AIAA JOURNAL

Vol. 47, No. 12, December 2009

Correlation-Based Transition Modeling for Unstructured

Parallelized Computational Fluid Dynamics Codes

Robin B. Langtry∗
The Boeing Company, Seattle, Washington 98124-2207
and
Florian R. Menter†
ANSYS Germany, 83624 Otterfing, Germany
DOI: 10.2514/1.42362
A new correlation-based transition model has been developed, which is built strictly on local variables. As a result,
the transition model is compatible with modern computational fluid dynamics techniques such as unstructured grids
and massively parallel execution. The model is based on two transport equations, one for intermittency and one for a
transition onset criterion in terms of momentum-thickness Reynolds number. A number of validation papers have
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

been published on the basic formulation of the model. However, until now the full model correlations have not been
published. The main goal of the present paper is to publish the full model and release it to the research community so
that it can continue to be further validated and possibly extended or improved. Included in this paper are a number of
test cases that can be used to validate the implementation of the model in a given computational fluid dynamics code.
The authors believe that the current formulation is a significant step forward in engineering transition modeling, as it
allows the combination of transition correlations with general-purpose computational fluid dynamics codes. There is
a strong potential that the model will allow the first-order effects of transition to be included in everyday industrial
computational fluid dynamics simulations.

Nomenclature
= wall shear stress

Cf = skin friction coefficient, 2

=
0:5Uref 
= absolute value of vorticity,
2
ij
ij 1=2
k = turbulent kinetic energy
ij = vorticity tensor, 0:5
@ui =@xj  @uj =@xi 
Rex = Reynolds number, LUref = ! = specific turbulence dissipation rate
Re = momentum-thickness Reynolds number, U0 =
Ret = transition onset momentum-thickness Reynolds Subscripts
number (based on freestream conditions), t U0 =
~ t s = streamline
Re = local transition onset momentum-thickness Reynolds
t = transition onset
number (obtained from a transport equation)
RT = viscosity ratio
Ry = wall-distance-based turbulent Reynolds number
Rv = vorticity Reynolds number I. Introduction
S = absolute value of strain rate,
2Sij Sij 1=2
Sij
Tu
=
=
strain-rate tensor, 0:5
@ui =@xj  @uj =@xi )
turbulence intensity, 100
2k=31=2 =U
I N THE past few decades, a significant amount of progress has
been made in the development of reliable turbulence models that
can accurately simulate a wide range of fully turbulent engineering
U = local velocity flows. The efforts by different groups have resulted in a spectrum of
Uo = local freestream velocity models that can be used in many different applications, while bal-
Uref = inlet reference velocity ancing the accuracy requirements and the computational resources
u0 = local fluctuating streamwise velocity available to a computational fluid dynamics (CFD) user. However,
x=C = axial distance over axial chord the important effect of laminar-turbulent transition is not included in
y = distance to nearest wall the majority of today’s engineering CFD simulations. The reason for
y = distance in wall coordinates, y
= this is that transition modeling does not offer the same wide spectrum
 = boundary-layer thickness of CFD-compatible model formulations that are currently available
 = momentum thickness for turbulent flows, even though a large body of publications is
 = pressure gradient parameter,
2 =
dU=ds available on the subject. There are several reasons for this unsatis-
 = molecular viscosity factory situation.
t = eddy viscosity The first is that transition occurs through different mechanisms in
 = density different applications. In aerodynamic flows, transition is typically
the result of a flow instability (Tollmien–Schlichting waves or, in the
case of highly swept wings, crossflow instability), where the resul-
Received 24 November 2008; revision received 9 June 2009; accepted for ting exponential growth of two-dimensional waves eventually results
publication 30 June 2009. Copyright © 2009 by the American Institute of in a nonlinear breakdown to turbulence. Transition occurring due to
Aeronautics and Astronautics, Inc. All rights reserved. Copies of this paper Tollmien–Schlichting waves is often referred to as natural transition
may be made for personal or internal use, on condition that the copier pay the [1]. In turbomachinery applications, the main transition mechanism
$10.00 per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood
is bypass transition [2] imposed on the boundary layer by high levels
Drive, Danvers, MA 01923; include the code 0001-1452/09 and $10.00 in
correspondence with the CCC.
of turbulence in the freestream. The high freestream turbulence levels
∗
Senior Engineer, Enabling Technology and Research, P.O. Box 3703, are for instance generated by upstream blade rows. Another impor-
Mail Code 67-LF; robin.b.langtry@boeing.com. tant transition mechanism is separation-induced transition [3], where
† a laminar boundary layer separates under the influence of a pressure
Development Manager, Fluids Business Unit, Staudenfeldweg 12;
florian.menter@ansys.com. gradient and transition develops within the separated shear layer
2894
 LANGTRY AND MENTER 2895

(which may or may not reattach). As well, a turbulent boundary layer that none of these methods can meet all of the aforementioned
can relaminarize under the influence of a strong favorable pressure requirements.
gradient [4]. Although the importance of transition phenomena for The only transition models that have historically been compatible
aerodynamic and heat transfer simulations is widely accepted, it is with modern CFD methods are the low-Re models [10,11]. However,
difficult to include all of these effects in a single model. they typically suffer from a close interaction with the transition
The second complication arises from the fact that conventional capability and the viscous sublayer modeling, and this can prevent an
Reynolds-averaged Navier–Stokes (RANS) procedures do not lend independent calibration of both phenomena [12,13]. At best, the low-
themselves easily to the description of transitional flows, where both Re models can only be expected to simulate bypass transition which
linear and nonlinear effects are relevant. RANS averaging eliminates is dominated by diffusion effects from the freestream. This is because
the effects of linear disturbance growth and is therefore difficult to the standard low-Re models rely exclusively on the ability of the wall
apply to the transition process. Although methods based on the damping terms to capture the effects of transition. Realistically, it
stability equations such as the en method of Smith and Gamberoni [5] would be very surprising if these models that were calibrated for
and van Ingen [6] avoids this limitation, they are not compatible with viscous sublayer damping could faithfully reproduce the physics of
general-purpose CFD methods as typically applied in complex transitional flows. It should be noted that there are several low-Re
geometries. The reason is that these methods require a priori models where transition prediction was considered specifically
knowledge of the geometry and the grid topology. In addition, they during the model calibration [14–16]. However, these model formu-
involve numerous nonlocal operations (e.g., tracking the disturbance lations still exhibit a close connection between the sublayer behavior
growth along each streamline) that are difficult to implement into and the transition calibration. Recalibration of one functionality also
today’s CFD methods [7]. This is not to argue against the stability
changes the performance of the other. It is therefore not possible to
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

