Lecture 3 Forced Convection

14.5 Release

Heat Transfer Modeling using

ANSYS FLUENT

2013 ANSYS, Inc.

March 28, 2013

Release 14.5

Outline
Introduction Heat Transfer Coefficient
Laminar and Turbulent Boundary Layers
Modelling Heat Transfer The Reynolds Analogy

Turbulence Modelling and Dynamic and Thermal Wall Functions

Case Study - Modelling Heat Transfer for Non-Equilibrium and
Complex Flows

Post Processing

2013 ANSYS, Inc.

March 28, 2013

Release 14.5

Heat Transfer Coefficient

Influence of:

Mechanism

Fluid

h (W/m2K)

Natural
Convection

Gases

5 30

Water

100 1000

Geometry, fluid properties, etc.

Gas

Importance of the boundary layer

Water
Forced
Convection

T0

Ts

Oil

Local heat flux

T
y

300 12,000
50 1,700

Liquid metal

6,000 110,000

Boiling

3,000 60,000

Condensation

5,000 110,000

Phase Change

y 0 x k f

10 300

hx Ts T0
y 0

1
( x) h( x) Tp T0 dx h Tp T0
y 0
L0
L

Mean heat flux

2013 ANSYS, Inc.

March 28, 2013

Release 14.5

Boundary Layers
Important parameters are bulk velocity, bulk temperature, and
pressure gradient
U 0 , T0
Laminar

Transition

laminar

Turbulent

turbulent

u
~
u

Dimensionless variables:
U0

2013 ANSYS, Inc.

March 28, 2013

~ T Ts
T
T0 Ts

Release 14.5

Boundary Layers
Laminar boundary layer
Mixing is characterized by the ratio of viscous boundary layer thickness to
thermal boundary layer thickness.

Turbulent boundary layer

Mixing is primarily governed by turbulence.

1
T

Heat transfer coefficient use an available correlation for the friction

coefficient, Cf
Laminar Boundary Layers (exact)
C f ,x

0.664
Re1x/ 2

Nu x 0.332 Re1x/ 2 Pr1/ 3

Turbulent Boundary Layers (empirical correlations)

C f ,x
2013 ANSYS, Inc.

March 28, 2013

0.0592
Re1x/ 5

Nu x 0.0296 Re 4x / 5 Pr1/ 3
5

Release 14.5

Modeling Turbulent Heat Transfer

RANS equations

u u u

T T T

u
P u

u
v

uv
y
x y y
x

T
T
C p u
v
y
x

k
C p vT
y y

Boussinesq approximation for Reynolds stresses

Turbulent viscosity, T, is calculated from some turbulence model:

uv T

u
y

By analogy, PrT = 0.85 (from experimental data)

vT DT
2013 ANSYS, Inc.

March 28, 2013

T
y

DT
6

T
PrT
Release 14.5

Turbulent Boundary Layer Structure

Velocity profile exhibits layered
structure identified from
dimensional analysis

Inner layer
U
yU
2.5 ln
5.45
U

Viscous sublayer Viscous forces

dominate, velocity depends on , w,
, y.

U
U

Outer
layer

Outer layer Depends on mean flow

characteristics

Viscous
sublayer

Overlap layer Log law applies

Dissipation dominates production

in the viscous sublayer region.
March 28, 2013

Upper limit
Depends on
Reynolds number

60

Loss

Fully-Developed Pipe Flow

yU

Dissipation of k

Gain

TKE production and dissipation are

nearly equal in the overlap layer
(turbulent equilibrium)

2013 ANSYS, Inc.

Fully turbulent
or
Buffer layer
Log-law
region
or
Blending
region

Diffusion of k
Production of k

10

30

yU

Release 14.5

Effects of Transition
Spurious jump of Cf and h at transition from
laminar to turbulent flows (Rex > 5e5)

h(x)

Natural transition is a complex phenomenon

(for RANS)

(x)
U , T

RANS: k-kL-w and Transition SST models can

be used for natural transition, bypass
transition, separation induced transition

Tw

xc

Use if extent of laminar flow region is

significant
U , T

Laminar

Transition

Turbulent

laminar

2013 ANSYS, Inc.

March 28, 2013

turbulent

Release 14.5

Boundary Layer Heat Transfer

Impact on numerical modeling

Use of wall functions for y+ >> 1 (when hypothesis are fulfilled)

Sensitivity of the results to y+ (transition, low-Re effect) and Pr

When hypothesis fails (Non-equilibrium boundary layers, recirculation,

stagnation, transition), we need to correctly resolve both the momentum
AND thermal viscous sub-layer (y+ < 1)

This is straightforward for Pr ~ 1 or Pr < 1.

