UFR 4-16 Test Case: Difference between revisions

From KBwiki
Jump to navigation Jump to search
 
(309 intermediate revisions by 2 users not shown)
Line 1: Line 1:
= Flow in a 3D diffuser =
= Flow in a 3D diffuser =
{{UFRHeader
{{UFRHeader
Line 64: Line 63:
resolution of 0.4 x 0.4 x 0.4 mm. Interested readers are referred to
resolution of 0.4 x 0.4 x 0.4 mm. Interested readers are referred to
[[UFR_4-16_References#7|Cherry ''et al.'' (2008]],
[[UFR_4-16_References#7|Cherry ''et al.'' (2008]],
[[UFR_4-16_References#7|2009)]]
[[UFR_4-16_References#8|2009)]]
for more details about the measurement technique.
for more details about the measurement technique.


Line 81: Line 80:


<div id="figure6"></div>
<div id="figure6"></div>
{|align="center" width="558"
|[[Image:UFR4-16_figure6.jpg|548px]]
|-
|'''Figure 6:''' 3D diffuser arrangement in a medical  magnetic  resonance  imaging device. Courtesy of J. Eaton (Stanford University)
|}
===Mean velocity and Reynolds stress measurements===
Cherry ''et&nbsp;al.'' provided a detailed reference database comprising  the
three-component mean velocity field and the streamwise Reynolds  stress  component
field within the  entire  diffuser  section.  Both  diffuser  configurations
considered  are  characterized  by  a  three-dimensional  boundary-layer
separation, but the slightly different expansion geometries caused the  size
and shape of the separation bubble exhibiting a  high  degree  of  geometric
sensitivity to the dimensions of the diffuser as illustrated in
[[UFR_4-16_Test_Case#figure7|Figs.  7]],  [[UFR_4-16_Test_Case#figure8|8]]
and [[UFR_4-16_Test_Case#figure9|9]]
<div id="figure7"></div>
{|align="center" width="750"
|[[Image:UFR4-16_figure7.png|740px]]
|-
|'''Figure 7:''' Streamwise velocity contours in a plane parallel to the top wall, from [[UFR_4-16_References#7|Cherry ''et&nbsp;al.'' (2008)]]
|}
<div id="figure8"></div>
{|align="center" width="750" style="border: 1px solid black;" border="1"
|align="center" colspan="3"|[[Image:UFR4-16_figure8a.png|493px]]
|-
|colspan="3"|[[Image:UFR4-16_figure8b.png|750px]]
|-
|[[Image:UFR4-16_figure8c.png|247px]]||[[Image:UFR4-16_figure8d.png|247px]]||[[Image:UFR4-16_figure8e.png|247px]]
|-
|colspan="3"|'''Figure 8:'''  Measured  streamwise  velocity  contours  in  the  central  plane (upper) of the '''Diffuser 1''' and in the three selected  cross-sectional  slices positioned at different distances from  the  diffuser  inlet  (lower;  their locations are denoted by thick white lines in the upper figure).  Note  that the black line indicates zero streamwise velocity and the  purple  and  pink regions are reverse flow. The velocity values are  normalized  by  the  bulk inlet velocity being ''V<sub>ref</sub>=1&nbsp;m/s''. Courtesy of J. Eaton (Stanford University)
|}
<div id="figure9"></div>
{|align="center" width="750" style="border: 1px solid black" border="1"
|colspan="3" align="center"|[[Image:UFR4-16_fig9a.png|493px]]
|-
|colspan="3"|[[Image:UFR4-16_fig9b.png|750px]]
|-
|valign="bottom"|[[Image:UFR4-16_fig9c.png|246px]]
|valign="bottom"|[[Image:UFR4-16_fig9d.png|246px]]
|valign="bottom"|[[Image:UFR4-16_fig9e.png|246px]]
|-
|colspan="3"|'''Figure 9:'''  Measured  streamwise  velocity  contours  in  the  central  plane (upper) of the '''Diffuser 2''' and in the three selected  cross-sectional  slices positioned at different distances from  the  diffuser  inlet  (lower;  their locations are denoted by thick white lines in the upper figure).  Note  that the black line indicates zero streamwise velocity and the  purple  and  pink regions are reverse flow. The velocity values are  normalized  by  the  bulk inlet velocity being ''V<sub>ref</sub>=1&nbsp;m/s''. Courtesy of J. Eaton (Stanford University)
|}
<div id="figure10"></div>
{|align="center" width="750" style="border: 1px solid black"
!x/h=2!!x/h=5!!x/h=8!!x/h=12!!x/h=15
|-
|valign="bottom"|[[Image:UFR4-16_fig10a.png|137px]]
|valign="bottom"|[[Image:UFR4-16_fig10b.png|139px]]
|valign="bottom"|[[Image:UFR4-16_fig10c.png|148px]]
|valign="bottom"|[[Image:UFR4-16_fig10d.png|154px]]
|valign="bottom"|[[Image:UFR4-16_fig10e.png|162px]]
|-
|align="center" colspan="5"|[[Image:UFR4-16_fig10f.png]]
|-
|colspan="5"|'''Figure 10:''' Measurements of turbulent streamwise stress components  taken  in cross-sectional slices of the diffuser 1 perpendicular  to  the  mean  flow. The region of highest turbulence (red areas; ignore the  red  areas  at  the bottom walls) follows the shear layer between forward and reverse flow.  ''h''=1cm represents the height of the inflow duct. Courtesy of J. Eaton (Stanford University)
|}
===Pressure measurements===
In addition
[[UFR_4-16_References#8|Cherry ''et&nbsp;al.'' (2009)]]
provided the  pressure  distribution  along
the bottom non-deflected wall of diffuser 1 at different  Reynolds  numbers.
Complementary to the Reynolds number 10000 (for which the entire flow  field
was measured), two higher Reynolds numbers &mdash; 20000 and  30000  &mdash;  were  also
considered, [[UFR_4-16_Test_Case#figure11|Fig. 11]]. The surface  pressure  distribution  was  evaluated  to
yield the coefficient
<math>C_p=(p-p_\textrm{ref})/(0.5\rho U_\textrm{bulk}^2)</math>;
the  reference  pressure  was  taken  at  the
position ''x/L&nbsp;=&nbsp;0.05''. The pressure curve  exhibits  a  development  typical  of
flow in diverging ducts.  The  pressure  decrease  in  the  inflow  duct  is
followed by a steep pressure increase already at the very end of the  inflow
duct  and  especially  at  the  beginning  of  the  diffuser  section.  The
transition from the initial strong pressure rise to  its  moderate  increase
occurs at ''x/L&nbsp;&#8776;&nbsp;0.3'', (''x/h=4.5'')
corresponding to the position where about  5%  of
the entire cross-section is occupied by the flow reversal  (see  e.g.,
[[UFR_4-16_Test_Case#figure19|Fig. 19]].
The onset of separation causes a certain contraction of the flow cross&#8208;section,
leading  to  a  weakening  of  the  deceleration  intensity  and,
accordingly, to a slower pressure increase. The region  characterized  by  a
monotonic pressure rise  was  reached  in  the  remainder  of  the  diffuser
section.
<div id="figure11"></div>
{|align="center" width="750"
{|align="center" width="750"
|[[Image:UFR4-16_figure6.jpg|740px]]
|[[Image:UFR4-16_figure11.png|740px]]
|-
|'''Figure 11:''' Pressure recovery coefficients relative to the  pressure  on  the bottom wall of the diffuser 1 inlet in a  range  of  flow  Reynolds  number. L=15 cm represents the length of Diffuser 1, from [[UFR_4-16_References#8|Cherry ''et&nbsp;al.'' (2009)]]
 
|}
 
===Measurements uncertainties===
(adopted  from  [[UFR_4-16_References#7|Cherry ''et&nbsp;al.'',  2008, IJHFF, Vol. 29(3)]])
 
Elkins  et  al.  (2004)  estimated  the  maximum  relative  uncertainty  of
individual mean velocity measurements to be about 10% of the measured  value
in a similar highly  turbulent  flow.  However,  comparisons  to  PIV  in  a
backward facing step flow (Elkins et al.,  2007)  show  that  only  a  small
percentage of MRV velocity samples deviate by that much and  most  are  much
more  accurate.  To  test  this,  the  streamwise  velocity  component  was
integrated over 250 cross-sections of the MRV  data  and  the  results  were
compared to the known volume flow rate. This  indicated  an  uncertainty  in
the integral of less than 2% with a 95% confidence level.
 