approaches, as they are an essential part of the desired “spectrum” of

introduce additional experimental information without a substantial
transition models required for the vastly different application areas
reformulation of the entire model.
and accuracy requirements. However, much like in turbulence
The engineering alternative to low-Re transition models are empir-
modeling, it is important to develop engineering models that can be
applied in day-to-day operations by design engineers on complicated ical correlations such as those of [17–19]. They typically correlate
3-D geometries. the transition momentum-thickness Reynolds number to local
It should be noted that, at least for 2-D flows, the efforts of various freestream conditions such as the turbulence intensity and pressure
groups have resulted in a number of engineering design tools gradient. These models are relatively easy to calibrate and are often
intended to model transition for very specific applications. The most sufficiently accurate to capture the major effects of transition. In
notable efforts are those of Drela and Giles [8] who developed the X- addition, correlations can be developed for the different transition
FOIL code which can be used for modeling transition on 2-D airfoils, mechanisms, ranging from bypass to natural transition as well as
and the MISES code of Youngren and Drela [9], which is used for crossflow instability or roughness. The main shortcoming of these
modeling transition on 2-D turbomachinery blade rows. Both of models lies in their inherently nonlocal formulation. They typically
these codes use a viscous–inviscid coupling approach which allows require information on the integral thickness of the boundary layer
the classical boundary-layer formulation tools to be used. Transition and the state of the flow outside the boundary layer. Although these
prediction is accomplished using either an en method or an empirical models have been used successfully in special-purpose turbomachi-
correlation, and both of these codes are used widely in their res- nery codes, the nonlocal operations involved with evaluating the
pective design communities. A 3-D wing or blade design is per- boundary-layer momentum thickness and determining the free-
formed by stacking the 2-D profiles (with the basic assumption that stream conditions have precluded their implementation into general-
spanwise flow is negligible) to create the geometry at which point a purpose CFD codes.
3-D CFD analysis is performed. Transition simulations based on linear stability analysis such as the
Closer inspection shows that hardly any of the current transition en method are the lowest closure level available where the actual
models are CFD compatible. Most formulations suffer from nonlocal instability of the flow is simulated. In the simpler models previously
operations that cannot be carried out (with reasonable effort) in described, the physics are introduced through the calibration of the
general-purpose CFD codes. This is because modern CFD codes use model constants. However, even the en method is not free from
mixed elements and massive parallel execution and do not provide empiricism. This is because the transition n factor is not universal and
the infrastructure for computing integral boundary-layer parameters depends on the wind-tunnel freestream/acoustic environment and
or allow the integration of quantities along the direction of external also the smoothness of the test model surface. The main obstacle to
streamlines. Even if structured boundary-layer grids are used (typi- the use of the en model is that the required infrastructure needed to
cally hexahedra), the codes are based on data structures for unstruc- apply the model is very complicated. The stability analysis is typi-
tured meshes. The information on a body-normal grid direction is cally based on velocity profiles obtained from highly resolved
therefore not easily available. In addition, most industrial CFD boundary-layer codes that must be coupled to the pressure distri-
simulations are carried out on parallel computers using a domain bution of a RANS CFD code [7]. The output of the boundary-layer
decomposition methodology. This means in the most general case method is then transferred to a stability method, which then provides
that boundary layers can be split and computed on different proces- information back to the turbulence model in the RANS solver. The
sors, prohibiting any search or integration algorithms. Consequently, complexity of this setup is mainly justified for special applications
the main requirements for a fully CFD-compatible transition model where the flow is designed to remain close to the stability limit for
are as follows: drag reduction, such as laminar wing design.
1) Allow the calibrated prediction of the onset and the length of Large eddy simulation and direct numerical simulations are sui-
transition. table tools for transition prediction [20], although the proper
2) Allow the inclusion of different transition mechanisms. specification of the external disturbance level and structure poses
3) Be formulated locally (no search or line-integration operations). substantial challenges. Unfortunately, these methods are far too
4) Avoid multiple solutions (same solution for initially laminar or costly for engineering applications. They are currently used mainly
turbulent boundary layer). as research tools and substitutes for controlled experiments.
5) Do not affect the underlying turbulence model in fully turbulent Despite its complexity, transition should not be viewed as outside
regimes. the range of RANS methods. In many applications, transition is
6) Allow a robust integration down to the wall with similar enforced within a narrow area of the flow due to geometric features
convergence as the underlying turbulence model. (e.g., steps or gaps), pressure gradients, and/or flow separation. Even
7) Be formulated independent of the coordinate system. relatively simple models can capture these effects with sufficient
8) Be applicable to three-dimensional boundary layers.Consider- engineering accuracy. The challenge to a proper engineering model is
ing the main classes of engineering transition models (stability therefore mainly in the formulation of a model that can be imple-
analysis, correlation-based models, low-Re models), one finds mented into a general RANS environment.
 2896 LANGTRY AND MENTER

In this paper, a novel approach to simulating laminar to turbu-

lent transition is described that can be implemented into a general
RANS environment. The central idea behind the new approach is
that Van Driest and Blumer’s [21] vorticity Reynolds number
concept can be used to provide a link between the transition onset
Reynolds number from an empirical correlation and the local
boundary-layer quantities. As a result, the model avoids the need
to integrate the boundary-layer velocity profile to determine the
onset of transition and this idea was first proposed by Menter
et al. [22].
The vorticity or alternatively the strain-rate Reynolds number
which is used in the present model is defined as follows:



y2

@u

y2 Fig. 2 Relative error between the maximum value of vorticity Reynolds
Rev   S (1)

@y
 number Rev and the momentum-thickness Reynolds number Re
as a
function of boundary-layer shape factor H.
where y is the distance from the nearest wall. Because the vorticity
Reynolds number depends only on density, viscosity, wall distance,
and the vorticity (or shear strain rate), it is a local property and can be
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

easily computed at each grid point in an unstructured, parallel For strong adverse pressure gradients, the difference between the
Navier–Stokes code. momentum-thickness and vorticity Reynolds number can become
A scaled profile of the vorticity Reynolds number is shown in significant, particularly near separation (H  3:5). However, the
Fig. 1 for a Blasius boundary layer. The scaling is chosen to have a trend with experiments is that adverse pressure gradients reduce the
maximum of one inside the boundary layer. This is achieved transition momentum-thickness Reynolds number. In practice, if a
by dividing the Blasius velocity profile by the corresponding constant transition momentum-thickness Reynolds number is
momentum-thickness Reynolds number and a constant of 2.193. In specified, the transition model is not very sensitive to adverse
other words, the maximum of the profile is proportional to the pressure gradients and an empirical correlation such as that of Abu-
momentum-thickness Reynolds number and can therefore be related Ghannam and Shaw [17] is necessary to predict adverse pressure
to the transition correlations [22] as follows: gradient transition accurately. In fact, the increase in vorticity
Reynolds number with increasing shape factor can actually be used
max
Rev  to predict separation-induced transition. This is one of the main
Re  (2) advantages of the present approach because the standard definition
2:193 of momentum-thickness Reynolds number is not suitable in
separated flows.
Based on this observation, a general framework can be built, which The function Rev can be used on physical reasoning, by arguing
can serve as a local environment for correlation-based transition that the combination of y2 S is responsible for the growth of
models. disturbances inside the boundary layer, whereas   = is res-
When the laminar boundary layer is subjected to strong pressure ponsible for their damping. As y2 S grows with the thickness of the
gradients, the relationship between momentum-thickness and boundary layer and  stays constant, transition will take place once a
vorticity Reynolds number described by Eq. (2) changes due to the critical value of Rev is reached. The connection between the growth
change in the shape of the profile. The relative difference between of disturbances and the function Rev was shown by Van Driest and
momentum-thickness and vorticity Reynolds number, as a function Blumer [21] in comparison with experimental data. As well, Langtry
of shape factor H, is shown in Fig. 2. For moderate pressure gradients and Sjolander [15] found that the location in the boundary layer
(2:3 < H < 2:9), the difference between the actual momentum- where Rev was largest correspondeds surprisingly well to the loca-
thickness Reynolds number and the maximum of the vorticity Rey- tion where the peak growth of disturbances was occurring, at least for
nolds number is less than 10%. Based on boundary-layer analysis, a bypass transition. The models proposed by Langtry and Sjolander
shape factor of 2.3 corresponds to a pressure gradient parameter  of [15] and Walters and Leylek [16], use Rev in physics-based
approximately 0.06. Because the majority of experimental data on arguments based on these observations of disturbance growth in the
transition in favorable pressure gradients falls within that range (see, boundary layer during bypass transition. These models appear
for example, [17]), the relative error between momentum-thickness superior to conventional low-Re models, as they implicitly contain
and vorticity Reynolds number is not of great concern under those information of the thickness of the boundary layer. Nevertheless, the
conditions. close integration of viscous sublayer damping and transition
prediction does not easily allow for an independent calibration of
both submodels.
In the present approach first described in [22–25], the main idea is
to use a combination of the strain-rate Reynolds number with
experimental transition correlations using standard transport equa-
tions. Because of the separation of viscous sublayer damping and
transition prediction, the new method has provided the flexibility for
introducing additional transition effects with relative ease. Currently,
the main missing extensions are crossflow instabilities and high-
speed flow correlations and these do not pose any significant obsta-
cles. The concept of linking the transition model with experimental
data has proven to be an essential strength of the model and this is
difficult to achieve with closures based on a physical modeling of
these diverse phenomena.
The present transition model is built on a transport equation for
intermittency, which can be used to trigger transition locally. In
addition to the transport equation for the intermittency, a second
Fig. 1 Scaled vorticity Reynolds number Rev profile in a Blasius transport equation is solved for the transition onset momentum-
boundary layer. thickness Reynolds number. This is required to capture the non-
 LANGTRY AND MENTER 2897

