment (FBCE)

for the International Space Station

Issam Mudawar, Chirag Kharangate, Lucas O’Neill & Chris Konishi

Purdue University Boiling and Two-Phase Flow Laboratory (PU-BTPFL)

Mechanical Engineering Building, 585 Purdue Mall

West Lafayette, IN 47907

Mohammad Hasan, Henry Nahra, Nancy Hall & R. Balasubramaniam

NASA Glenn Research Center Fluid Physics and Transport Branch

21000 Brookpark Rd, Cleveland, OH 44135

Jeffrey Mackey

Vantage Partners, LLC 3000 Aerospace Parkway

Brook Park, OH 44142

1

https://ntrs.nasa.gov/search.jsp?R=20150023462 2018-05-12T00:31:51+00:00Z

www.mudawar.com

Boiling and Two-Phase Flow Laboratory (BTPFL)ASGSR 2015 November 2015 2

1 ge 10 ge

Moon

0.17 ge

Mars

0.38 ge

Satellites

Earth Orbiting Vehicles

Earth Orbiting Station

Martian Base

Astronaut Suit

Space Vehicle

Fighter AircraftAsteroid Landing

Examples of Systems Demanding Predictive Models of Effects of Gravity on

Two-Phase Flow and Heat Transfer

μ ge

www.mudawar.com


High-Capacity

Condensation Facility

Mini/micro-channel


One-g Flow

Boiling Facility

Falling-Film

Heating/Evaporation Facility

Parabolic Flight Flow

Boiling Facility

Parabolic Flight


NASA-Supported Facilities at Boiling & Two-Phase Flow Laboratory

3

www.mudawar.com


The proposed research aims to develop an integrated two-phase flow

boiling/condensation facility for the International Space Station (ISS) to serve as

primary platform for obtaining two-phase flow and heat transfer data in

microgravity.

Overriding objectives are to:

1. Obtain flow boiling database in long-duration microgravity environment

2. Obtain flow condensation database in long-duration microgravity environment

3. Develop experimentally validated, mechanistic model for microgravity flow boiling critical heat flux (CHF) and dimensionless criteria to predict minimum flow velocity required to ensure gravity-independent CHF

4. Develop experimentally validated, mechanistic model for microgravity annular condensation and dimensionless criteria to predict minimum flow velocity required to ensure gravity-independent annular condensation; also develop correlations for other condensation regimes in microgravity

Overriding Objectives of FBCE

4

www.mudawar.com


Consists of:

nPFH sub-loop

Water sub-loop

Contains three test modules:

Flow Boiling Module (FBM)

Condensation Module CM-HT for heat

transfer measurements

Condensation Module CM-FV for flow

visualization

Condensation Module CM-HT

Condensation Module CM-FV

Layout of FBCE

Flow Boiling Module (FBM)

5

www.mudawar.com


Primary 2014 – 2015 Research Topics

1. Both One ge and Parabolic flight flow boiling

experiments using single-sided and double-sided heat

walls

2. Modeling of CHF for single-sided and double-sided

heated walls at different orientations in Earth gravity

and in microgravity

3. Computational modeling of condensing film

6

www.mudawar.com


Flow Boiling Facility

LED Light

Source

LED Power Supply

Hard

Drives for

Flow Viz

Computerfor Flow Viz

FBM H1 Thermocouples

MirrorCamera

FBM Heater Temperature

Controller

FBM Heater Power Meters

FBM H1

Thermocouples

Camera DataAcquisition System

FBM Inlet FBM Outlet

FBM HeaterPower Supply & Meters

Reservoir

FBM Heater Temperature

Controller

FBM HeaterPower Supply & Meters

FBM Inlet FBM Outlet

Control Valve

Filter

Turbine Flow Meter

Preheater

Pump

Air

Liquid to AirHeat Exchanger

Flow Boiling Module

T

Watt-meter

T

Control Valve

Variac

T

Watt-meter

High-Speed

Camera

Variac

P P

T

1 atm

Control Valve

7

www.mudawar.com


Support Plates(Aluminum)