Measurements of turbulent normal stresses in  Diffuser  1  were  also  taken
using the MR technique described by Elkins et al.  (2007).  This  method  is
based on diffusion imaging principles in which the turbulence causes a  loss
of net magnetization signal from a voxel in the flow. This  causes  a  decay
in signal strength which can be related to  turbulent  velocity  statistics.
Elkins et al found this method to be accurate within 20% everywhere  in  the
FOV and within 5% in regions of  high  turbulence.  Three  turbulence  scans
were completed using three different magnetic field gradient strengths.  For
each gradient strength, 30 scans were  completed  and  averaged.  The  three
averaged data sets were then averaged to obtain a final data set.
 
===Experimental data available===
Velocity, (and streamwise Reynolds stress components  for  diffuser  1)  and
coordinate  data  for  both  diffusers  are  available    here. <!--at http://stanford.edu/~echerry/-->
 
{|
<!--
|[http://stanford.edu/~echerry/AnnularDiffuserData.zip AnnularDiffuserData.zip] (contains 4 files, 134 MBytes)
-->
|-
|[https://kbwiki-images.s3.amazonaws.com/2/28/Diffuser_1_data.zip Diffuser 1 data.zip] (contains 1 file, 8.4 MBytes)
|-
<!--
|-
|[http://stanford.edu/~echerry/Diffuser%201%20data.zip Diffuser 1 data.zip] (contains 1 file, 8.4 MBytes)
-->
|-
|[https://kbwiki-images.s3.amazonaws.com/4/47/Diffuser_1_turbulence.zip Diffuser 1 turbulence.zip] (contains 1 file, 5.3 MBytes)
|-
|-
|'''Figure 6:''' 3D diffuser arrangement in a medical  magnetic  resonance  imaging device. Courtesy of J. Eaton (Stanford University)
|[https://kbwiki-images.s3.amazonaws.com/8/82/Diffuser_2_data.zip Diffuser 2 data.zip] (contains 1 file, 4 MBytes)
|}
|}
The data are seven 3D  matlab  matrices.  The
x, y, and z matrices give the coordinates of each point  in  the  coordinate
system shown in Fig. 5. The units are meters. The Vx, Vy,  and  Vz  matrices
give the corresponding velocity components for  each  point  in  m/sec.  The
matrix mg gives the relative signal magnitude detected by the MRI machine.
Experimental data for the pressure coefficient in Diffuser 1 for the  inflow
Reynolds  number  Re=10000  are  available [[Media:UFR4-16_Cp_Re=10000.xlsx|here]]
([[Media:UFR4-16_Cp_Re=10000.xlsx|Cp_Re=10000.xlsx]]).  The
coordinate system is the same as the  coordinate  system  described  in  the
corresponding manuscript (see below). <math>{\ C_p}</math> is defined as
<math>\left(p-p_\textrm{ref}\right)/\left(\frac{1}{2} \rho V^2\right)</math>,
where <math>{\ P_\textrm{ref}}</math> is  the  pressure  at  ''x=0.05''  (see  Fig.  11)
at  the  midpoint
(''z/B=0.5'') of the bottom flat wall  (opposite  the  wall  expanding  at  11.3
degrees), <math>{\ \rho}</math> is the density, and <math>{\ V}</math>
is the bulk inlet velocity. The data  were
taken in a line along the bottom wall of Diffuser 1  at  constant  y  and  z
coordinates. L (=15 cm) indicates the length of the diffuser.
''Please acknowledge the authors of the experiment when using their database!''


== CFD Methods ==
== CFD Methods ==
{{Demo_UFR_Test_CFD}}
The flow in the present diffuser configuration was intensively  investigated
computationally in the  framework  of  two  ERCOFTAC-SIG15  Workshops and in
the European ATAAC project.
The workshops focussing on both 3D diffuser  configurations  (denoted  by  SIG15
Case 13.2-1 and SIG15 Case  13.2-2,  respectively)  were  organized  at  the
Technical University of  Graz,  Austria  in  September,  2008  and  the  "LA
Sapienza" University of Rome, Italy in September,  2009.  The  corresponding
reports are published in  the  ERCOFTAC  Bulletin  Issues,
[[UFR_4-16_References#30|Steiner  ''et&nbsp;al.'' (2009)]]
and [[UFR_4-16_References#15|Jakirli&#x107; ''et&nbsp;al.'' (2010b)]],
see [[UFR_4-16_References#0|"List of References"]].
Both in the workshops and in the ATAAC project a wide range of
turbulence models in both LES and RANS  frameworks as  well  as  some  novel
Hybrid LES/RANS formulations have been employed. The computational  database
was furthermore enriched by the results of a Direct Numerical Simulation  of
the first diffuser performed by
[[UFR_4-16_References#24|Ohlsson ''et&nbsp;al.''  (2010)]].  The  list  of  all
computational contributions  to the workshops including  some  basic  information  about  the
methods and models used  and  corresponding  grid  resolution  is  given  in
the following [[UFR_4-16_Test_Case#table1|tables 1]]  and  [[UFR_4-16_Test_Case#table2|table 2]].
For  more  computational  details,  interested
readers are referred to the "[[UFR_4-16_Evaluation#Available_CFD_results:_ERCOFTAC_SIG15_Workshop_Proceedings|workshop proceedings]]"
&mdash; see  the  corresponding
links at the end of the Section "[[UFR_4-16_Evaluation#Evaluation_of_the_results|Evaluation of the results]]".
The results from the ATAAC project and information on the methods used can be obtained through the links
[https://kbwiki-images.s3.amazonaws.com/c/c8/ATAAC_D3-2-36_excerpt3DDiffuser.pdf ATAAC_D3-2-36_excerpt3DDiffuser.pdf]
(excerpt from an ATAAC report) and
[[https://kbwiki-images.s3.amazonaws.com/a/a1/ATAAC_finalWorkshop_ST04-Diffuser-ANSYS.pdf ATAAC_finalWorkshop_ST04-Diffuser-ANSYS.pdf]]
(PowerPoint presentation at ATAAC final workshop).
 
 
<div id="table1"></div>
{|align="center" width="750"
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_table1.png|740px]]
|-
|'''Table 1:''' SIG15 Case 13.2-1 (Diffuser 1) &mdash; contributors  and  methods  (note that the DNS  grid  comprises  220  million  cells  in  the  follow-up  work published in [[UFR_4-16_References#24|Ohlsson ''et&nbsp;al.'' (2010)]])
|-
|'''N.B.''' ITS is "Institut f&uuml;r Thermische Str&ouml;mungsmaschinen"
|}
 
 
<div id="table2"></div>
{|align="center" width="750"
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_table2.png|740px]]
|-
|'''Table 2:''' SIG15 Case 13.2-2 (Diffuser 2) &mdash; contributors and methods
|-
|'''N.B.''' ITS is "Institut f&uuml;r Thermische Str&ouml;mungsmaschinen"
|}
 
 
===Direct numerical simulation of the flow in a 3D  diffuser===
 
(adopted from [[UFR_4-16_References#24|Ohlsson ''et&nbsp;al.'', 2010, JFM, Vol. 650]])
 
In this part of the present contribution the DNS study of  the  flow  in  3D
Diffuser 1 performed recently by
[[UFR_4-16_References#24|Ohlsson ''et&nbsp;al.'' (2010)]] will be described  in
more details. Ohlsson ''et&nbsp;al.'' participated  with  this  contribution  at  the
14th SIG15  Workshop  on  Refined  Turbulence  Modeling
([[UFR_4-16_References#15|Jakirli&#x107;  ''et&nbsp;al.'', 2010b]]).
Accordingly, their results are also part of the  CFD  methods/models
evaluation &mdash; along with the results obtained  by  different  LES,  RANS  and
Hybrid LES/RANS models (see the [[UFR_4-16_Evaluation#Evaluation_of_the_results|next chapter]]).
In  addition,  as  the  DNS
provided a very comprehensive database comprising all  three  mean  velocity
components  (and  associated  integral  characteristics  such  as  surface
pressure and friction factor) and all  six  Reynolds  stress  components  as
well  as  a  certain  insight  into  the  physics  (not  detected  by  the
experimental  investigation)  one  can  regard  it  also  as  a  reference
investigation, as we do presently.
 
