Aerodynamics and Heat Transfer for Airfoil-Endwall Junctures in Gas Turbine Engines

Stephen LynchKaren TholeVirginia Tech Experimental and

Computational Convection LaboratoryMechanical Engineering DepartmentVirginia Tech

fillet

Motivation, past studies, and research objectives

Experimental facilities and techniques

Discussion of results

This talk evaluates endwall heat transfer and shear stress with filleted vanes in gas turbine engines

Horseshoe vortex

Secondary flows augment wall heat transfer and increase aerodynamic losses for a gas turbine

Secondary flow model presented by Langston (1980)

Experimental measurements,Kang, et al. (1999)

Secondary flows augment wall heat transfer and increase aerodynamic losses for a gas turbine

Secondary flow model presented by Langston (1980)

0.0

0.1

0.2

0.3

0.4

0.5

0.00 0.05 0.10 0.15 0.20 0.25

Z/S

y/P

U/Uin = 1

Experimental measurements,Kang, et al. (1999)

Experimental measurements,Kang, et al. (1999)

Horseshoe vortex

Passage vortex

Various leading edge/endwall junction geometries have been investigated with promising results

Zess and Thole (2001)

Shih and Lin (2002)

Sauer, Müller, andVogeler (2000)

Various leading edge/endwall junction geometries have been investigated with promising results

Zess and Thole (2001)

Becz, Majewski, and Langston (2003)

Shih and Lin (2002)

Sauer, Müller, andVogeler (2000) Lethander, Thole,

and Zess (2003)

Various leading edge/endwall junction geometries have been investigated with promising results

Zess and Thole (2001)

Becz, Majewski, and Langston (2003)

Shih and Lin (2002)

Sauer, Müller, andVogeler (2000)

Mahmood, Gustafson, and Acharya (2005)

Lethander, Thole, and Zess (2003)

Han and Goldstein (2004)

The goal was to find how fillet reduce aerodynamic losses and heat transfer for gas turbine vanes

Project Objectives1) Evaluate endwall heat transfer and shear stress with an unfilleted vane2) Evaluate endwall heat transfer and shear stress with a filleted vane

We conducted this research in a closed-loop, low speed wind tunnel

PW6000 turbine nozzle guide vane

Geometry:

9X engineScale:

59.4 cm (23.4”)Chord (C):

45.7 cm (18.0”)Pitch (P):

55.0 cm (21.7”)Span (S):

Linear cascade test section

Primary heat exchanger

Axial fan

To measure temperatures on a constant heat flux surface, we used infrared thermography

Spatial transformation of fillet surface

Oil film interferometry (OFI) can provide measurements of wall shear stress

xΔ

Cf,y

Cf,x

Airflow direction

~10 mm

3.8 cm (1.5”)

Oil film interferometry (OFI) can provide measurements of wall shear stress

xΔ

Interference between light rays (out of phase)

Reflective surface (nickel foil, 0.05 mm thick)

Dow Corning 200 silicone oil

Airflow direction

Low-pressure sodium vapor lamp (monochromatic)

Cf,y

Cf,x

Airflow direction

~10 mm

3.8 cm (1.5”)

iθ

hoil

~600 nm

πφΔ

2=,x

OFI was benchmarked for channel flow and implemented in the vane cascade

0

0.002

0.004

0.006

0.008

2.5 104 3 104 3.5 104 4 104 4.5 104

Colebrook CorrelationOil Film Interferometry

ReDh

2m

wU2

1=fρτ

OFI was benchmarked for channel flow and implemented in the vane cascade

0

0.002

0.004

0.006

0.008

2.5 104 3 104 3.5 104 4 104 4.5 104

Colebrook CorrelationOil Film Interferometry

ReDh

2m

wU2

1=fρτ

OFI was benchmarked for channel flow and implemented in the vane cascade

0

0.002

0.004

0.006

0.008

2.5 104 3 104 3.5 104 4 104 4.5 104

Colebrook CorrelationOil Film Interferometry

ReDh

2m

wU2

1=fρτ

Adding a fillet to the airfoil-endwall junction changes the endwall heat transfer

∞UCh

=Stp

∞ ρ

Heat transfer coefficients are lowered on the pressure side of the passage

s (m)

∞UCh

=Stp

∞ ρ

0 0.1 0.2 0.3 0.4 0.5 0.6

No filletLinear fillet

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

Heat transfer coefficients are lowered on the pressure side of the passage

s (m)

∞UCh

=Stp

∞ ρ

0 0.1 0.2 0.3 0.4 0.5

No filletLinear fillet

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

Heat transfer coefficients are lowered on the pressure side of the passage

s (m)

∞UCh

=Stp

∞ ρ

0 0.1 0.2 0.3 0.4 0.5

No filletLinear fillet

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

The OFI method shows unique endwall flow patterns caused by secondary flow

Cf = 0.150Cf = 0.020

No fillet

Linear fillet

Cf = 0.020

Adding a linear fillet increases skin friction, but reduces flow turning at the passage exit

No fillet

Linear fillet

Cf = 0.020 Cf = 0.150

Adding a linear fillet increases skin friction, but reduces flow turning at the passage exit

0 0.1 0.2 0.3 0.4 0.5

No filletLinear fillet

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

2w

f U21C

∞ρτ

=

s (m)

The linear fillet increases wall shear stress through the center of the passage

2w

f U21C

∞ρτ

=∞UC

h=St

p∞ ρ

Wall shear stress and heat transfer coefficients do not appear to be related for a turbine vane

No fillet No fillet

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0 0.1 0.2 0.3 0.4 0.5

No Fillet, St*Pr(2/3)

Linear Fillet, St*Pr(2/3)

No Fillet, Cf /2

Linear Fillet, Cf /2

s (m)

( )3/2f Pr*St , 2C

The Reynolds analogy is not valid along the center of the passage

In conclusion, adding a linear fillet reduces endwall surface area

Adding a linear fillet marginally changes the heat transfer distribution

Wall shear increases with the fillet, but endwall flow turning is reduced

Wall shear stress and heat transfer coefficients do not appear to be related in the vane passage

Seawolf Submarine Class

Questions?