When Pr is greater than 1, the thermal sublayer is much thinner than the
viscous sublayer.

Small sensitivity to grid resolution (provided that the momentum boundary

layer is correctly predicted

2013 ANSYS, Inc.

y+ 1 and ~10 cells for 1 < y+ < 30

March 28, 2013

Release 14.5

BL Heat Transfer Example Abrupt

Pipe Expansion
Abrupt pipe expansion (non-equilibrium boundary layer, recirculation,
wall heat transfer)
Flow

Re D 40,750

Mesh: y+ ~ 1, 50
Inlet: Fully-developed turbulent pipe flow.
Models:

RKE with EWT, SST k

Enhanced wall treatment (for y+ ~ 1 mesh)

Standard wall functions (for y+ ~ 50 mesh)
Both equilibrium and non-equilibrium wall functions were studied.
J. Baughn, M. Hoffman, R. Takahashi, and B. Launder (1984), Local Heat Transfer Downstream of an Abrupt Expansion
in a Circular Channel with Constant Wall Heat Flux, ASME J. Heat Transfer, Vol. 106, No. 4, pp. 789796.
2013 ANSYS, Inc.

March 28, 2013

10

Release 14.5

BL Heat Transfer Example Pipe

Expansion
Local Nusselt number compared to the Dittus-Boelter correlation (valid
for pipe flows).
Nu DB 0.023 Re0.8 Pr 0.4

Nu
Nu DB

Nu
Nu DB

x/ H

x/ H

2013 ANSYS, Inc.

March 28, 2013

11

Release 14.5

BL Heat Transfer 2D Turbine Blade

The suction side BL undergoes a laminar-to-turbulent transition.
Computed using several near-wall models and low Re models.

Two-layer zonal model

k models both with and without the transitional flow option

2013 ANSYS, Inc.

March 28, 2013

12

Release 14.5

BL Heat Transfer 2D Turbine Blade

St

h
U C p

Stanton
Number
(St)

Heat Transfer Prediction on the Suction Side

2013 ANSYS, Inc.

March 28, 2013

13

Release 14.5

Turbine Blade Heat Transfer with

Transition Models

VPI Turbine

Hybrid Mesh: 24,386 cells

Re = 23,000, Uin = 5.85 m/s, Tin = 20 C, Chord = 59.4 cm
Air with constant properties
Inlet turbulent intensity = 10%

Both models do a good job of predicting transition point and heat transfer
coefficient
2013 ANSYS, Inc.

March 28, 2013

14

Release 14.5

Example Impinging Jet

Relevant dimensionless parameters
Height-to-diameter ratio, H/D
Reynolds number, Re

T0

Prandtl number, Pr

Quantities analyzed
Surface heat transfer coefficient
h( x )

Tp T0

Nu x

h( x ) L
kf

Tp or

Nusselt number

2013 ANSYS, Inc.

March 28, 2013

15

Release 14.5

Characteristics of Impinging Jet Flow

Modeling challenge complex flow

Free jet turbulence

Stagnation point
Boundary layer

Strong streamline curvature

Transition (?)

Free jet

Wall jet

Stagnation zone
Boundary layer
and transition

2013 ANSYS, Inc.

March 28, 2013

16

Release 14.5

TKE Production at Stagnation Point

Physically, decreased production of turbulence is observed at the
stagnation point.
Two-equation models tend to overestimate TKE production at the
stagnation point

Standard k

Realizable k

RNG k

Can the production of turbulent kinetic energy be reduced?