Outer ChannelPlates (Lexan)

FlowStraightener

O-rings

O-ringsCopper Slabs with Resistive Heaters

FC-72 Outlet

FC-72 Inlet

ChannelSide Wall Plate

(Lexan)

Transparent Walls

Ld Lh Le

FC-72Inlet

FC-72Outlet

Flow Straightener

H

Resistive Heater

Resistive Heater

Oxygen-Free Copper Slab

Oxygen-Free Copper Slab

W

Channel Height: H = 5.0 mmChannel Width: W = 2.5 mm

Development Length: Ld = 327.9 mmHeated Length: Lh = 114.6 mmExit Length: Le = 60.9 mm

FC-72 Outlet

FC-72 Inlet

P

T

PP

P P

T

Flow Boiling Module

8

www.mudawar.com


Heated Wall Design

9

www.mudawar.com


Top wall heating

Bottom wall heating

Double-sided heating

Horizontal Flow Boiling – Heated Wall Configurations

H1: q”w

Liquid

FlowVapor

W

HLiquid

Vapor

Layer

ge

Vapor Layer 1

H1: q”w

H2: q”w

Liquid

Vapor

FlowVapor

W

HLiquid

Vapor Layer 2

Flow

Vapor

Liquid

H2: q”wW

HH – δ1 Liquid

Vapor Layer

ge

ge

δ1

H – δ1 – δ2

δ1

δ2

H – δ1 – δ2

δ1

10

www.mudawar.com


Slightly subcooled flow at low velocity

51%

85%

CHF (28.2 W/cm2)

51%

87%

CHF (9.3 W/cm2)

59%

86%

CHF (18 W/cm2)

Bottom

Wall Heated

Top Wall

Heated

Top & Bottom

Walls Heated

Flowge

G = 394.8 – 403.4 kg/m2s (U = 0. 25 m/s)

ΔTsub,in= 3.6 - 5.1°C

Horizontal Flow Boiling – Flow Visualization

11

www.mudawar.com


∆Tsub,in = 24.5 – 31.0°C

∆Tsub,in = 3.3 – 7.7°C

xe,in = 0.03 – 0.18

G [kg/m2s]

CH

F [

W/c

m2]

0 1000 2000 30000

10

20

30

40

Horizontal Flow Boiling – Experimental CHF Results

Bottom heating

Top heating

Double-sided heating

12

www.mudawar.com


Separated Flow Model for Double-Sided Heating

13

www.mudawar.com


Interfacial Lift-off CHF Model

14

www.mudawar.com


CHF Model Predictions

U [m/s]

0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

CH

F [

W/c

m2]

10

15

20

25

30

35

40

45

H1

H2

MAE

Double-Sided

Heating Data

5.0%

5.7%

CHF Model Predictions1 ge

(Vertical Upflow)

µge

5

0

FC-72pout = 118.2 – 148.3 kPa∆Tsub,in = 2.8 – 8.1°C

0 0.5 1.0 1.5 2.0 2.50

10

20

30

40

U [m/s]

CH

F [

W/c

m2]

FC-72

pin = 99.3 – 161.8 kPa

∆Tsub,in = 1.9 – 8.4°C

Bottom heating (MAE = 5.6%)

Top heating (MAE = 27.4%)

Double-sided heating (MAE = 8.1%)

Microgravity &

One ge Vertical UpflowOne ge Horizontal Flow

15

www.mudawar.com


Single-sided Heating Double-sided Heating

Single-sided versus Double-sided Heating in Earth Gravity

θ = 90°

135° 45°

0°180°

225° 315°

270°

Upflow

Downflow

Upward-facing heater

Downward-facing heater

ge 135°

180°

225°

θ = 90°

270°

Upflow

Downflow

ge

Upward-facing heaterDownward-facing heater

0°

45°

315°

16

www.mudawar.com


CHF Predictions for Single-sided versus Double-sided Heating in Earth Gravity

0°

45°

90°

135°

180°

225°

270°

315°

0 5 10 15 20 30

5

10

15

20

CHF (W/cm2)