The Direct Numerical Simulation of the Diffuser  1  was  performed  using  a
massively parallel high-order  spectral  element  code.  The  incompressible
Navier-Stokes  equations  are  solved  using  a  Legendre-polynomial-based
spectral-element method, implemented  in  the  code  nek5000,  developed  by
Fischer et al. (2008). The computational domain shown in
[[UFR_4-16_Test_Case#figure12|Fig. 12]] is  set  up
in close agreement with the diffuser geometry in the  experiment
(see  [[UFR_4-16_Test_Case#figure3|Fig.3]])
and consists of the inflow development duct of almost  63  duct  heights,
''h'', (starting at the  non-dimensional  coordinate  ''x''&nbsp;=&nbsp;-62.9),
the  diffuser
expansion located at ''x''&nbsp;=&nbsp;0  and  the  converging  section  upstream  of  the
outlet. The corners resulting  from  the  diffuser  expansion  are  smoothly
rounded with a radius of 6.0 in accordance  with  the  experimental  set-up.
The maximum dimensions are ''Lx =105.4 h, Ly =[h, 4h], Lz =[3.33  h,  4h]''.  In
the inflow duct, laminar flow undergoes natural transition by the use of  an
unsteady trip forcing (see e.g. Schlatter et al., 2009),  which  avoids  the
use of artificial turbulence and eliminates artificial temporal  frequencies
which may arise from inflow recycling  methods  (Herbst  et  al.,  2007).  A
'sponge region' is added at the end of the contraction in order to  smoothly
damp out  turbulent  fluctuations,  thereby  eliminating  spurious  pressure
waves. It is followed by a homogeneous Dirichlet condition for the  pressure
and a homogeneous Neumann condition for the velocities.  The  resolution  of
approximately 220 million grid points is obtained by  a  total  of  127750
local tensor product domains (elements)  with  a  polynomial  order  of  11,
respectively, resulting in &Delta;z<sup>+</sup><sub>max</sub>&asymp;11.6,
&Delta;y<sup>+</sup><sub>max</sub>&asymp;13.2 and
&Delta;x<sup>+</sup><sub>max</sub>&asymp;19.5  in  the
duct center and the first grid point being  located  at  z<sup>+</sup>&asymp;0.074
and  y<sup>+</sup>&asymp;0.37, respectively (note that at the time of the  ERCOFTAC  SIG15  workshop
the DNS grid had 172 million grid points,
see Chapter "[[UFR_4-16_Evaluation#172million|Evaluation]]"). It  was
verified that this resolution yields accurate results in  turbulent  channel
flow simulations. In the diffuser, the grid is linearly  stretched  in  both
directions, but since the mean resolution requirements  decreases  with  the
velocity, which decreases linearly with the area expansion,  the  resolution
in the entire domain is hence satisfactory. The simulation was performed  on
the Blue Gene/P at ALCF, Argonne National Laboratory (32768  cores  and  a
total of 8 million core hours) and on the cluster 'Ekman' at PDC,  Stockholm
(2048 cores and a total of 4  million  core  hours).  Thirteen  flow-through
times, ''t U<sub>b</sub>/L''=13, based on bulk velocity, ''U<sub>b</sub>'', and diffuser length,  ''L=15&nbsp;h'',
were simulated in order to let the  flow  settle  to  an  equilibrium  state
before turbulent  statistics  were  collected  over  approximately  ''U<sub>b</sub>t/L''=21
additional flow-through times.
 
In addition to the three-dimensional mean velocity field, all  six  Reynolds
stress components were evaluated, as well as the surface pressure  and  skin
friction distribution along the bottom wall.
 
 
<div id="figure12"></div>
{|align="center" width="750"
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure12.png|740px]]
|-
|'''Figure 12:''' DNS grid of the  diffuser  1  geometry  showing  the  development region, diffuser expansion, converging section and outlet. From  [[UFR_4-16_References#24|Ohlsson  ''et&nbsp;al.'' (2010)]]
|}
 
 
[[UFR_4-16_Test_Case#figure13|Fig. 13]] illustrates the mean axial flow field obtained  by  DNS  along  with
the experimental data and [[UFR_4-16_Test_Case#figure14|Fig. 14]] depicts the skin friction evolution  along
the bottom diffuser wall at the midpoint
(''z/B=0.5''; ''C<sub>f</sub>=&tau;<sub>wall</sub>&nbsp;/&nbsp;(0.5&rho;U<sup>2</sup><sub>bulk</sub>'')&nbsp;).  The  latter  DNS
result was  evaluated  exclusively  for  the  needs  of  the  14th  ERCOFTAC
Workshop; it is not part of the DNS database which can be downloaded [[UFR_4-16_Test_Case#kth_data|here]].
 
 
<div id="figure13"></div>
{|align="center" width="750"
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure13.png|740px]]
|-
|'''Figure 13:''' Cross-flow planes of streamwise velocity component  at  2,  5,  8 and 15 ''h'' downstream of the diffuser throat. Left: DNS. Right: Experiment  by [[UFR_4-16_References#7|Cherry ''et&nbsp;al.'' (2008)]]. Each streamwise position has its  own  colour  bar  on the right. Contour lines are  spaced  ''0.1 V<sub>ref</sub>''  apart.  Thick  black  lines correspond to the zero velocity contour. From [[UFR_4-16_References#24|Ohlsson ''et&nbsp;al.'' (2010)]]
 
 
<div id="figure14"></div>
{|align="center" width="750"
|[[Image:UFR4-16_figure14.jpg|740px]]
|-
|'''Figure 14:''' Friction coefficient at the bottom wall of the  diffuser  1  with ''L=15 cm'' representing the length of Diffuser  1.  The  LES  and  HLR  (Hybrid LES/RANS) results are from [[UFR_4-16_References#14|Jakirli&#x107; ''et&nbsp;al.''  (2010a)]]  and  [[UFR_4-16_References#17|John-Puthenveetil (2012)]].
 
===Available DNS data (last update: 2010-11-23)===
<div id="kth_data"></div>
The entire  digitalized  DNS  database  (as  well  as  some  high-resolution
images)    as    listed    below    can    be      downloaded      from
http://www.mech.kth.se/~johan/data/index.html. The links given below are to local wiki copies of the files.
 
This directory contains velocity profiles  and  mean  fluctuations  obtained
from the DNS by
[[UFR_4-16_References#24|Ohlsson ''et&nbsp;al.'', (JFM 650 307&ndash;318)]]
This  simulation  was
designed to match the experiment by
[[UFR_4-16_References#7|Cherry ''et&nbsp;al.'', 2008,  IJHFF  29(3)]].  The
data is given at the same streamwise and spanwise locations  as  during  the
14th  ERCOFTAC  SIG15  Workshop  on  Turbulence  Modelling,  held  in  Rome,
September 2009.
 
The full Reynolds stress budgets will be available in the future.
 
''The data is free to use; please include a proper reference to  the  original publications.''
 
In case of any questions, e.g. related to the data, or whether you wish  for
additional  data  not  presented  here,  please  contact
Johan  Malm, ([mailto:johan@mech.kth.se johan@mech.kth.se]),
Philipp  Schlatter  ([mailto:pschlatt@mech.kth.se pschlatt@mech.kth.se])
or  Dan Henningson ([mailto:henning@mech.kth.se henning@mech.kth.se])
 
====Visualizations and computational mesh====
[[Media:UFR4-16_planes_press0000.jpeg|Crossflow planes with instantaneous streamwise velocity]]
 
[[Media:UFR4-16_planes_nogrid_iso_vec.jpeg|Crossflow planes with instantaneous streamwise velocity and  isosurfaces  of streamwise velocity]]
 
[[Media:UFR4-16_press_persp0010.jpeg|Instantaneous  streamwise  velocity  in  a  spanwise  midplane  with  some isosurfaces of instantaneous pressure]]
 
[[Media:UFR4-16_mesh_paper.png|Mesh]]
 
====Digitalized DNS database: full 3D  mean  velocity  and  Reynolds stress fields====
Explanation:
*E.g., c13.2_Ucont2_KTH_DNS denotes the files comprising the contours of the axial velocity - Ucont - at the streamwise position ''x/h=2''  for  the case 13.2 (ERCOFTAC SIG15 denotation). The same data are given  at  the streamwise locations ''x/h=2, 5, 8, 12'' and ''15''
*E.g., c13.2_urms2_KTH_DNS.txt denotes the file comprising the  contours of the root-mean-square values of the  streamwise  stress  component  &mdash; urms &mdash; at the streamwise position ''x/h=2''. The same data are given at the streamwise locations ''x/h=2, 5, 8, 12'' and ''15''
*The file c13.2_cp_KTH_DNS.txt denotes the file comprising the  pressure coefficient distribution at  the  bottom  flat  wall  at  the  midpoint ''z/B=0.5''
*The file c13.2_z0250_x-2_KTH_DNS.txt denotes the  file  comprising  all three mean velocity components, kinetic energy of  turbulence  and  all six Reynolds  stress  components  at  the  streamwise  position  ''x/h=-2'' (inflow duct) and the spanwise position ''z/B=0.25''.  The  same  data  are given at the streamwise locations ''x/h=0, 2, 4, 6, 8, 10, 12, 14,  15.5, 17, 18.5, 20'' and ''21.5'' at the  following  spanwise  locations  ''z/B=0.25, 0.5, 0.75'' and ''0.875''.
 