2013 ANSYS, Inc.

March 28, 2013

17

Release 14.5

Impinging Jet Example

Turbulent kinetic energy transport equation:
Dk

Dt x j

ui
T k

T
x

x j

k
j

Diffusion

ui u j

j xi

Production

Dissipation

Modification of production term (Menter, 1992):

Pk T 2

Text user interface command is

define/models/viscous/turbulent-expert/

2013 ANSYS, Inc.

March 28, 2013

18

Release 14.5

Effect of Modified Production Term

-Based Production

Default Production

k Model

2013 ANSYS, Inc.

March 28, 2013

19

Release 14.5

Flow Calculations (y+ = 1)

The following RANS models were evaluated:
Standard k (SKE)
RNG k (RKE) Minimizes TKE at stagnation point.
Standard k (KW) Laminar/turbulent transition in boundary layer.

Modified k (KWW) Production of TKE based on rotation rate, .

V2F model Accounts for near-wall anisotropy by solving a transport equation
for (v')2

Flow characteristics
Prandtl number: Pr = 0.7
Reynolds number: Re = 23,000
Height-to-diameter ratio: H/D = 2.0 and 6.0

2013 ANSYS, Inc.

March 28, 2013

20

Release 14.5

Impinging Jet: Velocity Profiles

Results: H/D = 2, Re = 23,000

D
r/D = 1

KWW
r/D = 2

Mean velocity
profiles

V2F

RNG

2013 ANSYS, Inc.

March 28, 2013

21

Release 14.5

Results from Two-Equation Models

Results: H/D = 2, Re = 23,000
Nu*

TKE*
SKE
RNG

SKE
KWW
RNG

KWW

Nusselt Number
Re = 23,000

2013 ANSYS, Inc.

March 28, 2013

Turbulent Kinetic Energy

Re = 23,000

22

Release 14.5

Comparison of k and V2F models

Results: H/D = 2, Re = 23,000
Nu*

TKE*
V2F
V2F

KWW
KWW

Nusselt Number
Re = 23,000

2013 ANSYS, Inc.

March 28, 2013

Turbulent Kinetic Energy

Re = 23,000

23

Release 14.5

Results from Two-Equation Models

Results: H/D = 6, Re = 23.000
Nu*

Nu*
RNG
SKE

V2F
KWW

KWW

Nusselt Number
Re = 23,000
2013 ANSYS, Inc.

March 28, 2013

Nusselt Number
Re = 23,000
24

Release 14.5

Mixed Convection around a Wall-Mounted

Cylinder
Re = 40,000 (subcritical flow)

Laminar BL with turbulent wake

0.642 m

Bluff Body

Massive separation, vortex shedding

ReD = 40,000

g
2m

Turbulence model

600 W

SST k with y+ = 1
LES with dynamic Smagorinsky subgrid

12 m

model

3 million cell mesh

2.13 m

Courtesy CEA/EDF

Mixed convection (buoyancy is important)

Boussinesq approximation
Cylinder covered by a 5 mm thick steel layer
Fluid/Solid coupled thermal simulation

D >> d so use of shell conduction is appropriate

2013 ANSYS, Inc.

March 28, 2013

25

Release 14.5

Flow Regimes for Flow Past Cylinders

Re < 50

Laminar wake
No separation

Steady separation bubble

50 < Re < 5000 :

Von-Karman street (laminar BL)

Oscillating Krman vortex wake

5,000 < Re < 200,000:

Laminar BL prior to separation ( =

80). Sub-critical regime

Re > 200,000 Drag Crisis

Laminar boundary layer

with wide turbulent wake

Turbulent boundary layer prior to

Turbulent boundary
layer with narrow
turbulent wake

separation ( = 120).
2013 ANSYS, Inc.

March 28, 2013

26

Release 14.5

Subgrid Scale Viscosity Models

FLUENT 14.5 offers the
following subgrid scale models
to be used with LES:

Viscous Model

Smagorinsky model
WALE model

Dynamic Smagorinsky model

Dynamic subgrid kinetic energy
transport model
Wall Modeled LES (WMLES)

2013 ANSYS, Inc.

March 28, 2013

27

Release 14.5

Results Surface Temperature

Exp. From CEA/EDF
160

140

Temprature (C)

120

100

80

60

SST450

SST1250

SST 1750

Exp 450

Exp 1250

Exp 1750

LES 450

LES1250

LES 1750

40

20
-180

-160

-140

-120

-100

-80

-60

-40

-20

Angle (C)