local influence of the turbulence intensity, which changes due to the The transport equation for the intermittency  reads
decay of the turbulence kinetic energy in the freestream, as well as   
due to changes in the freestream velocity outside the boundary layer. @
 @
Uj  @  @
  P  E   t (3)
This second transport equation is an essential part of the model as it @t @xj @xj f @xj
ties the empirical correlation to the onset criteria in the intermittency
equation. Therefore, it allows the model to be used in general The transition sources are defined as follows:
geometries and over multiple airfoils, without additional information
P1  Flength ca1 SFonset 0:5
1  ce1  (4)
on the geometry. The intermittency function is coupled with the shear
stress transport (SST) k-!-based turbulence model [26]. It is used to
where S is the strain-rate magnitude. Flength is an empirical cor-
turn on the production term of the turbulent kinetic energy
relation that controls the length of the transition region, and FOnset
downstream of the transition point based on the relation between
controls the transition onset location. Both are dimensionless
transition momentum-thickness and strain-rate Reynolds number. As
functions that are used to control the intermittency equation in the
the strain-rate Reynolds number is a local property, the present
boundary layer. The destruction/relaminarization source is defined as
formulation avoids another very severe shortcoming of the
follows:
correlation-based models, namely their limitation to 2-D flows. It
therefore allows the simulation of transition in 3-D flows originating E  ca2 
Fturb
ce2   1 (5)
from different walls. The formulation of the intermittency has also
been extended to account for the rapid onset of transition caused by where
is the vorticity magnitude. The transition onset is controlled
separation of the laminar boundary layer [Eq. (17)]. In addition, the by the following functions:
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

model can be fully calibrated with internal or proprietary transition

onset and transition length correlations. The correlations can also be y2 S
ReV  (6)
extended to flows with rough walls or to flows with crossflow 
instability. It should be stressed that the proposed transport equations
do not attempt to model the physics of the transition process (unlike, Rev
e.g., turbulence models), but form a framework for the implemen- Fonset1  (7)
tation of correlation-based models into general-purpose CFD 2:193  Rec
methods. To distinguish the present concept from physics-based
transition modeling, it is called local correlation-based transition
modeling. 4
Fonset2  min
max
Fonset1 ; Fonset1 ; 2:0 (8)

II. Transition Model Formulation k

RT  (9)
The present transition model formulation is described very briefly !
for completeness; a detailed description of the model and its
development can be found in Langtry [27]. It should be noted that a
  3 
few changes have been made to the model since it was first published RT
[23] to improve the predictions of natural transition. These include Fonset3  max 1  ;0 (10)
2:5
the following:
1) A new transition onset correlation results in improved
predictions for both natural and bypass transition.
2) A modification to the separation-induced transition modifi- Fonset  max
Fonset2  Fonset3 ; 0 (11)
cation prevents it from causing early transition near the separation
point. Rec is the critical Reynolds number where the intermittency first
3) Some adjustments of the model coefficients better account for starts to increase in the boundary layer. This occurs upstream of the
flow history effects on the transition onset location. transition Reynolds number Re ~ t, and the difference between the two
It was expected that different groups will make numerous must be obtained from an empirical correlation. Both the Flength and
improvements to the model and, consequently, a naming convention Rec correlations are functions of Re~ t .
was introduced in [23] to keep track of the various model versions. Based on the T3B, T3A, T3A-, and the Schubauer and Klebanof
The basic model framework (transport equations without any test cases [28], a correlation for Flength based on Ret from an
correlations) was called the -Re transition model. The version empirical correlation is defined as

8
> 1 4 ~ 6 ~ 2 ~ t < 400
> 398:189  10 
119:270  10 Ret 
132:567  10 Ret ;
>
<
Re
2 ~ 5 ~ 2 ~ 3t ; ~ t < 596
263:404 
123:939  10 Ret 
194:548  10 Ret 
101:695  108 Re 400 Re
Flength  (12)
>
> ~ 4
0:5 
Ret  596:0  3:0  10 ; 596 Re ~ t < 1200
>
: 0:3188; 1200 Re ~ t

number given in [23] was called CFX-v-1.0. Based on this naming In certain cases, such as transition at higher Reynolds numbers, the
convention, the present model with the preceding modifications will ~ t scalar will often decrease to very small values in the boundary
Re
be referred to as the -Re model, CFX-v-1.1. The present transition ~ t , this can
layer shortly after transition. Because Flength is based on Re
model is briefly summarized in the following pages. result in a local increase in the source term for the intermittency
 2898 LANGTRY AND MENTER

equation, which in turn can show up as a sharp increase in the skin Outside the boundary layer, the source term Pt is designed to force
friction. The skin friction does eventually return back to the fully ~ t to match the local value of Ret calculated
the transported scalar Re
turbulent value, however, this effect is unphysical. It appears to be from the empirical correlation [Eqs. (35) and (36)]. The source term
caused by a sharp change in the y in the viscous sublayer where the is defined as follows:
intermittency decreases back to its minimum value due to the 
destruction term [Eq. (5)]. The effect can be eliminated by forcing Pt  ct
Ret  Re ~ t 
1:0  Ft  (22)
Flength to always be equal to its maximum value (in this case 40.0) in t
the viscous sublayer. The modification for doing this is shown next.
The modification does not appear to have any effect on the predicted 500
t (23)
transition length. An added benefit is that, at higher Reynolds num- U2
bers, the model now appears to predict the skin friction overshoot
measured by experiments: where t is a time scale, which is present for dimensional reasons. The
time scale was determined based on dimensional analysis with the
R! 2 main criteria being that it had to scale with the convective and
Fsublayer  e
0:4  (13)
diffusive terms in the transport equation. The blending function Ft is
used to turn off the source term in the boundary layer and allow the
y2 ! ~ t to diffuse in from the freestream. Ft is equal
transported scalar Re
R!  (14)
500 to zero in the freestream and one in the boundary layer. The Ft
blending function is defined as follows:
    