 
{|align="center" border="1" cellpadding="5"
|[[Media:c13.2_cp_KTH_DNS.txt|c13.2_cp_KTH_DNS.txt]]
|-
|[[Media:c13.2_Ucont2_KTH_DNS.txt|c13.2_Ucont2_KTH_DNS.txt]]||[[Media:c13.2_urms2_KTH_DNS.txt|c13.2_urms2_KTH_DNS.txt]]
|-
|[[Media:c13.2_Ucont5_KTH_DNS.txt|c13.2_Ucont5_KTH_DNS.txt]]||[[Media:c13.2_urms5_KTH_DNS.txt|c13.2_urms5_KTH_DNS.txt]]
|-
|[[Media:c13.2_Ucont8_KTH_DNS.txt|c13.2_Ucont8_KTH_DNS.txt]]||[[Media:c13.2_urms8_KTH_DNS.txt|c13.2_urms8_KTH_DNS.txt]]
|-
|[[Media:c13.2_Ucont12_KTH_DNS.txt|c13.2_Ucont12_KTH_DNS.txt]]||[[Media:c13.2_urms12_KTH_DNS.txt|c13.2_urms12_KTH_DNS.txt]]
|-
|[[Media:c13.2_Ucont15_KTH_DNS.txt|c13.2_Ucont15_KTH_DNS.txt]]||[[Media:c13.2_urms15_KTH_DNS.txt|c13.2_urms15_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x-2_KTH_DNS.txt|c13.2_z0250_x-2_KTH_DNS.txt]]||[[Media:c13.2_z0500_x-2_KTH_DNS.txt|c13.2_z0500_x-2_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x-2_KTH_DNS.txt|c13.2_z0750_x-2_KTH_DNS.txt]]||[[Media:c13.2_z0875_x-2_KTH_DNS.txt|c13.2_z0875_x-2_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x0_KTH_DNS.txt|c13.2_z0250_x0_KTH_DNS.txt]]||[[Media:c13.2_z0500_x0_KTH_DNS.txt|c13.2_z0500_x0_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x0_KTH_DNS.txt|c13.2_z0750_x0_KTH_DNS.txt]]||[[Media:c13.2_z0875_x0_KTH_DNS.txt|c13.2_z0875_x0_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x2_KTH_DNS.txt|c13.2_z0250_x2_KTH_DNS.txt]]||[[Media:c13.2_z0500_x2_KTH_DNS.txt|c13.2_z0500_x2_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x2_KTH_DNS.txt|c13.2_z0750_x2_KTH_DNS.txt]]||[[Media:c13.2_z0875_x2_KTH_DNS.txt|c13.2_z0875_x2_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x4_KTH_DNS.txt|c13.2_z0250_x4_KTH_DNS.txt]]||[[Media:c13.2_z0500_x4_KTH_DNS.txt|c13.2_z0500_x4_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x4_KTH_DNS.txt|c13.2_z0750_x4_KTH_DNS.txt]]||[[Media:c13.2_z0875_x4_KTH_DNS.txt|c13.2_z0875_x4_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x6_KTH_DNS.txt|c13.2_z0250_x6_KTH_DNS.txt]]||[[Media:c13.2_z0500_x6_KTH_DNS.txt|c13.2_z0500_x6_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x6_KTH_DNS.txt|c13.2_z0750_x6_KTH_DNS.txt]]||[[Media:c13.2_z0875_x6_KTH_DNS.txt|c13.2_z0875_x6_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x8_KTH_DNS.txt|c13.2_z0250_x8_KTH_DNS.txt]]||[[Media:c13.2_z0500_x8_KTH_DNS.txt|c13.2_z0500_x8_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x8_KTH_DNS.txt|c13.2_z0750_x8_KTH_DNS.txt]]||[[Media:c13.2_z0875_x8_KTH_DNS.txt|c13.2_z0875_x8_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x10_KTH_DNS.txt|c13.2_z0250_x10_KTH_DNS.txt]]||[[Media:c13.2_z0500_x10_KTH_DNS.txt|c13.2_z0500_x10_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x10_KTH_DNS.txt|c13.2_z0750_x10_KTH_DNS.txt]]||[[Media:c13.2_z0875_x10_KTH_DNS.txt|c13.2_z0875_x10_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x12_KTH_DNS.txt|c13.2_z0250_x12_KTH_DNS.txt]]||[[Media:c13.2_z0500_x12_KTH_DNS.txt|c13.2_z0500_x12_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x12_KTH_DNS.txt|c13.2_z0750_x12_KTH_DNS.txt]]||[[Media:c13.2_z0875_x12_KTH_DNS.txt|c13.2_z0875_x12_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x14_KTH_DNS.txt|c13.2_z0250_x14_KTH_DNS.txt]]||[[Media:c13.2_z0500_x14_KTH_DNS.txt|c13.2_z0500_x14_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x14_KTH_DNS.txt|c13.2_z0750_x14_KTH_DNS.txt]]||[[Media:c13.2_z0875_x14_KTH_DNS.txt|c13.2_z0875_x14_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x155_KTH_DNS.txt|c13.2_z0250_x155_KTH_DNS.txt]]||[[Media:c13.2_z0500_x155_KTH_DNS.txt|c13.2_z0500_x155_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x155_KTH_DNS.txt|c13.2_z0750_x155_KTH_DNS.txt]]||[[Media:c13.2_z0875_x155_KTH_DNS.txt|c13.2_z0875_x155_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x17_KTH_DNS.txt|c13.2_z0250_x17_KTH_DNS.txt]]||[[Media:c13.2_z0500_x17_KTH_DNS.txt|c13.2_z0500_x17_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x17_KTH_DNS.txt|c13.2_z0750_x17_KTH_DNS.txt]]||[[Media:c13.2_z0875_x17_KTH_DNS.txt|c13.2_z0875_x17_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x185_KTH_DNS.txt|c13.2_z0250_x185_KTH_DNS.txt]]||[[Media:c13.2_z0500_x185_KTH_DNS.txt|c13.2_z0500_x185_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x185_KTH_DNS.txt|c13.2_z0750_x185_KTH_DNS.txt]]||[[Media:c13.2_z0875_x185_KTH_DNS.txt|c13.2_z0875_x185_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x20_KTH_DNS.txt|c13.2_z0250_x20_KTH_DNS.txt]]||[[Media:c13.2_z0500_x20_KTH_DNS.txt|c13.2_z0500_x20_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x20_KTH_DNS.txt|c13.2_z0750_x20_KTH_DNS.txt]]||[[Media:c13.2_z0875_x20_KTH_DNS.txt|c13.2_z0875_x20_KTH_DNS.txt]]
|-
|[[Media:c13.2_z0250_x215_KTH_DNS.txt|c13.2_z0250_x215_KTH_DNS.txt]]||[[Media:c13.2_z0500_x215_KTH_DNS.txt|c13.2_z0500_x215_KTH_DNS.txt]]
|[[Media:c13.2_z0750_x215_KTH_DNS.txt|c13.2_z0750_x215_KTH_DNS.txt]]||[[Media:c13.2_z0875_x215_KTH_DNS.txt|c13.2_z0875_x215_KTH_DNS.txt]]
|}
<br/>
<br/>
 
==Reference DNS and  experimental  data:  mean  flow  and  turbulence evolution==
The following figures display and compare  the  reference  experimental  and
DNS database results (the present results can be  analysed  along  with  the
axial velocity contours shown in
[[UFR_4-16_Test_Case#figure7|Figs. 7]] &ndash;
[[UFR_4-16_Test_Case#figure9|9]] and [[UFR_4-16_Test_Case#figure13|13]]):
 
 
In order to get an impression about the mean flow structure  and  about  the
available reference DNS and experimental results
[[UFR_4-16_Test_Case#figure15|Figs.  15]]  and
[[UFR_4-16_Test_Case#figure16|16]]  display
the velocity field  development  in  the  vertical  central  plane  of  both
diffusers (''z/B=0.5''; ''B=3.33 cm'' is the width of the inflow duct)  typical  for
the flow in an expanding duct. The bulk flow exhibits deceleration,  leading
to an asymmetry of the velocity  profile,  particularly  so  for  the  axial
velocity  component.  The  effect  of  the  adverse  pressure  gradient  is
especially visible in the flow region along the upper  expanding  wall.  The
velocity profile approaches gradually the form characterizing  a  separating
flow, exhibiting regions of the zero velocity gradient  (at  separation  and
reattachment points) and profile inflection. The through-flow, that  is  the
flow in the positive streamwise direction, is characterized by  a  spreading
dictated by the pressure  gradient  arising  from  the  geometry  expansion.
Accordingly, the position of  the  reduced  velocity  maximum  is  gradually
shifted towards the upper wall, eventually reaching the  center  (''y/h=2'')  of
the straight outlet channel (with the height ''4h''). In this  post-reattachment
zone the velocity profile exhibits a fairly  flattened  form,  being  almost
symmetric.  The  consequence  of  the  velocity-profile  flattening  is  a
continuous monotonic  decrease  of  the  wall  shear  stress.  The  velocity
profile evolution is similar in  other  longitudinal  vertical  planes.  The
specific differences are related to the vicinity of both  bottom  and  upper
walls, especially the latter upper wall, where the flow  passes  regions  of
three-dimensional flow reversal.
 