2013 ANSYS, Inc.

March 28, 2013

28

Release 14.5

Visualisation d
l'chauffement de
2500

G=40cm

Results Wake (x = 0.4 m)

2250

2000

SST

LES

Exp.
1750

2750

12,0-14,0

Visualisation de
l'chauffement de

2500

G=40cm

1500 Hauteur (mm)

2250

10,0-12,0
8,0-10,0
6,0-8,0

1250
2000

4,0-6,0
1000

1750

12,0-14,0
2,0-4,0
10,0-12,0

750

1500 Hauteur (mm)

0,0-2,0

8,0-10,0
6,0-8,0

1250

H (mm)
500

-800

-500

-321
0
321
Largeur (mm)

-800

2013 ANSYS, Inc.

March 28, 2013

29

-500

500

250
800

1000

750

4,0-6,0
2,0-4,0
0,0-2,0

500

-321
0
321
Largeur (mm)

W
(mm)

500

250
800

Release 14.5

Results Wake (x = 0.5 m)

2750

Visualisation
l'chauffement
2500

G=50cm

Exp.

LES

SST

2250

2750

2000

Visual

l'chauffe
8,0-10,0
2500

1750

G=50cm

6,0-8,0
1500 Haute ur 2250
(mm)

4,0-6,0
1250

2000

1000

1750

2,0-4,0
0,0-2,0

8,0-10

6,0-8,0

750

1500 Hauteur (mm)

H (mm)

500

1250
-800

-500

-321
0
321
Large ur (mm)

500

250
800

2,0-4,0

1000

0,0-2,0

750

500

-800

-500

-321
0
321
Largeur (mm)

500

250
800

W (mm)
2013 ANSYS, Inc.

March 28, 2013

30

4,0-6,0

Release 14.5

2500

G=75cm

Results Wake (x = 0.75 m)

Visual
l'chauffe

2250

2000

SST

Exp.

LES

1750

2750

2500

6,0-8,0Visualisati
l'chauffemen
1500 Hauteur (mm) 4,0-6,0
2250

G=75cm

2000

1250

2,0-4,0
0,0-2,0

1750

6,0-8,0

1000

1500 Hauteur (mm) 4,0-6,0

1250

750

H (mm)

2,0-4,0
0,0-2,0

1000

500

750

500

-800

-500

-321
0
321
Largeur (mm)

-800

250
500
-500
-321 800
0
321

500

250
800

Largeur (mm)

W (mm)
2013 ANSYS, Inc.

March 28, 2013

31

Release 14.5

Results Wake (x = 1.5 m)

SST

LES

Exp.
2750

V
l'c

2500

G=1m50

2250

2000

1750

1500 Hauteur (mm)

H (mm)
1250

1000

750

500

-800

-500

-321
0
321
Largeur (mm)

500

250
800

W
(mm)
2013 ANSYS, Inc.

March 28, 2013

32

Release 14.5

3,0-4
2,0-3
1,0-2
0,0-1

Results Conclusions
Wall temperature comparable between RANS/LES
More accurate wake prediction with LES
CPU time required

RANS Days
LES Weeks
In this case fluid/solid thermal coupling and large difference between
characteristic time scales induce expensive unsteady calculations

2013 ANSYS, Inc.

March 28, 2013

33

Release 14.5

Large Eddy Simulation Applications

Compute unsteady temperature
field
Explicit representation of mixing

Accurate min/max fluctuations

Application examples
Thermal fatigue
Fluid-structure interaction (FSI)

2013 ANSYS, Inc.

March 28, 2013

34

Release 14.5

Appendix: Lecture 3
Forced Convection
14.5 Release

2013 ANSYS, Inc.

March 28, 2013

35

Release 14.5

Reynolds Analogy
U 0 , T0

Boundary Layer Equations:

Laminar

Transition

laminar

Turbulent

turbulent

~
2~
~
~

P
1

u
~
u ~ v ~ ~
x
y
x Re ~
x2

u~ ~
x ,0 v~ ~
x ,0 0
u~ ~
x , u / U

~
~
2~
T
T
1 T
u~ ~ v ~
x
y Re Pr ~
x2

~
~
T ~
x ,0 0 T ~
x , 1

Wall Fluxes:

2
Cf
Re L
2013 ANSYS, Inc.

March 28, 2013

u~
~
y

~
y 0

36

~
T
Nu ~
y

~
y 0

Release 14.5

Boundary Layers
Reynolds analogy

If dP/dx ~ 0, Pr ~ 1 (constant properties)

u~ ~ u~
1 2u~
~
u ~ v ~
x
y Re ~
x2

u~ ~
x , 0 v~ ~
x , 0 0
u~ ~
x , 1

~
~
2~

T
1

T
u~ ~ v~ ~
x
y Re Pr ~
x2

~
T ~
x , 0 0
~
T ~
x , 1

In dimensionless form, equations are of the same form. Thus, the

solutions for dimensionless velocity and dimensionless temperature
should be equivalent.
Cf
Re L
Cf
Nu
St
2
2
2013 ANSYS, Inc.

March 28, 2013

37

Release 14.5

Some Definitions

u
w
y
T
qw k
y

2
T lmix

2013 ANSYS, Inc.

March 28, 2013

U
y 0

qw
T
Cp U

y 0

u
y

lmix y

38

Release 14.5

Wall Functions
BL Momentum RANS equations

u 0 u 0 P 0 u
u
v

uv
x
y
x y y

u
uv constant
y
total

BC at the wall (y = 0):

Mixing length model

2013 ANSYS, Inc.

March 28, 2013

total

y 0

uv T
39

u
y

U 2
y 0

u
u
y2
y
y

Release 14.5

Boundary Layers
Viscous sublayer
0

uv U 2
y
total

u y

Turbulent region
0

uv U 2
y
total

2013 ANSYS, Inc.

March 28, 2013

40

1
ln y C

Release 14.5

Boundary Layers
Boundary layer energy equation
0

T
T
C p u
v
y
x

k
C p v T
y y

T
k
C p v T constant
y
qtotal

BC at the wall (y = 0):

Reynolds analogy:
2013 ANSYS, Inc.

March 28, 2013

qtotal

y 0

T
y

C p T U
y 0

T T y 2 u T
v T

PrT y PrT y y
41

Release 14.5

Boundary Layers
Viscous sublayer
0

T
C p v T C p T U
y
qtotal

T Pr y

Turbulent region
0

T
C p v T C p T U
y

PrT
ln y f Pr

qtotal

2013 ANSYS, Inc.

March 28, 2013

42

Release 14.5

Wall Functions in FLUENT

Non-equilibrium effect and pressure gradient effect
T C1/ 4 k 1/ 2 y

Use Prandtl-Komolgorov eddy-viscosity model

y
U

1
ln E y

C1/ 4 k P1/ 2 yP

U P C1/ 4 k P1/ 2
w /

Keep pressure gradient in boundary layer equations (partially cancel the

inertial terms)
~
U C1/ 4 k 1/ 2
U 2

1/ 4 1/ 2
1 C k
ln E

yv
yv
C1/ 4 k P1/ 2

y y yv yv2
1 dP yv
~
U U
ln

2 dx k yv k
2013 ANSYS, Inc.

March 28, 2013

43

Release 14.5

Wall Functions in FLUENT

Jayatilleke: Wide range of Prandtl number

Tp C P C1/ 4 k P1/ 2
q

2
1/ 4 1/ 2

Pr
U
C
kP
p

Pr y

2 q

1/ 4 1/ 2

C
1

kP
2
2
Pr ln E y P

Pr
U

Pr

Pr
U
t
t
P
t
c

2 q
k

for y yT

for y yT

Pr 3 / 4

Pr
1 1 0.28 exp 0.007

P 9.24
Prt
Prt

2013 ANSYS, Inc.

March 28, 2013

44

Release 14.5

Turbulent Thermal Boundary Layers

The wall laws are also functions of Prandtl number.
Viscous sublayer thickness defined as the intersection between viscous
and logarithmic law.
y yT f (Pr, PrT )

y+ ~ 10 for Momentum and for Pr = 1

20

60

Pr = 1

Pr = 7

16

40

T*

T*

12
8

20

4
0

0
0.1

10

100

0.1

1000

March 28, 2013

10

100

1000

y*

y*
2013 ANSYS, Inc.

45

Release 14.5