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

Flength  Flength
1  Fsublayer   40:0  Fsublayer (15) y 4   1=ce2 2

Ft  min max Fwake  e
 ; 1:0  ; 1:0
1:0  1=ce2
~ t is defined as follows:
The correlation between Rec and Re (24)

8
>
> ~ t 
396:035  102 
120:656  104 Re~ t 
868:230  106 Re
~ 2t
< Re
Rec  9 ~ 3 12 ~ 4

696:506  10 Ret 
174:105  10 Ret ; ~ t 1870
Re (16)
>
>
: Re
~ t 
593:11 
Re
~ t  1870:0  0:482; ~ t > 1870
Re

The constants for the intermittency equation are ~ t 

Re 15 50
y
BL  ; BL   ;   BL (25)
ce1  1:0; ca1  2:0 ce2  50; ca2  0:06;  1:0 U 2 BL U
f

The modification to the intermittency for predicting separation-

induced transition is
!y2 Re! 2
      Re!  ; Fwake  e
1E5 (26)
Rev 
sep  min s1 max 0;  1 Freattach ; 2 Ft (17)
3:235Rec
The Fwake function ensures that the blending function is not active in
RT 4
the wake regions downstream of an airfoil/blade.
Freattach  e
20  (18) ~ t equation are
The model constants for the Re

ct  0:03; t  2:0 (27)

eff  max
; sep  (19)
The boundary condition for Re ~ t at a wall is zero flux. The boundary
~ t at an inlet should be calculated from the empirical
condition for Re
s1  2 (20) correlation [Eqs. (35) and (36)] based on the inlet turbulence
intensity.
The model constants in Eq. (17) have been adjusted from those of The empirical correlation for transition onset is based on the
Menter et al. [23] and Langtry et al. [24] to improve the predictions of following parameters:
separated flow transition. See Langtry [27] for a detailed discussion
of the changes to the model from the Menter et al. [23] and Langtry 2 dU
et al. [24] versions. The main difference is the constant that controls   (28)
 ds
the relation between Rev and Rec was changed from 2.193, its value
for a Blasius boundary layer, to 3.235, the value at a separation point
p
where the shape factor H is 3.5 (see Fig. 2). The boundary condition 2k=3
for  at a wall is zero normal flux, while for an inlet  is equal to 1.0. Tu  100 (29)
U
An inlet  equal to 1.0 is necessary to preserve the original turbulence
model’s freestream turbulence decay rate. Where dU=ds is the acceleration along the streamwise direction and
The transport equation for the transition momentum-thickness can be computed by taking the derivative of the velocity U in the x, y,
Reynolds number Re ~ t reads and z directions and then summing the contribution of these
  derivatives along the streamwise flow direction:
@
Re ~ t  @
Uj Re ~ t  @ ~ t
@Re
  Pt 
    (21)
@xj t t 1
@t @xj @xj U 
u2  v2  w2 2 (30)
 LANGTRY AND MENTER 2899

 
dU 1 2 1 du dv dw A minimum turbulence intensity of 0.027% results in a transition

u  v2  w2 2  2u  2v  2w (31) momentum-thickness Reynolds number of 1450, which is the largest
dx 2 dx dx dx
experimentally observed flat-plate transition Reynolds number based
on the Sinclair and Wells [29] data. For cases where larger transition
  Reynolds number are believed to occur (e.g., aircraft in flight) this
dU 1 2 1 du dv dw limiter may need to be adjusted downward.

u  v2  w2 2  2u  2v  2w (32)
dy 2 dy dy dy The empirical correlation is used only in the source term [Eq. (22)]
of the transport equation for the transition onset momentum-
thickness Reynolds number. Equations. (35–38) must be solved
  iteratively because the momentum thickness t is present in the left-
dU 1 2 1 du dv dw hand side of the equation and also in the right-hand side in the

u  v2  w2 2  2u  2v  2w (33)
dz 2 dz dz dz pressure gradient parameter . In the present work, an initial guess
for the local value of t was obtained based on the zero pressure
gradient solution of Eqs. (35) and (36) and the local values of U, ,
  and . With this initial guess, Eqs. (35–38) were solved by iterating
dU dU dU dU

u=U 
v=U 
w=U (34) on the value of t and convergence was obtained in less then 10
ds dx dy dz iterations using a shooting point method.
The transition model interacts with the SST turbulence model [26]
The use of the streamline direction is not Galilean invariant.
However, this deficiency is inherent to all correlation-based models, as follows:
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

as their main variable, the turbulence intensity, is already based on the   

@ @ @ @k
local freestream velocity and does therefore violate Galilean invari-
k 
uj k  P~ k  D~ k    k t (39)
@t @xj @xj @xj
ance. This is not problematic, as the correlations are defined with
respect to a wall boundary layer and all velocities are therefore
relative to the wall. Nevertheless, multiple moving walls in one P~ k  eff Pk ; D~ k  min
max
eff ; 0:1; 1:0Dk (40)
domain will likely require additional information.
The empirical correlation has been modified from [23] to improve
p
the predictions of natural transition. The predicted transition Rey- y k Ry 8
nolds number as a function of turbulence intensity is shown in Fig. 3. Ry  ; F3  e
120 ; F1  max
F1orig ; F3  (41)

For pressure gradient flows, the model predictions are similar to the
Abu-Ghannam and Shaw [17] correlation. The empirical correlation where Pk and Dk are the original production and destruction terms for
is defined as follows: the SST model and F1orig is the original SST blending function. Note
  that the production term in the ! equation is not modified. The
0:2196
Ret  1173:51  589:428Tu  F
 ; Tu 1:3 (35) rationale behind the preceding model formulation is given in detail
Tu2 in [23].
To capture the laminar and transitional boundary layers correctly,
Ret  331:50Tu  0:56580:671 F
 ; Tu > 1:3 (36) the grid must have a y of approximately one at the first grid point off
the wall. If the y is too large (i.e., >5), then the transition onset
location moves upstream with increasing y . All simulations have
Tu 1:5
F
   1  12:986  123:662  405:6893 e1:5 ;  0 been performed with CFX-5 using a bounded second-order upwind
(37) biased discretization for the mean flow, turbulence, and transition
equations.
Tu
F
   1  0:2751  e35:0  e 0:5  ;  > 0 (38)
III. Test Cases
For numerical robustness, the acceleration parameters, the turbul- The remaining part of the paper will give an overview of some of
ence intensity, and the empirical correlation should be limited as the public-domain test cases that have been computed with the model
follows: described in the previous section. This naturally requires a compact
 0:1  0:1 Tu 0:027 Ret 20 representation of the simulations. Most of the cases are described in
far more detail in [27], including grid refinement and sensitivity
studies.

