Jump to navigation Jump to search
 
(15 intermediate revisions by the same user not shown)
Line 12: Line 12:
then expanding into a diffuser, whose upper  wall  and  one  side  wall  are
then expanding into a diffuser, whose upper  wall  and  one  side  wall  are
appropriately deflected, has been  investigated  experimentally  
appropriately deflected, has been  investigated  experimentally  
(Cherry  ''et al.'', 2008, 2009) and computationally  by  means  of  DNS  (Direct  Numerical
([[UFR_4-16_References#7|Cherry  ''et al.'', 2008]], [[UFR_4-16_References#8|2009]])
Simulation; Ohlsson ''et al.'', 2009, 2010), LES  (Large-Eddy  Simulation)  as
and computationally  by  means  of  DNS  (Direct  Numerical
Simulation;
[[UFR_4-16_References#23|Ohlsson ''et al.'', 2009]], [[UFR_4-16_References#24|2010]]),
LES  (Large-Eddy  Simulation)  as
well as by different hybrid LES/RANS  and  RANS  (Reynolds-Averaged  Navier-Stokes) models.
well as by different hybrid LES/RANS  and  RANS  (Reynolds-Averaged  Navier-Stokes) models.
The results  of  the  computational  studies  were  analysed
The results  of  the  computational  studies  were  analysed
along with the experimental reference  database  in  the  framework  of  two
along with the experimental reference  database  in  the  framework  of  two
workshops on ''“Refined  Turbulence  Modelling”''
workshops on ''“Refined  Turbulence  Modelling”''
(Steiner ''et al.'', 2009  and
([[UFR_4-16_References#30|Steiner ''et al.'', 2009]] and
Jakirlic ''et al.'', 2010) organized by the ERCOFTAC Special Interest  Group  on
[[UFR_4-16_References#15|Jakirlić ''et al.'', 2010b]])
Turbulence Modelling (SIG15). Two three-dimensional diffuser  configurations
organized by the ERCOFTAC Special Interest  Group  on
Turbulence Modelling (SIG15).
The flow was also a test case in the European ATAAC project.
Two three-dimensional diffuser  configurations
differing in terms of the values of the expansion angles  —  the  upper-wall
differing in terms of the values of the expansion angles  —  the  upper-wall
expansion angle is reduced from 11.3° (diffuser 1) to 9° (diffuser  2);  the
expansion angle is reduced from 11.3° (diffuser 1) to 9° (diffuser  2);  the
Line 27: Line 33:
geometry led to substantial changes in the flow structure  with  respect  to
geometry led to substantial changes in the flow structure  with  respect  to
the onset, location, shape and  size  of  the  three-dimensional  separation
the onset, location, shape and  size  of  the  three-dimensional  separation
pattern associated with the corner separation and corner reattachment, Fig.
pattern associated with the corner separation and corner reattachment,
1. The inflow in both  considered  cases  is  characterized  by  a  Reynolds
[[UFR_4-16#figure1|Fig. 1]].
The inflow in both  considered  cases  is  characterized  by  a  Reynolds
number Re<sub>h</sub>=10000, based on the inlet duct height.
number Re<sub>h</sub>=10000, based on the inlet duct height.


Line 42: Line 49:
|[[Image:UFR4-16_figure1b.png|center|740px]]
|[[Image:UFR4-16_figure1b.png|center|740px]]
|-
|-
|'''Figure 1:''' Instantaneous velocity  field  in  both  diffuser  configurations, diffuser 1 (upper) and diffuser  2  (lower),  obtained  by  a  zonal  hybrid LES/RANS  model,  illustrating  different  flow  separation  patterns  (from Jakirlic et al., 2010a). Whereas the separation zone spreads over the  upper wall in the diffuser 1, it occupies the deflected side wall in the  diffuser 2
|'''Figure 1:''' Instantaneous velocity  field  in  both  diffuser  configurations, diffuser 1 (upper) and diffuser  2  (lower),  obtained  by  a  zonal  hybrid LES/RANS  model,  illustrating  different  flow  separation  patterns  (from [[UFR_4-16_References#14|Jakirli&#x107; ''et&nbsp;al.'', 2010a]]). Whereas the separation zone spreads over the  upper wall in the diffuser 1, it occupies the deflected side wall in the  diffuser 2
|}
|}


Line 49: Line 56:
----
----
{{ACContribs
{{ACContribs
|authors=Suad Jakirlic
|authors=Suad Jakirli&#x107;, Gisa  John-Puthenveettil
|organisation=Technische Universit&auml;t Darmstadt
|organisation=Technische Universit&auml;t Darmstadt
}}
}}

Latest revision as of 08:27, 30 April 2013

Flow in a 3D diffuser

Front Page

Description

Test Case Studies

Evaluation

Best Practice Advice

References

Confined flows

Underlying Flow Regime 4-16

Abstract

The incompressible flow developing fully in a three-dimensional duct and then expanding into a diffuser, whose upper wall and one side wall are appropriately deflected, has been investigated experimentally (Cherry  et al., 2008, 2009) and computationally by means of DNS (Direct Numerical Simulation; Ohlsson et al., 2009, 2010), LES (Large-Eddy Simulation) as well as by different hybrid LES/RANS and RANS (Reynolds-Averaged Navier-Stokes) models. The results of the computational studies were analysed along with the experimental reference database in the framework of two workshops on “Refined Turbulence Modelling” (Steiner et al., 2009 and Jakirlić et al., 2010b) organized by the ERCOFTAC Special Interest Group on Turbulence Modelling (SIG15). The flow was also a test case in the European ATAAC project. Two three-dimensional diffuser configurations differing in terms of the values of the expansion angles — the upper-wall expansion angle is reduced from 11.3° (diffuser 1) to 9° (diffuser 2); the side-wall expansion angle is increased from 2.56° (diffuser 1) to 4° (diffuser 2) — were considered. These slight modifications in the diffuser geometry led to substantial changes in the flow structure with respect to the onset, location, shape and size of the three-dimensional separation pattern associated with the corner separation and corner reattachment, Fig. 1. The inflow in both considered cases is characterized by a Reynolds number Reh=10000, based on the inlet duct height.

The primary objective of the present contribution is twofold:

  • to provide further insight into the physics of the separation of the three-dimensional boundary layer generated at the intersection of two deflected walls. This separation is the consequence of an adverse pressure gradient evoked by the duct expansion under different conditions and
  • a comparative assessment of different modelling approaches in terms of their capability to accurately capture the size and shape of the three-dimensional flow separation pattern and associated mean flow and turbulence features.


UFR4-16 figure1a.png
UFR4-16 figure1b.png
Figure 1: Instantaneous velocity field in both diffuser configurations, diffuser 1 (upper) and diffuser 2 (lower), obtained by a zonal hybrid LES/RANS model, illustrating different flow separation patterns (from Jakirlić et al., 2010a). Whereas the separation zone spreads over the upper wall in the diffuser 1, it occupies the deflected side wall in the diffuser 2





Contributed by: Suad Jakirlić, Gisa John-Puthenveettil — Technische Universität Darmstadt

Front Page

Description

Test Case Studies

Evaluation

Best Practice Advice

References


© copyright ERCOFTAC 2024