 
<div id="figure15"></div>
{|align="center" width="750" cellspacing="0"
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure15a.jpg|740px]]
|-
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure15b.jpg|740px]]
|-
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure15c.jpg|740px]]
|-
|'''Figure 15:''' Diffuser 1 - Evolution of the  profiles  of  all  three  velocity components in the vertical plane  ''x-y''  at  the  central  spanwise  locations ''z/B=1/2'' obtained experimentally and by means of DNS
|}
 
 
<div id="figure16"></div>
{|align="center" width="750" cellspacing="0"
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure16a.jpg|740px]]
|-
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure16b.jpg|740px]]
|-
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure16c.jpg|740px]]
|-
|'''Figure 16:''' Diffuser 2 - Evolution of the  profiles  of  all  three  velocity components in the vertical plane  ''x-y''  at  the  central  spanwise  locations ''z/B=1/2'' obtained experimentally
|}
 
 
In the  top  part  of  [[UFR_4-16_Test_Case#figure17|Fig.  17 &ndash; upper]]
the  development  of  the  streamwise
turbulence  intensity  is  shown  for  diffuser  1.  The  lowest  turbulence
intensity is situated in  the  region  coinciding  with  the  mean  velocity
maximum &mdash; flow zone with approximately zero velocity gradient  &mdash;  along  the
entire diffuser section. The Reynolds stress profiles exhibit their  highest
values in the regions with the most intensive flow  deformation.  These  are
the near-wall layer in the attached-flow regions and  the  flow  zone  along
the  shear  layer  bordering  the  recirculation  zone.  The  peak  of  the
turbulence intensity originating from the boundary layer at the  top  inflow
duct wall increases initially,  after  the  strong  rise  in  pressure  (see
pressure coefficient development in
[[UFR_4-16_Test_Case#figure11|Fig. 11]]),  and  weakens  slightly  after
flow transition to  the  second  part  of  the  diffuser  section,  that  is
characterized by a decreasingly adverse pressure  gradient.  The  streamwise
turbulence intensity in the outlet duct is uniformly  distributed  over  the
cross-section.
 
 
<div id="figure17"></div>
{|align="center" width="750" cellspacing="0"
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure17a.jpg|740px]]
|-
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure17b.jpg|740px]]
|-
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure17c.jpg|740px]]
|-
|style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure17d.jpg|740px]]
|-
|'''Figure 17:''' Diffuser 1 - Evolution of the profiles  of  the  Reynolds  stress components in the vertical plane  ''x-y''  at  the  central  spanwise  locations ''z/B=1/2'' obtained experimentally and by means of DNS
|}
 
 
[[UFR_4-16_Test_Case#figure18|Fig.  18]]
shows  contour  plots  of  the  axial  velocity  component  at  five
streamwise cross-sectional areas in both  diffuser  configurations  obtained
experimentally indicating the evolution of the flow separation pattern.  The
recirculation-zone development displayed here can be  analyzed  in  parallel
with the quantitative information about the fraction of the diffuser  cross-
sectional area occupied by  the  reverse  flow  depicted  in
[[UFR_4-16_Test_Case#figure19|Fig.  19]].  The
adverse pressure gradient is imposed onto the intersecting  boundary  layers
along flat walls upon  entering  the  diffuser  section.  According  to  the
experimental investigation the  boundary  layers  along  all  walls  are  of
comparable thickness. The separation starts immediately after the  beginning
of the diffuser section at ''x/L = 0 (x/h=0)''.  The  onset  of  separation  is
located in the upper-right diffuser corner, formed  by  the  deflected  side
wall and the top wall, see e.g. the position ''x/h&nbsp;&asymp;&nbsp;2 (x/L=0.13)'' in
[[UFR_4-16_Test_Case#figure19|Fig.  18]].
Initial growth of this corner bubble reveals its spreading  rate  along  the
two sloped walls being approximately of the  same  intensity,  see  position
''x/h=5''. As the adverse pressure  gradient  along  the  upper  wall  outweighs
significantly the one along the side wall due to  the  substantially  higher
angle of expansion in diffuser 1,  11.3&deg;  vs.  2.55&deg;,  the  separation  zone
spreads gradually over the entire top wall surface, see position ''x/h=8''.  The
behaviour is different in diffuser  2.  There  one  notes  a  strong  three-
dimensional nature of the separation pattern. The maximum occupation of  the
diffuser cross-sectional area by the flow reversal, around 22% and  15%  for
the diffusers 1  and  2,  respectively
([[UFR_4-16_Test_Case#figure19|Fig.  19]]),  is  documented  at  the
position ''x/h=12-17 (x/L=0.8-1.13)''. The thickness of the flow  reversal  zone
in the diffuser 1 (its dimension in the normal-to-wall direction) is  almost
constant over the diffuser width in this region, resembling approximately  a
2-D pattern. After this position the intensity  of  the  back-flow  weakens.
The experimental results indicate that the reattachment  region  is  located
within the straight outlet duct,
[[UFR_4-16_Test_Case#figure19|Fig. 19]]. The  separation  pattern  and  the
differences between diffuser 1 and 2 can also be seen clearly  from  the  3D
plots given in
[[UFR_4-16_Evaluation#figure26|Fig. 26]], which were obtained by LES.
 
 
<div id="figure18"></div>
{|align="center" width="750" cellspacing="0"
|colspan="2" style="border: 1px solid darkgray;"|[[Image:UFR4-16_figure18.png|740px]]
|-
|style="border: 1px solid darkgray;" align="center"|'''Diffuser 1'''
|style="border: 1px solid darkgray;" align="center"|'''Diffuser 2'''
|-
|colspan="2"|'''Figure 18:''' Comparison between experimentally obtained  iso-contours  of  the axial velocity field in the cross planes ''y-z''  at  five  selected  streamwise locations within the both diffuser section (the thick line denotes the zero-velocity line). From [[UFR_4-16_References#7|Cherry ''et&nbsp;al.'' (2008)]]
|}
 
 
 
<div id="figure19"></div>
{|align="center" width="750"
|[[Image:UFR4-16_figure19.png|740px]]
|-
|'''Figure 19:''' Fraction  of  the  cross-sectional  area  occupied  by  the  flow reversal in both diffuser configurations. From [[UFR_4-16_References#7|Cherry ''et&nbsp;al.'' (2008)]]
|}
 
 
<br/>
<br/>
----
----

Latest revision as of 10:57, 13 December 2021

Flow in a 3D diffuser

Front Page

Description

Test Case Studies

Evaluation

Best Practice Advice

References

Confined flows

Underlying Flow Regime 4-16

Test Case Study

Brief description of the test case studied

The diffuser shapes, dimensions and the coordinate system are shown in Fig. 3 and Fig. 4. Both diffuser configurations considered have the same fully‐developed flow at channel inlet but slightly different expansion geometries: the upper-wall expansion angle is reduced from 11.3° (Diffuser 1) to (Diffuser 2) and the side-wall expansion angle is increased from 2.56° (Diffuser 1) to (Diffuser2). The flow in the inlet duct (height h=1 cm, width B=3.33 cm) corresponds to fully-developed turbulent channel flow (enabled experimentally by a development channel being 62.9 channel heights long). The L=15h long diffuser section is followed by a straight outlet part (12.5h long). Downstream of this the flow goes through a 10h long contraction into a 1 inch diameter tube. The curvature radius at the walls transitioning between diffuser and the straight duct parts are 6 cm (Diffuser 1) and 2.8 cm (Diffuser 2). The bulk velocity in the inflow duct is in the x-direction resulting in the Reynolds number based on the inlet channel height of 10000. The origin of the coordinates (y=0, z=0) coincides with the intersection of the two non-expanding walls at the beginning of the diffuser's expansion (x=0). The working fluid is water (ρ=1000 kg/m3 and μ=0.001 Pas).