A. Flat-Plate Test Cases

The flat-plate test cases that were used to calibrate the model are
the European Research Community on Flow Turbulence and Com-
bustion (ERCOFTAC) T3 series of flat-plate experiments [12,13]
and the Schubauer and Klebanoff [28] flat-plate experiment, all of
which are commonly used as benchmarks for transition models. Also
included is a test case in which the boundary layer experiences a
strong favorable pressure gradient that causes it to relaminarize [30].
The inlet conditions for these test cases are summarized in Table 1.
The three cases T3A-, T3A, and T3B have zero pressure gradients
with different freestream turbulence intensity (FSTI) levels corres-
ponding to transition in the bypass regime. The Schubauer and
Klebanof (S&K) test case [28] has a low freestream turbulence
intensity and corresponds to natural transition. Figure 4 shows the
comparison of the model prediction with experimental data for these
cases. It also gives the corresponding FSTI values. In all simulations,
Fig. 3 Transition onset momentum-thickness Reynolds number Re
t the inlet turbulence levels were specified to match the experimental
predicted by the new correlation as a function of turbulence intensity Tu turbulence intensity and its decay rate. This was done by fixing the
for a flat plate with zero pressure gradient. inlet turbulence intensity and, via trial and error, adjusting the inlet
 2900 LANGTRY AND MENTER

Table 1 Inlet condition for the flat-plate test cases at 0.04 m upstream of plate leading edge
Case Inlet velocity, m=s Turbulence intensity, % t = Density, kg=m3 Dynamic viscosity, kg=ms
Inlet/leading edge value
T3A 5.4 3.3 12.0 1.2 1:8 105
T3B 9.4 6.5 100.0 1.2 1:8 105
T3A- 19.8 0.874 8.72 1.2 1:8 105
Schubauer and Klebanof [28] 50.1 0.3 1.0 1.2 1:8 105
T3C2 5.29 3.0 11.0 1.2 1:8 105
T3C3 4.0 3.0 6.0 1.2 1:8 105
T3C4 1.37 3.0 8.0 1.2 1:8 105
T3C5 9.0 4.0 15.0 1.2 1:8 105
Relaminarization 1.4 5.5 15 1.2 1:8 105

viscosity ratio (i.e., the ! inlet condition) to match the experimentally boundary layer with a turbulence intensity of 3%. The same effect
measured turbulence levels at various downstream locations. As the could have been accomplished with a small step or gap in the CFD
freestream turbulence increases, the transition location moves to geometry. Downstream of the trip, the boundary layer slowly rela-
lower Reynolds numbers. minarizes due to the strong favorable pressure gradient.
The T3C test cases consist of a flat plate with a favorable and For all of the flat-plate test cases, the agreement with the data is
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

adverse pressure gradient imposed by the opposite converging/ generally good, considering the diverse nature of the physical phe-
diverging wall. The wind-tunnel Reynolds number was varied for the nomena computed, ranging from bypass transition to natural tran-
four cases (T3C5, T3C3, T3C2, T3C4), thus moving the transition sition, separation-induced transition, and even relaminarization.
location from the favorable pressure at the beginning of the plate to
the adverse pressure gradient at the end. The cases are used to
demonstrate the transition models ability to predict transition under B. Turbomachinery Test Cases
the influence of various pressure gradients. Figure 5 details the results This section describes a few of the turbomachinery test cases that
for the pressure gradient cases. The effect of the pressure gradient on have been used to validate the transition model including a com-
the transition length is clearly visible with favorable pressure gra- pressor blade, a low-pressure turbine, and a high-pressure turbine. A
dients increasing the transition length and adverse pressure gradients summary of the inlet conditions is shown in Table 2.
reducing it. For the T3C4 case, the laminar boundary layer actually For the Zierke and Deutsch [31] compressor blade, transition on
separates and undergoes separation-induced transition. the suction side occurs at the leading edge due to a small leading-edge
The relaminarization test case is shown in Fig. 6. For this case, the separation bubble on the suction side. On the pressure side, transition
opposite converging wall imposes a strong favorable pressure occurs at about midchord. The turbulence contours and the skin
gradient that can relaminarize a turbulent boundary layer. In both the friction distribution are shown in Fig. 7. There appears to be a signi-
experiment and in the CFD prediction, the boundary layer was ficant amount of scatter in the experimental data; however, overall,
tripped near the plate leading edge. In the CFD computation, this was the transition model is predicting the major flow features correctly
accomplished by injecting a small amount of turbulent air into the (i.e., fully turbulent suction side, transition at midchord on the

Fig. 4 Results for flat-plate test cases with different freestream turbulence levels.
 LANGTRY AND MENTER 2901
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

Fig. 5 Results for flat-plate test cases where variation of the tunnel Reynolds number causes transition to occur in different pressure gradients (dp=dx).

The Pratt and Whitney PAK-B low-pressure turbine blade is a

particularly interesting airfoil because it has a loading profile similar
to the rotors found in many modern aircraft engines [32]. The low-
pressure rotors on modern aircraft engines are extremely challenging
flowfields. This is because, in many cases, the transition occurs in the
free shear layer of a separation bubble on the suction side [4]. The
onset of transition in the free shear layer determines whether or not
the separation bubble will reattach as a turbulent boundary layer and,
ultimately, whether or not the blade will stall. The present transition
model would therefore be of great interest to turbine designers if it
can accurately predict the transition onset location for these types of
flows.
Huang et al. [33] conducted experiments on the PAK-B blade
cascade for a range of Reynolds numbers and turbulence intensities.
The experiments were performed at the design incidence angle for
Fig. 6 Predicted skin friction Cf for a flat plate with a strong Reynolds numbers of 50,000; 75,000; and 100,000 based on inlet
acceleration that causes the boundary layer to relaminarize. velocity and axial chord length, with turbulence intensities of 0.08,
2.35, and 6.0% (which corresponded to values of 0.08, 1.6, and
2.85% at the leading edge of the blade). The computed pressure
pressure side). One important issue to note is the effect of streamwise coefficient distributions obtained with the transition model and fully
grid resolution on resolving the leading-edge laminar separation and turbulent model are compared to the experimental data for the 75,000
subsequent transition on the suction side. If the number of Reynolds number, 2.35% turbulent intensity case in Fig. 8. On the
streamwise nodes clustered around the leading edge is too low, the suction side, a pressure plateau due to a laminar separation with
model cannot resolve the rapid transition and a laminar boundary turbulent reattachment exists. The fully turbulent computation
layer on the suction side is the result. For the present study, 60 completely misses this phenomenon because the boundary layer
streamwise nodes were used between the leading edge and the remains attached over the entire length of the suction surface. The
x=C  0:1 location. transition model can predict the pressure plateau due to the laminar