51025305

10

25

30

250

20

1520

15

25

Heater Ha

30

0°

45°

90°

135°

225°

270°

315°

180°

Upward-facing Heater

Downward-facing Heater

20

CHF (W/cm2)

FC-72

pin = 100 kPa

ΔTsub,in = 3°C

Single-sided Heating Double-sided Heating

U = 2.0 m/s1.5 m/s1.0 m/s0.5 m/s

U = 2.0 m/s1.5 m/s1.0 m/s0.5 m/s

U = 2.0 m/s1.5 m/s1.0 m/s

10152035 30 25 302510 1555 0 35

5

0

10

15

20

25

30

35

10

20

35

25

30

15

5

17

www.mudawar.com


Data Sharing Plans

18

www.mudawar.com


Impetus for Data Sharing

NASA Office of Physical Science Informatics (PSI) tasked with

organizing and distributing databases to researchers in the field

Large databases (terabytes of data) will be generated from FBCE

ISS experiment

Purdue-Glenn team will create organization structure for FBCE

databases to be provided to (PSI)

Purdue is presently exploring most effective means for packaging

data for ease of use by other researchers using recent FBM

Earth’s gravity data as example

19

www.mudawar.com


Organizational Structure of FBM Database

Folder Name:Summer 2015 FBM Testing

Organizational Documents

Publications, Presentation,

and Summaries

Data Folders

20

www.mudawar.com


Data Folders

Contain four filetypes:

Text Files containing raw sensor data output by data acquisition system

Matlab Scripts for processing raw data

Excel Spreadsheets containing all relevant parameters (e.g., pressure drop, heat

transfer coefficient, CHF) output by processing script

Image Files for flow visualization

With subfolders used to group data by operating conditions

Data

Folders

T&B, 2 Phase,

3psig Accum

T&B, Near Sat,

3psig Accum

T&B, Near Sat,

Accum Atm

Raw Data (Text Files)

Data Processing

(Matlab Scripts)

Processed Data (Excel

Spreadsheets)

Flow Visualization (Image Files)

B, Subcooled,

Accum Atm

Full description of file paths and data file structures found in “Organizational Documents” folder

21

www.mudawar.com


Computational

Modeling

22

www.mudawar.com


Study flow

condensation using

CFD solver Fluent

Select an appropriate

phase change model

Study heat transfer and

fluid flow

characteristics over a

broad range of

Reynolds numbers

Lay foundation for

future computational

modeling of

complicated flow

boiling processes

Objectives

Condensation

Test Facility

ANSYS Fluent Modeling

Pre-Heater Power

Control

Data Acquisition

System

Pre-Heater

Condensation Module

Modular Liquid-liquid Cooling System

FC-72 Flow Loop Cart

23

www.mudawar.com


• Lagrangian

- Smoothed-Particle Hydrodynamics (SPH) Method: Gingold & Monaghan

(1977), Lucy (1977)

- Multiphase Particle-in-Cell (MP-PIC) Method: Harlow (1955)

• Eulerian

- Level-Set Method (LSM): Osher & Sethian (1955)

- Volume-Of-Fluid (VOF) Method: Hirt & Nichols (1981)

- Coupled Level-Set/Volume-Of-Fluid (CLSVOF) Method: Sussman & Puckett

(2000), Enright et al. (2002), Tomar et al. (2005)

• Eulerian-Lagrangian

- Front Tracking Method: Unverdi & Tryggvason(1992), Tryggvason et al. (2001)

Numerical Approaches to Modeling Two-Phase Systems

24

www.mudawar.com


Lee Model (1980)Wu et al.(2007). Yang et al.(2008), Fang et al.(2010) …

Kartuzova & Kassemi (2011), Magnini et al. (2013) …

Mao (2009). Michita & Thome( 2010), Sun et al. (2012)