UFR4-16 figure3.png
Figure 3: Geometry of the 3-D diffuser 1 considered (not to scale), Cherry et al. (2008); see also Jakirlić et al. (2010a)



UFR4-16 figure4.png
Figure 4: Geometry of the 3-D diffuser 2 considered (not to scale), Cherry et al. (2008).

Experimental investigation

Brief description of the experimental setup

The measurements were performed in a recirculating water channel using the method of magnetic resonance velocimetry (MRV), Fig. 5. MRV makes use of a technique very similar to that used in conventional medical magnetic resonance imaging (MRI), Fig. 6. Experiments were performed on a 1.5 Tesla magnet with resolution of 0.9 x 0.9 x 0.9 mm and a 7 Tesla magnet with resolution of 0.4 x 0.4 x 0.4 mm. Interested readers are referred to Cherry et al. (2008, 2009) for more details about the measurement technique.


UFR4-16 figure5a.png
UFR4-16 figure5b.png
Figure 5: Schematic of the experimental flow system (upper) and design of the 3D diffuser. Courtesy of J. Eaton (Stanford University)



UFR4-16 figure6.jpg
Figure 6: 3D diffuser arrangement in a medical magnetic resonance imaging device. Courtesy of J. Eaton (Stanford University)

Mean velocity and Reynolds stress measurements

Cherry et al. provided a detailed reference database comprising the three-component mean velocity field and the streamwise Reynolds stress component field within the entire diffuser section. Both diffuser configurations considered are characterized by a three-dimensional boundary-layer separation, but the slightly different expansion geometries caused the size and shape of the separation bubble exhibiting a high degree of geometric sensitivity to the dimensions of the diffuser as illustrated in Figs. 7, 8 and 9


UFR4-16 figure7.png
Figure 7: Streamwise velocity contours in a plane parallel to the top wall, from Cherry et al. (2008)


UFR4-16 figure8a.png
UFR4-16 figure8b.png
UFR4-16 figure8c.png UFR4-16 figure8d.png UFR4-16 figure8e.png
Figure 8: Measured streamwise velocity contours in the central plane (upper) of the Diffuser 1 and in the three selected cross-sectional slices positioned at different distances from the diffuser inlet (lower; their locations are denoted by thick white lines in the upper figure). Note that the black line indicates zero streamwise velocity and the purple and pink regions are reverse flow. The velocity values are normalized by the bulk inlet velocity being Vref=1 m/s. Courtesy of J. Eaton (Stanford University)


UFR4-16 fig9a.png
UFR4-16 fig9b.png
UFR4-16 fig9c.png UFR4-16 fig9d.png UFR4-16 fig9e.png
Figure 9: Measured streamwise velocity contours in the central plane (upper) of the Diffuser 2 and in the three selected cross-sectional slices positioned at different distances from the diffuser inlet (lower; their locations are denoted by thick white lines in the upper figure). Note that the black line indicates zero streamwise velocity and the purple and pink regions are reverse flow. The velocity values are normalized by the bulk inlet velocity being Vref=1 m/s. Courtesy of J. Eaton (Stanford University)


x/h=2 x/h=5 x/h=8 x/h=12 x/h=15
UFR4-16 fig10a.png UFR4-16 fig10b.png UFR4-16 fig10c.png UFR4-16 fig10d.png UFR4-16 fig10e.png
UFR4-16 fig10f.png
Figure 10: Measurements of turbulent streamwise stress components taken in cross-sectional slices of the diffuser 1 perpendicular to the mean flow. The region of highest turbulence (red areas; ignore the red areas at the bottom walls) follows the shear layer between forward and reverse flow. h=1cm represents the height of the inflow duct. Courtesy of J. Eaton (Stanford University)

Pressure measurements

In addition Cherry et al. (2009) provided the pressure distribution along the bottom non-deflected wall of diffuser 1 at different Reynolds numbers. Complementary to the Reynolds number 10000 (for which the entire flow field was measured), two higher Reynolds numbers — 20000 and 30000 — were also considered, Fig. 11. The surface pressure distribution was evaluated to yield the coefficient ; the reference pressure was taken at the position x/L = 0.05. The pressure curve exhibits a development typical of flow in diverging ducts. The pressure decrease in the inflow duct is followed by a steep pressure increase already at the very end of the inflow duct and especially at the beginning of the diffuser section. The transition from the initial strong pressure rise to its moderate increase occurs at x/L ≈ 0.3, (x/h=4.5) corresponding to the position where about 5% of the entire cross-section is occupied by the flow reversal (see e.g., Fig. 19. The onset of separation causes a certain contraction of the flow cross‐section, leading to a weakening of the deceleration intensity and, accordingly, to a slower pressure increase. The region characterized by a monotonic pressure rise was reached in the remainder of the diffuser section.


UFR4-16 figure11.png
Figure 11: Pressure recovery coefficients relative to the pressure on the bottom wall of the diffuser 1 inlet in a range of flow Reynolds number. L=15 cm represents the length of Diffuser 1, from Cherry et al. (2009)

Measurements uncertainties

(adopted from Cherry et al., 2008, IJHFF, Vol. 29(3))

Elkins et al. (2004) estimated the maximum relative uncertainty of individual mean velocity measurements to be about 10% of the measured value in a similar highly turbulent flow. However, comparisons to PIV in a backward facing step flow (Elkins et al., 2007) show that only a small percentage of MRV velocity samples deviate by that much and most are much more accurate. To test this, the streamwise velocity component was integrated over 250 cross-sections of the MRV data and the results were compared to the known volume flow rate. This indicated an uncertainty in the integral of less than 2% with a 95% confidence level.

Measurements of turbulent normal stresses in Diffuser 1 were also taken using the MR technique described by Elkins et al. (2007). This method is based on diffusion imaging principles in which the turbulence causes a loss of net magnetization signal from a voxel in the flow. This causes a decay in signal strength which can be related to turbulent velocity statistics. Elkins et al found this method to be accurate within 20% everywhere in the FOV and within 5% in regions of high turbulence. Three turbulence scans were completed using three different magnetic field gradient strengths. For each gradient strength, 30 scans were completed and averaged. The three averaged data sets were then averaged to obtain a final data set.

Experimental data available

Velocity, (and streamwise Reynolds stress components for diffuser 1) and coordinate data for both diffusers are available here.

Diffuser 1 data.zip (contains 1 file, 8.4 MBytes)
Diffuser 1 turbulence.zip (contains 1 file, 5.3 MBytes)
Diffuser 2 data.zip (contains 1 file, 4 MBytes)

The data are seven 3D matlab matrices. The x, y, and z matrices give the coordinates of each point in the coordinate system shown in Fig. 5. The units are meters. The Vx, Vy, and Vz matrices give the corresponding velocity components for each point in m/sec. The matrix mg gives the relative signal magnitude detected by the MRI machine.

Experimental data for the pressure coefficient in Diffuser 1 for the inflow Reynolds number Re=10000 are available here (Cp_Re=10000.xlsx). The coordinate system is the same as the coordinate system described in the corresponding manuscript (see below). is defined as , where is the pressure at x=0.05 (see Fig. 11) at the midpoint (z/B=0.5) of the bottom flat wall (opposite the wall expanding at 11.3 degrees), is the density, and is the bulk inlet velocity. The data were taken in a line along the bottom wall of Diffuser 1 at constant y and z coordinates. L (=15 cm) indicates the length of the diffuser.

Please acknowledge the authors of the experiment when using their database!

CFD Methods

The flow in the present diffuser configuration was intensively investigated computationally in the framework of two ERCOFTAC-SIG15 Workshops and in the European ATAAC project. The workshops focussing on both 3D diffuser configurations (denoted by SIG15 Case 13.2-1 and SIG15 Case 13.2-2, respectively) were organized at the Technical University of Graz, Austria in September, 2008 and the "LA Sapienza" University of Rome, Italy in September, 2009. The corresponding reports are published in the ERCOFTAC Bulletin Issues, Steiner et al. (2009) and Jakirlić et al. (2010b), see "List of References". Both in the workshops and in the ATAAC project a wide range of turbulence models in both LES and RANS frameworks as well as some novel Hybrid LES/RANS formulations have been employed. The computational database was furthermore enriched by the results of a Direct Numerical Simulation of the first diffuser performed by Ohlsson et al. (2010). The list of all computational contributions to the workshops including some basic information about the methods and models used and corresponding grid resolution is given in the following tables 1 and table 2. For more computational details, interested readers are referred to the "workshop proceedings" — see the corresponding links at the end of the Section "Evaluation of the results". The results from the ATAAC project and information on the methods used can be obtained through the links ATAAC_D3-2-36_excerpt3DDiffuser.pdf (excerpt from an ATAAC report) and [ATAAC_finalWorkshop_ST04-Diffuser-ANSYS.pdf] (PowerPoint presentation at ATAAC final workshop).