Table 2 Inlet conditions for the turbomachinery test cases

Case Rex  cUo =, 106 Mach  Uo =a where speed of Chord c, m FSTI, % t =
sound a 
RT0:5
Zierke and Deutsch [31] compressor 0.47 0.1 0.2152 0.18 2.0
incidence  1:5 deg
Pak-B low-pressure turbine blade 0.05, 0.075, 0.1 0.03 0.075 0.08, 2.35, 6.0 6.5–30
VKI MUR transonic guide vane 0.26 Inlet: 0.15 0.037 1.0, 6.5 11, 1000
Outlet: 1.06
 2902 LANGTRY AND MENTER

the ability of the model to capture the effects of Reynolds number and
turbulence intensity variations on the size of a laminar separation
bubble and the subsequent turbulent reattachment.
The surface heat transfer for the transonic von Karman Institute
(VKI) MUR 241 (FSTI  6:0%) and MUR 116 (FSTI  1:0%) test
cases [34] is shown in Fig. 10. The strong acceleration on the suction
side for the MUR 241 case keeps the flow laminar until a weak shock
at midchord, whereas for the MUR 116 case, the flow is laminar until
right before the trailing edge. Downstream of transition there appears
to be a significant difference between the predicted turbulent heat
transfer and the measured value. It is possible that this is the result of a
Mach number (inlet Mach number Mainlet  0:15, Maoutlet  1:089)
effect on the transition length [35]. At present, no attempt has been
made to account for this effect in the model. It can be incorporated in
future correlations, if found consistently important.
The pressure side heat transfer is of particular interest for this
case. For both cases, transition did not occur on the pressure side,
however, the heat transfer was significantly increased for the high
turbulence intensity case. This is a result of the large freestream
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

levels of turbulence which diffuse into the laminar boundary layer

and increase the heat transfer and skin friction. From a modeling
standpoint, the effect was caused by the large freestream viscosity
ratio necessary for MUR 241 to keep the turbulence intensity from
decaying below 6%, which is the freestream value quoted in the
experiment. The enhanced heat transfer on the pressure side was
also present in the experiment and the effect appears to be physical.
The model can predict this effect, as the intermittency does not
multiply the eddy viscosity but only the production term of the k
equation. The diffusive terms are therefore active in the laminar
region.

Fig. 7 Turbulence intensity contours (top) and Cf distribution against

experimental data (bottom) for the Zierke and Deutsch compressor [31]. C. Aeronautical Test Cases
This section describes a few of the aeronautical test cases that have
been used to validate the transition model including a 2-D airfoil, a
2-D three-element flap, and a transonic wing. A summary of the inlet
conditions is shown in Table 3.
The S809 airfoil is a 21% thick, laminar-flow airfoil that was
designed specifically for horizontal-axis wind turbine applications.
The airfoil profile is shown in Fig. 11. The experimental results
where obtained in the low-turbulence wind tunnel at the Delft
University of Technology [36,37]. The detailed CFD results can be
found in [38]. The predicted pressure distribution around the airfoil
for angles of attack (AoA) of 1 deg is shown in Fig. 12. For the 1 deg
AoA case, the flow is laminar for the first 0.5 chord of the airfoil on
both the suction and pressure sides. The boundary layers then
undergo a laminar separation and reattach as a turbulent boundary
layer and this is clearly visible in the experimental pressure distri-
bution plateaus. The fully turbulent computation obviously does not
capture this phenomenon, as the turbulent boundary layers remain
completely attached. Both the transitional CFD and X-Foil solutions
Fig. 8 Predicted blade loading for the Pak-B low-pressure turbine at a do predict the laminar separation bubble. However, X-Foil appears to
Reynolds number of 75,000 and a FSTI of 2.35%. slightly overpredict the reattachment location while the transitional
CFD simulation is in very good agreement with the experiment.
The predicted transition locations as a function of angle of attack
separation and the subsequent turbulent reattachment location. The are shown in Fig. 13. The experimental transition locations were
pressure side was predicted to be fully attached and laminar. obtained using a stethoscope method (Somers [37]). At the moderate
The computed pressure coefficient distributions for various angles of attack, all of the results appear be to within approximately
Reynolds numbers and freestream turbulence intensities compared to 5% chord of each other. The results obtained for the lift and drag
experimental data are shown in Fig. 9. In this figure, the comparisons polars are shown in Figs. 14 and 15. Between 0 and 9 deg, the lift
are organized such that the horizontal axis denotes the Reynolds coefficients Cl predicted by the transitional CFD and the X-Foil code
number whereas the vertical axis corresponds to the freestream are in very good agreement with the experiment, whereas the fully
turbulence intensity of the specific case. As previously pointed out, turbulent CFD results appear to underpredict the lift curve by appro-
the most important feature of this test case is the extent of the ximately 0.1. Between 0 and 9 deg, the drag coefficient Cd predicted
separation bubble on the suction side, characterized by the plateau in by the transitional CFD and X-Foil code are in very good agreement
the pressure distribution. The size of the separation bubble is actually with the experiment, whereas the fully turbulent CFD simulation
a complex function of the Reynolds number and the freestream significantly overpredicts the drag, as expected.
turbulence value. As the Reynolds number or freestream turbulence The McDonnell Douglas 30P-30N flap configuration was
decrease, the size of the separation and hence the pressure plateau originally a test case for the high-lift workshop/CFD challenge that
increases. The computations with the transition model compare well was held at the NASA Langley Research Center in 1993 [39]. It is a
with the experimental data for all of the cases considered, illustrating very complex test case for a transition model because of the large
 LANGTRY AND MENTER 2903
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

Fig. 9 Blade loading for the Pak-B low-pressure turbine at various FSTI and Reynolds numbers Re.

measured using hot films on the upper surface of the slat and flap and
on both the upper and lower surfaces of the main element. The skin
friction was also measured at various locations using a Preston tube
[39]. For the present comparison, the Reynolds number Re  9
106 and an angle of attack AoA  8 deg was selected. The free-
stream conditions for k and ! were selected to match the transition
location at the suction side of the slat. The other transition locations
are an outcome of the simulation.
A contour plot of the predicted turbulence intensity around the flap
is shown in Fig. 16. Also indicated are the various transition locations
that were measured in the experiment (Exp.) as well as the locations
predicted by the present transition model (CFD). In the compu-
tations, the onset of transition was judged as the location where the
skin friction first started to increase due to the production of turbulent
kinetic energy in the boundary layer. In general, the agreement
Fig. 10 Heat transfer for the VKI MUR 241 (FSTI
6:0%) and between the measured and predicted transition locations is very
MUR 116 (FSTI
1:0%) test cases. good. The largest error was observed on the lower surface of the main
element where the predicted transition location was too far down-
stream by approximately 6% of the cruise-airfoil chord.
changes in pressure gradient and the varying local freestream The DLR F-5 geometry is a 20 deg swept wing with a symmetrical
turbulence intensity around the different lifting surfaces. airfoil section that is supercritical at a freestream Mach number of
The experiment was performed in NASA Langley’s low- 0.82. The experiment was performed at the DLR by Sobieczky [40]
turbulence pressure tunnel and the transition locations were and consists of a wing mounted to the tunnel sidewall (which is

Table 3 Inlet conditions for the aeronautical test cases

Case Rex , 106 Mach Chord, m FSTI, % t =
S809 airfoil 2.0 0.1 1 0.2 10
30P-30N flap, AoA  8:1 deg 9.0 0.2 0.5588 0.6 2.5
DLR F-5 wing, AoA  2 deg 1.5 0.82 0.15 0.5 10
 2904 LANGTRY AND MENTER

Fig. 11 S809 airfoil profile.