Phase Change Models

k

eff= a

gk

g+a

fk

f

Sg

= - Sf

= ria

gr

g

T -Tsat( )

Tsat

Sg

= - Sf

= ria

fr

f

T -Tsat( )

Tsat

where

where

for condensation (T < Tsat)

for evaporation (T > Tsat)

Rankine-Hugoniot jump condition

Schrage Model (1953)

25

www.mudawar.com


Rankine-Hugoniot jump condition

Schrage Model (1953)

Lee Model (1980)

Wu et al. (2007). Yang et al.(2008), Fang et al. (2010) …

Kartuzova & Kassemi (2011), Magnini et al. (2013)…

Pros:

1. Ease of implementation

Mao (2009). Michita & Thome (2010), Sun et al. (2012) …

Phase Change Models

Cons:

1. Allows for phase change only along

interface

2. Cannot maintain saturation

temperature

Pros:

1. Successfully used for

evaporating & condensing films

Cons:

1. Requires use of empirical coefficient γ

2. Allows for phase change only along

interface

Pros:

1. Ease of implementation

2. Successfully used for

condensation processes

Cons:

1. Not applicable for subcooled boiling

2. Requires use of empirical coefficient ri

26

www.mudawar.com


Computational Domain

Computational Domain, Governing Equations and Boundary Conditions

Boundary Condition

Axisymmetric centerline

k-ω SST turbulence model

Inlet uniform velocity from experimental data

Wall heat flux from experimental data

- Liquid Phase:

Lee model (2008)

Governing Equations

z

Modeled axi-symmetricregion

y

Centerline

Outlet

Inlet

Dh/2

r

Dh/2

zy

u Inlet

Outlet

Center-line

Cooling wall

Uniform mesh

xe = 0

FC-72

g

Gradually finer mesh near wall

L

r

Continuity Equations

Momentum Equation

Energy Equation

- Vapor Phase:

Sg

= - Sf

= ria

gr

g

T -Tsat( )

Tsat

Q = h

fgS

f

27

www.mudawar.com


Void Fraction Results: Climbing Film Regime

28

Gravity

Flow

Climbing film

Wall

3.25 mm

G = 106.5 kg/m2s

xe,in = 1.16

www.mudawar.com


Predictions of Average Heat Transfer Coefficient

0 50 100 150 200 250 3000

200

400

600

800

1000

1200

1400

G [kg/m2s]

havg

[W/m

2K

]

_

Experimental

Computational

FC-72

29

www.mudawar.com


Future Computational

Modeling

30

www.mudawar.com


Need for Development of In-House Code

Fluent is able to accurately replicate experimental results, but

“tuning” necessary for phase change model means it is not a

reliable predictive tool

Difficult to work with Fluent because solver code is proprietary

Fluent very robust and can tackle wide range of problems,

making it slower than a dedicated research code

Fluent does not utilize cutting edge multi-phase computational

techniques

31

www.mudawar.com


Current Work

Current Work

Working with Prof. Carlo Scalo’s

group at Purdue University to

develop a 2-D code using best

available computational techniques

Performing comparisons with Fluent

to quantify in-house solver

performance

Future Work

Use proposed 2-D code to run select cases which can be represented reasonably well by 2-D domains (e.g., axi-symmetric flow condensation, slug flow)

Scale 2-D code up to 3-D, highly parallelized solver, which will include turbulence effects, to be run on Purdue supercomputing cluster

Begin comparing data for transient cases with 3-D geometry to prior experimental studies

32

www.mudawar.com

Boiling and Two-Phase Flow Laboratory (BTPFL)ASGSR 2014 October 2014

The Flow Boiling and Condensation Experiment is supported by

NASA grant NNX13AB01G.

This work was supported by NASA Space Technology Research

Fellowship NNX15AP29H.

Thanks to Professor Scalo and Victor Sousa for creating the Fluent-

comparison figures.

Acknowledgements

33