UFR4-16 table1.png
Table 1: SIG15 Case 13.2-1 (Diffuser 1) — contributors and methods (note that the DNS grid comprises 220 million cells in the follow-up work published in Ohlsson et al. (2010))
N.B. ITS is "Institut für Thermische Strömungsmaschinen"


UFR4-16 table2.png
Table 2: SIG15 Case 13.2-2 (Diffuser 2) — contributors and methods
N.B. ITS is "Institut für Thermische Strömungsmaschinen"


Direct numerical simulation of the flow in a 3D diffuser

(adopted from Ohlsson et al., 2010, JFM, Vol. 650)

In this part of the present contribution the DNS study of the flow in 3D Diffuser 1 performed recently by Ohlsson et al. (2010) will be described in more details. Ohlsson et al. participated with this contribution at the 14th SIG15 Workshop on Refined Turbulence Modeling (Jakirlić et al., 2010b). Accordingly, their results are also part of the CFD methods/models evaluation — along with the results obtained by different LES, RANS and Hybrid LES/RANS models (see the next chapter). In addition, as the DNS provided a very comprehensive database comprising all three mean velocity components (and associated integral characteristics such as surface pressure and friction factor) and all six Reynolds stress components as well as a certain insight into the physics (not detected by the experimental investigation) one can regard it also as a reference investigation, as we do presently.

The Direct Numerical Simulation of the Diffuser 1 was performed using a massively parallel high-order spectral element code. The incompressible Navier-Stokes equations are solved using a Legendre-polynomial-based spectral-element method, implemented in the code nek5000, developed by Fischer et al. (2008). The computational domain shown in Fig. 12 is set up in close agreement with the diffuser geometry in the experiment (see Fig.3) and consists of the inflow development duct of almost 63 duct heights, h, (starting at the non-dimensional coordinate x = -62.9), the diffuser expansion located at x = 0 and the converging section upstream of the outlet. The corners resulting from the diffuser expansion are smoothly rounded with a radius of 6.0 in accordance with the experimental set-up. The maximum dimensions are Lx =105.4 h, Ly =[h, 4h], Lz =[3.33 h, 4h]. In the inflow duct, laminar flow undergoes natural transition by the use of an unsteady trip forcing (see e.g. Schlatter et al., 2009), which avoids the use of artificial turbulence and eliminates artificial temporal frequencies which may arise from inflow recycling methods (Herbst et al., 2007). A 'sponge region' is added at the end of the contraction in order to smoothly damp out turbulent fluctuations, thereby eliminating spurious pressure waves. It is followed by a homogeneous Dirichlet condition for the pressure and a homogeneous Neumann condition for the velocities. The resolution of approximately 220 million grid points is obtained by a total of 127750 local tensor product domains (elements) with a polynomial order of 11, respectively, resulting in Δz+max≈11.6, Δy+max≈13.2 and Δx+max≈19.5 in the duct center and the first grid point being located at z+≈0.074 and y+≈0.37, respectively (note that at the time of the ERCOFTAC SIG15 workshop the DNS grid had 172 million grid points, see Chapter "Evaluation"). It was verified that this resolution yields accurate results in turbulent channel flow simulations. In the diffuser, the grid is linearly stretched in both directions, but since the mean resolution requirements decreases with the velocity, which decreases linearly with the area expansion, the resolution in the entire domain is hence satisfactory. The simulation was performed on the Blue Gene/P at ALCF, Argonne National Laboratory (32768 cores and a total of 8 million core hours) and on the cluster 'Ekman' at PDC, Stockholm (2048 cores and a total of 4 million core hours). Thirteen flow-through times, t Ub/L=13, based on bulk velocity, Ub, and diffuser length, L=15 h, were simulated in order to let the flow settle to an equilibrium state before turbulent statistics were collected over approximately Ubt/L=21 additional flow-through times.

In addition to the three-dimensional mean velocity field, all six Reynolds stress components were evaluated, as well as the surface pressure and skin friction distribution along the bottom wall.


UFR4-16 figure12.png
Figure 12: DNS grid of the diffuser 1 geometry showing the development region, diffuser expansion, converging section and outlet. From Ohlsson et al. (2010)


Fig. 13 illustrates the mean axial flow field obtained by DNS along with the experimental data and Fig. 14 depicts the skin friction evolution along the bottom diffuser wall at the midpoint (z/B=0.5; Cfwall / (0.5ρU2bulk) ). The latter DNS result was evaluated exclusively for the needs of the 14th ERCOFTAC Workshop; it is not part of the DNS database which can be downloaded here.


UFR4-16 figure13.png
Figure 13: Cross-flow planes of streamwise velocity component at 2, 5, 8 and 15 h downstream of the diffuser throat. Left: DNS. Right: Experiment by Cherry et al. (2008). Each streamwise position has its own colour bar on the right. Contour lines are spaced 0.1 Vref apart. Thick black lines correspond to the zero velocity contour. From Ohlsson et al. (2010)


UFR4-16 figure14.jpg
Figure 14: Friction coefficient at the bottom wall of the diffuser 1 with L=15 cm representing the length of Diffuser 1. The LES and HLR (Hybrid LES/RANS) results are from Jakirlić et al. (2010a) and John-Puthenveetil (2012).

Available DNS data (last update: 2010-11-23)

The entire digitalized DNS database (as well as some high-resolution images) as listed below can be downloaded from http://www.mech.kth.se/~johan/data/index.html. The links given below are to local wiki copies of the files.

This directory contains velocity profiles and mean fluctuations obtained from the DNS by Ohlsson et al., (JFM 650 307–318) This simulation was designed to match the experiment by Cherry et al., 2008, IJHFF 29(3). The data is given at the same streamwise and spanwise locations as during the 14th ERCOFTAC SIG15 Workshop on Turbulence Modelling, held in Rome, September 2009.

The full Reynolds stress budgets will be available in the future.

The data is free to use; please include a proper reference to the original publications.

In case of any questions, e.g. related to the data, or whether you wish for additional data not presented here, please contact Johan Malm, (johan@mech.kth.se), Philipp Schlatter (pschlatt@mech.kth.se) or Dan Henningson (henning@mech.kth.se)

Visualizations and computational mesh

Crossflow planes with instantaneous streamwise velocity

Crossflow planes with instantaneous streamwise velocity and isosurfaces of streamwise velocity

Instantaneous streamwise velocity in a spanwise midplane with some isosurfaces of instantaneous pressure

Mesh

Digitalized DNS database: full 3D mean velocity and Reynolds stress fields

Explanation:

  • E.g., c13.2_Ucont2_KTH_DNS denotes the files comprising the contours of the axial velocity - Ucont - at the streamwise position x/h=2 for the case 13.2 (ERCOFTAC SIG15 denotation). The same data are given at the streamwise locations x/h=2, 5, 8, 12 and 15
  • E.g., c13.2_urms2_KTH_DNS.txt denotes the file comprising the contours of the root-mean-square values of the streamwise stress component — urms — at the streamwise position x/h=2. The same data are given at the streamwise locations x/h=2, 5, 8, 12 and 15
  • The file c13.2_cp_KTH_DNS.txt denotes the file comprising the pressure coefficient distribution at the bottom flat wall at the midpoint z/B=0.5
  • The file c13.2_z0250_x-2_KTH_DNS.txt denotes the file comprising all three mean velocity components, kinetic energy of turbulence and all six Reynolds stress components at the streamwise position x/h=-2 (inflow duct) and the spanwise position z/B=0.25. The same data are given at the streamwise locations x/h=0, 2, 4, 6, 8, 10, 12, 14, 15.5, 17, 18.5, 20 and 21.5 at the following spanwise locations z/B=0.25, 0.5, 0.75 and 0.875.