Fig. 15 Drag coefficient Cd polar for the S809 airfoil.

assumed to have transitioned far upstream of the wing). At the root,

the wing was designed to blend smoothly into the wall, thus
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

eliminating the horseshoe vortex that usually develops there. The

experimental measurements consist of wing-mounted static taps at
various spanwise locations and flow visualization of the surface shear
using a sublimation technique. The experimental flow visualization
is shown in Fig. 17 (right). Based on the flow visualization and the
pressure measurements, a diagram of the flowfield around the wing
was constructed and can be seen in Fig. 17 (middle). From the
measurements, the boundary layer is laminar until about 60% chord
where a shock causes the laminar boundary layer to separate and
reattach as a turbulent boundary layer. The contours of skin friction
and the surface streamlines predicted by the transition model are
shown in Fig. 17 (left). From the skin friction, the laminar separation
Fig. 12 Pressure distribution Cp for the S809 airfoil at 1 deg angle of and turbulent reattachment can be clearly seen and both appear to be
attack. in good agreement with the experimental diagram from about 20%
span out to the wing tip. Near the wing–body intersection, the experi-
ments indicate earlier transition than the simulations. This might be
due to the omission of the crossflow instability in the transition
model.

IV. Conclusions
In this paper, various methods for transition prediction in general-
purpose CFD codes have been discussed. In addition, the require-
ments that a model has to satisfy to be suitable for implementation
into a general-purpose CFD code have been listed. The main criterion
is that nonlocal operations must be avoided. A new concept of
transition modeling termed local correlation-based transition model
(LCTM) was introduced. It combines the advantages of locally
formulated transport equations with the physical information
contained in empirical correlations. The   Req transition model is
representative of that modeling concept. The model is based on two
new transport equations (in addition to the k and ! equations), one for
Fig. 13 Transition location (xt=c) vs angle of attack for the S809 airfoil.
intermittency and one for a transition onset criterion in terms of
momentum-thickness Reynolds number. The proposed transport
equations do not attempt to model the physics of the transition
process (unlike, e.g., turbulence models), but form a framework for
the implementation of transition correlations into general-purpose
CFD methods.
An overview of the -Req model formulation has been given along
with the publication of the full model including some previously
undisclosed empirical correlations that control the predicted tran-
sition length. The main goal of the present paper was to publish the
full model and release it to the research community so that it can
continue to be further validated and possibly extended. Included in
this article are a number of test cases that can be used to validate the
implementation of the model in a given CFD code.
The present transition model accounts for transition due to free-
stream turbulence intensity, pressure gradients, and separation. It is
fully CFD-compatible and does not negatively affect the conver-
gence of the solver. Current limitations of the model are that
Fig. 14 Lift coefficient Cl polar for the S809 airfoil. crossflow instability or roughness are not included in the correlations
 LANGTRY AND MENTER 2905
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

Fig. 16 Contour of turbulence intensity Tu around the McDonnell Douglas 30P-30N flap as well as the measured (Exp.) and predicted (CFD) transition
locations (x=c) as a function of the cruise-airfoil chord (c
0:5588 m). Also indicated is the relative error between the experiment and the predicted
transition locations.

Fig. 17 DLR-F5 wing with transition: simulations (left), experiment (middle and right).

and that the transition correlations are formulated non-Galilean

invariant. These limitations are currently being investigated and can Acknowledgments
be removed in principle.
An overview of the test cases computed with the new model has The model development and validation was done over the course
been given. Because of the nature of the paper, the presentation of of the first author’s Ph.D. thesis and employment at ANSYS CFX.
each individual test case had to be brief. More details on the test case The project was funded by General Electric Aircraft Engines and
setup, boundary conditions, grid resolutions, etc., can be found in the General Electric Global Research, and the authors would like to
literature. The purpose of the overview was to show that the model thank Stefan Voelker and Bill Solomon of General Electric for their
can handle a wide variety of geometries and physically diverse support and numerous thoughtful discussions throughout the course
problems. of the model development. We would also like to thank G. Huang of
The authors believe that the current model is a significant step Wright State University and B. Suzen of North Dakota State
forward in engineering transition modeling. Through the use of University who have supported the original model development with
transport equations instead of search or line-integration algorithms, their extensive know-how and in-house codes. Finally, the authors
the model formulation offers a flexible environment for engineering would also like to thank Chris Rumsey from the NASA Langley
transition predictions that is fully compatible with the infrastructure Research Center for supplying the geometry and experimental data
of modern CFD methods. As a result, the model can be used in any for the McDonnell Douglas 30P-30N flap, and Helmut Sobieczky of
general-purpose CFD method without special provisions for the DLR for his helpful discussions on the DLR F-5 test case.
geometry and grid topology. The authors believe that the LCTM
concept of combining transition correlations with locally formulated
transport equations has a strong potential for allowing the first-order References
effects of transition to be included into today’s industrial CFD [1] Schlichting, H., Boundary Layer Theory, 7th ed., McGraw–Hill, New
simulations. York, 1979.
 2906 LANGTRY AND MENTER

[2] Morkovin, M. V., “On the Many Faces of Transition,” Viscous Drag [22] Menter, F. R., Esch, T., and Kubacki, S., “Transition Modelling Based
Reduction, edited by C. S. Wells, Plenum, New York, 1969, pp. 1–31. on Local Variables,” Proceedings of the 5th International Symposium
[3] Malkiel, E., and Mayle, R. E., “Transition in a Separation Bubble,” on Engineering Turbulence Modelling and Measurements, Elsevier,
Journal of Turbomachinery, Vol. 118, No. 4, 1996, pp. 752–759. Amsterdam, 2002, pp. 555–564.
[4] Mayle, R. E., “The Role of Laminar-Turbulent Transition in Gas [23] Menter, F. R., Langtry, R. B., Likki, S. R., Suzen, Y. B., Huang, P. G.,
Turbine Engines,” Journal of Turbomachinery, Vol. 113, No. 4, 1991, and Völker, S., “A Correlation Based Transition Model Using Local
pp. 509–537. Variables Part 1: Model Formulation,” Journal of Turbomachinery,
doi:10.1115/1.2929110 Vol. 128, No. 3, 2006, pp. 413–422.
[5] Smith, A. M. O., and Gamberoni, N., “Transition, Pressure Gradient [24] Langtry, R. B., Menter, F. R., Likki, S. R., Suzen, Y. B., Huang, P. G.,
and Stability Theory,” Douglas Aircraft Co., Rept. ES 26388, Long and Völker, S., “A Correlation Based Transition Model Using Local
Beach, CA, 1956. Variables Part 2: Test Cases and Industrial Applications,” Journal of
[6] van Ingen, J. L., “A Suggested Semi-Empirical Method for the Turbomachinery, Vol. 128, No. 3, 2006, pp. 423–434.
Calculation of the Boundary Layer Transition Region,” Univ. of Delft, doi:10.1115/1.2184353
Dept. Aerospace Engineering, Rept. VTH-74, Delft, The Netherlands, [25] Menter, F. R., Langtry, R. B., and Völker, S., “Transition Modelling for
1956. General Purpose CFD Codes,” Flow, Turbulence and Combustion,
[7] Stock, H. W., and Haase, W., “Navier-Stokes Airfoil Computations with Vol. 77, Nos. 1–4, 2006, pp. 277–303.
eN Transition Prediction Including Transitional Flow Regions,” AIAA doi:10.1007/s10494-006-9047-1
Journal, Vol. 38, No. 11, 2000, pp. 2059–2066. [26] Menter, F. R., “Two-Equation Eddy-Viscosity Turbulence Models for
doi:10.2514/2.893 Engineering Applications,” AIAA Journal, Vol. 32, No. 8, 1994,
[8] Drela, M., and Giles, M. B., “Viscous-Inviscid Analysis of Transonic pp. 1598–1605.
and Low Reynolds Number Airfoils,” AIAA Journal, Vol. 25, No. 10, doi:10.2514/3.12149
[27] Langtry, R. B., “A Correlation-Based Transition Model Using Local
Downloaded by MONASH UNIVERSITY on November 20, 2013 | http://arc.aiaa.org | DOI: 10.2514/1.42362