c13.2_cp_KTH_DNS.txt
c13.2_Ucont2_KTH_DNS.txt c13.2_urms2_KTH_DNS.txt
c13.2_Ucont5_KTH_DNS.txt c13.2_urms5_KTH_DNS.txt
c13.2_Ucont8_KTH_DNS.txt c13.2_urms8_KTH_DNS.txt
c13.2_Ucont12_KTH_DNS.txt c13.2_urms12_KTH_DNS.txt
c13.2_Ucont15_KTH_DNS.txt c13.2_urms15_KTH_DNS.txt
c13.2_z0250_x-2_KTH_DNS.txt c13.2_z0500_x-2_KTH_DNS.txt c13.2_z0750_x-2_KTH_DNS.txt c13.2_z0875_x-2_KTH_DNS.txt
c13.2_z0250_x0_KTH_DNS.txt c13.2_z0500_x0_KTH_DNS.txt c13.2_z0750_x0_KTH_DNS.txt c13.2_z0875_x0_KTH_DNS.txt
c13.2_z0250_x2_KTH_DNS.txt c13.2_z0500_x2_KTH_DNS.txt c13.2_z0750_x2_KTH_DNS.txt c13.2_z0875_x2_KTH_DNS.txt
c13.2_z0250_x4_KTH_DNS.txt c13.2_z0500_x4_KTH_DNS.txt c13.2_z0750_x4_KTH_DNS.txt c13.2_z0875_x4_KTH_DNS.txt
c13.2_z0250_x6_KTH_DNS.txt c13.2_z0500_x6_KTH_DNS.txt c13.2_z0750_x6_KTH_DNS.txt c13.2_z0875_x6_KTH_DNS.txt
c13.2_z0250_x8_KTH_DNS.txt c13.2_z0500_x8_KTH_DNS.txt c13.2_z0750_x8_KTH_DNS.txt c13.2_z0875_x8_KTH_DNS.txt
c13.2_z0250_x10_KTH_DNS.txt c13.2_z0500_x10_KTH_DNS.txt c13.2_z0750_x10_KTH_DNS.txt c13.2_z0875_x10_KTH_DNS.txt
c13.2_z0250_x12_KTH_DNS.txt c13.2_z0500_x12_KTH_DNS.txt c13.2_z0750_x12_KTH_DNS.txt c13.2_z0875_x12_KTH_DNS.txt
c13.2_z0250_x14_KTH_DNS.txt c13.2_z0500_x14_KTH_DNS.txt c13.2_z0750_x14_KTH_DNS.txt c13.2_z0875_x14_KTH_DNS.txt
c13.2_z0250_x155_KTH_DNS.txt c13.2_z0500_x155_KTH_DNS.txt c13.2_z0750_x155_KTH_DNS.txt c13.2_z0875_x155_KTH_DNS.txt
c13.2_z0250_x17_KTH_DNS.txt c13.2_z0500_x17_KTH_DNS.txt c13.2_z0750_x17_KTH_DNS.txt c13.2_z0875_x17_KTH_DNS.txt
c13.2_z0250_x185_KTH_DNS.txt c13.2_z0500_x185_KTH_DNS.txt c13.2_z0750_x185_KTH_DNS.txt c13.2_z0875_x185_KTH_DNS.txt
c13.2_z0250_x20_KTH_DNS.txt c13.2_z0500_x20_KTH_DNS.txt c13.2_z0750_x20_KTH_DNS.txt c13.2_z0875_x20_KTH_DNS.txt
c13.2_z0250_x215_KTH_DNS.txt c13.2_z0500_x215_KTH_DNS.txt c13.2_z0750_x215_KTH_DNS.txt c13.2_z0875_x215_KTH_DNS.txt



Reference DNS and experimental data: mean flow and turbulence evolution

The following figures display and compare the reference experimental and DNS database results (the present results can be analysed along with the axial velocity contours shown in Figs. 79 and 13):


In order to get an impression about the mean flow structure and about the available reference DNS and experimental results Figs. 15 and 16 display the velocity field development in the vertical central plane of both diffusers (z/B=0.5; B=3.33 cm is the width of the inflow duct) typical for the flow in an expanding duct. The bulk flow exhibits deceleration, leading to an asymmetry of the velocity profile, particularly so for the axial velocity component. The effect of the adverse pressure gradient is especially visible in the flow region along the upper expanding wall. The velocity profile approaches gradually the form characterizing a separating flow, exhibiting regions of the zero velocity gradient (at separation and reattachment points) and profile inflection. The through-flow, that is the flow in the positive streamwise direction, is characterized by a spreading dictated by the pressure gradient arising from the geometry expansion. Accordingly, the position of the reduced velocity maximum is gradually shifted towards the upper wall, eventually reaching the center (y/h=2) of the straight outlet channel (with the height 4h). In this post-reattachment zone the velocity profile exhibits a fairly flattened form, being almost symmetric. The consequence of the velocity-profile flattening is a continuous monotonic decrease of the wall shear stress. The velocity profile evolution is similar in other longitudinal vertical planes. The specific differences are related to the vicinity of both bottom and upper walls, especially the latter upper wall, where the flow passes regions of three-dimensional flow reversal.


UFR4-16 figure15a.jpg
UFR4-16 figure15b.jpg
UFR4-16 figure15c.jpg
Figure 15: Diffuser 1 - Evolution of the profiles of all three velocity components in the vertical plane x-y at the central spanwise locations z/B=1/2 obtained experimentally and by means of DNS


UFR4-16 figure16a.jpg
UFR4-16 figure16b.jpg
UFR4-16 figure16c.jpg
Figure 16: Diffuser 2 - Evolution of the profiles of all three velocity components in the vertical plane x-y at the central spanwise locations z/B=1/2 obtained experimentally


In the top part of Fig. 17 – upper the development of the streamwise turbulence intensity is shown for diffuser 1. The lowest turbulence intensity is situated in the region coinciding with the mean velocity maximum — flow zone with approximately zero velocity gradient — along the entire diffuser section. The Reynolds stress profiles exhibit their highest values in the regions with the most intensive flow deformation. These are the near-wall layer in the attached-flow regions and the flow zone along the shear layer bordering the recirculation zone. The peak of the turbulence intensity originating from the boundary layer at the top inflow duct wall increases initially, after the strong rise in pressure (see pressure coefficient development in Fig. 11), and weakens slightly after flow transition to the second part of the diffuser section, that is characterized by a decreasingly adverse pressure gradient. The streamwise turbulence intensity in the outlet duct is uniformly distributed over the cross-section.


UFR4-16 figure17a.jpg
UFR4-16 figure17b.jpg
UFR4-16 figure17c.jpg
UFR4-16 figure17d.jpg
Figure 17: Diffuser 1 - Evolution of the profiles of the Reynolds stress components in the vertical plane x-y at the central spanwise locations z/B=1/2 obtained experimentally and by means of DNS


Fig. 18 shows contour plots of the axial velocity component at five streamwise cross-sectional areas in both diffuser configurations obtained experimentally indicating the evolution of the flow separation pattern. The recirculation-zone development displayed here can be analyzed in parallel with the quantitative information about the fraction of the diffuser cross- sectional area occupied by the reverse flow depicted in Fig. 19. The adverse pressure gradient is imposed onto the intersecting boundary layers along flat walls upon entering the diffuser section. According to the experimental investigation the boundary layers along all walls are of comparable thickness. The separation starts immediately after the beginning of the diffuser section at x/L = 0 (x/h=0). The onset of separation is located in the upper-right diffuser corner, formed by the deflected side wall and the top wall, see e.g. the position x/h ≈ 2 (x/L=0.13) in Fig. 18. Initial growth of this corner bubble reveals its spreading rate along the two sloped walls being approximately of the same intensity, see position x/h=5. As the adverse pressure gradient along the upper wall outweighs significantly the one along the side wall due to the substantially higher angle of expansion in diffuser 1, 11.3° vs. 2.55°, the separation zone spreads gradually over the entire top wall surface, see position x/h=8. The behaviour is different in diffuser 2. There one notes a strong three- dimensional nature of the separation pattern. The maximum occupation of the diffuser cross-sectional area by the flow reversal, around 22% and 15% for the diffusers 1 and 2, respectively (Fig. 19), is documented at the position x/h=12-17 (x/L=0.8-1.13). The thickness of the flow reversal zone in the diffuser 1 (its dimension in the normal-to-wall direction) is almost constant over the diffuser width in this region, resembling approximately a 2-D pattern. After this position the intensity of the back-flow weakens. The experimental results indicate that the reattachment region is located within the straight outlet duct, Fig. 19. The separation pattern and the differences between diffuser 1 and 2 can also be seen clearly from the 3D plots given in Fig. 26, which were obtained by LES.


UFR4-16 figure18.png
Diffuser 1 Diffuser 2
Figure 18: Comparison between experimentally obtained iso-contours of the axial velocity field in the cross planes y-z at five selected streamwise locations within the both diffuser section (the thick line denotes the zero-velocity line). From Cherry et al. (2008)


UFR4-16 figure19.png
Figure 19: Fraction of the cross-sectional area occupied by the flow reversal in both diffuser configurations. From Cherry et al. (2008)





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

Front Page

Description

Test Case Studies

Evaluation

Best Practice Advice

References


© copyright ERCOFTAC 2024