1987, pp. 1347–1355.

doi:10.2514/3.9789 Variables for Unstructured Parallelized CFD Codes,” Ph.D. Thesis,
[9] Youngren, H., and Drela, M., “Viscous-Inviscid Method for Univ. of Stuttgart, Stuttgart, Germany, 2006, http://elib.uni-stuttgart.de/
Preliminary Design of Transonic Cascades,” AIAA Paper No. 91- opus/volltexte/2006/2801/.
2364, 1991. [28] Schubauer, G. B., and Klebanoff, P. S., “Contribution on the Mechanics
[10] Jones, W. P., and Launder, B. E., “The Calculation of Low Reynolds of Boundary Layer Transition,” NACA TN 3489, 1955.
Number Phenomena with a Two-Equation Model of Turbulence,” [29] Sinclair, C., and Wells, C. S., Jr., “Effects of Freestream Turbulence on
International Journal of Heat and Mass Transfer, Vol. 16, 1973, Boundary-Layer Transition,” AIAA Journal, Vol. 5, No. 1, 1967,
pp. 1119–1130. pp. 172–174.
doi:10.1016/0017-9310(73)90125-7 doi:10.2514/3.3931
[11] Rodi, W., and Scheuerer, G., “Calculation of Laminar-Turbulent [30] McIlroy, H. M., and Budwig, R. S., “The Boundary Layer over Turbine
Boundary Layer Transition on Turbine Blades,” AGARD CP 390, Blade Models with Realistic Rough Surfaces,” ASME Turbo Expo,
1984. American Society of Mechanical Engineers, GT2005-68342, 2005.
[12] Savill, A. M., “Some Recent Progress in the Turbulence Modeling of [31] Zierke, W. C., and Deutsch, S., “The Measurement of Boundary Layers
By-Pass Transition,” Near-Wall Turbulent Flows, edited by R. M. C. on a Compressor Blade in Cascade,” Vols. 1–2, NASA CR 185118,
So, C. G. Speziale, and B. E. Launder, Elsevier, New York, 1993, p. 829. 1989.
[13] Savill, A. M., One-Point Closures Applied to Transition, Turbulence [32] Dorney, D. J., Lake, J. P., King, P. L., and Ashpis, D. E., “Experimental
and Transition Modeling, edited by M. Hallbäck, Kluwer, Dordrecht, and Numerical Investigation of Losses in Low-Pressure Turbine Blade
The Netherlands, 1996, pp. 233–268. Rows,” AIAA Paper AIAA-2000-0737, 2000.
[14] Wilcox, D. C. W., “Simulation of Transition with a Two-Equation [33] Huang, J., Corke, T. C., Thomas, F. O., “Plasma Actuators for
Turbulence Model,” AIAA Journal, Vol. 32, No. 2, 1994, pp. 247–255. Separation Control of Low Pressure Turbine Blades,” AIAA Paper
doi:10.2514/3.59994 No. AIAA-2003-1027, 2003.
[15] Langtry, R. B., and Sjolander, S. A., “Prediction of Transition for [34] Arts, T., Lambert de Rouvroit, M., and Rutherford, A. W., “Aero-
Attached and Separated Shear Layers in Turbo Machinery,” AIAA Thermal Investigation of a Highly Loaded Transonic Linear Turbine
Paper 2002-3643, 2002. Guide Vane Cascade,” von Karman Inst. for Fluid Dynamics, TN 174,
[16] Walters, D. K., and Leylek, J. H., “A New Model for Boundary-Layer 1990.
Transition Using a Single-Point RANS Approach,” American Society [35] Steelant, J., and Dick, E., “Modeling of Laminar-Turbulent Transition
of Mechanical Engineers, IMECE2002-HT-32740, 2002. for High Freestream Turbulence,” Journal of Fluids Engineering,
[17] Abu-Ghannam, B. J., and Shaw, R., “Natural Transition of Boundary Vol. 123, No. 1, 2001, pp. 22–30.
Layers: The Effects of Turbulence, Pressure Gradient, and Flow doi:10.1115/1.1340623
History,” Journal of Mechanical Engineering Science, Vol. 22, No. 5, [36] Somers, D. M., “Design and Experimental Results for the S809 Airfoil,”
1980, pp. 213–228. Airfoils, Inc., State College, PA, 1989.
doi:10.1243/JMES_JOUR_1980_022_043_02 [37] Somers, D. M., “Design and Experimental Results for the S809 Airfoil,”
[18] Mayle, R. E., “The Role of Laminar-Turbulent Transition in Gas National Renewable Energy Lab., SR-440-6918, Jan. 1997.
Turbine Engines,” Journal of Turbomachinery, Vol. 113, No. 4, 1991, [38] Langtry, R. B., Gola, J., and Menter, F. R., “Predicting 2-D Airfoil and
pp. 509–537. 3-D Wind Turbine Rotor Performance Using a Transition Model for
doi:10.1115/1.2929110 General CFD Codes,” AIAA Paper 2006-0395, 2006.
[19] Suzen, Y. B., Huang, P. G., Hultgren, L. S., and Ashpis, D. E., [39] Klausmeyer, S. M., and Lin, J. C., “Comparative Results from a CFD
“Predictions of Separated and Transitional Boundary Layers Under Challenge over a 2-D Three-Element High-Lift Airfoil,” NASA,
Low-Pressure Turbine Airfoil Conditions Using an Intermittency TM 112858, 1997.
Transport Equation,” Journal of Turbomachinery, Vol. 125, No. 3, [40] Sobieczky, H., “DLR, F5: Test Wing for CFD and Applied
July 2003, pp. 455–464. Aerodynamics,” Test Case B-5 in AGARD FDP Advisory Rept. AR 303:
doi:10.1115/1.1580159 Test Cases for CFD Validation, AGARD, 1994.
[20] Durbin, P. A., Jacobs, R. G., and Wu, X., “DNS of Bypass Transition,” [41] Fashifar, A., and Johnson, M. W., “An Improved Boundary Layer
Closure Strategies for Turbulent and Transitional Flows, edited by B. Transition Correlation,” American Society of Mechanical Engineers
E. Launder, and N. D. Sandham, Cambridge Univ. Press, Cambridge, Paper ASME-92-GT-245, 1992.
England, U.K., 2002, pp. 449–463. [42] Schubauer, G. B., and Skramstad, H. K., “Laminar-Boundary-Layer
[21] Van Driest, E. R., and Blumer, C. B., “Boundary Layer Transition: Oscillations and Transition on a Flat Plate,” NACA Rept. 909, 1948.
Freestream Turbulence and Pressure Gradient Effects,” AIAA Journal,
Vol. 1, No. 6, June 1963, pp. 1303–1306. M. Glauser
doi:10.2514/3.1784 Associate Editor