Slide Training 1 PDF
Slide Training 1 PDF
This software and related documentation are proprietary to Siemens Product Lifecycle Management Software lnc.
Trademarks
Siemens and the Siemens logo are registered trademarks of Sjemens AG. Teamcenter is a trademark or registered
trademark of Siemens Product Lifecycle Management Software lnc. or its subsidiaries in the United States and in other
countries. All other trademarks, registered trademarks, or service marks belong to their respective holders.
STARCCM8 C |d S M S
Description
! Launch STAR-CCM+.
i. In the menu, select File > Load and browse to the folder lnputFiles >
01 Essentialconcepts.
qy` Select mixer.sin and click Open.
I Right-click the Scenes node and select Open All Scenes.
EE--_
xpeb®/ev:froimfti:d><v*xpndg&e>qu#>b«rffl.)zr®+rAadeiAiFouZ3r"IqucArfe:ov`
Which nods ih the tree makes the connection between mesh and physics?
Client-Gul
Server -Computations
--_.-` client-Server
Reconnected
No client is
client and server connected to the Server
running server
Page 8 CTO9101_CM12_1
Page9 CTO9101 CM121
smlt I MNS
gr a Geomem/
S ii Continua
giv ® A,r
` H Partshle5hes
er ® stag,
fr ii Regions
Contains all of the simulation F # F,Bid
giv a Boundaries
objects in the .sim file. giv Gi Feature curves
giv ii Phvs[cs conditions
ah a Physics Values
er ii lntgrface5
objects/nodes are created in the gr ii Derived Parts
gr ii So[v€rs
simulation. a- ii Stopping Criteria
E Solutior\ Histories
tr ii Solution vie:us
gr ii Reports
er Ea Monitors
tr ji Plots
tr a Scenes
-Gi Summaries
tr ii] P`epresentations
th E] Tools
cL I De/eL.ti !FluLc!/solid!
c> rm Outl¢,
9 I Punrler|
Properties can be modified ~ jE Ph/si£=Co/}drt[or)5
i iB Ph'is/I;`Jalug3
Probe
|R + Reset iF:ii?i-;I-.j[8__; =F_6_T_::i_l=_`|-:.lil-:1:i:i ffigH
Commonly used actions have shortcuts in the Toolbars at the top of the Gul.
Page 15 CTO9101_CM12_1
Tin lgy a d M S
•dll I(I
il
So I
-`(a.&:f?- ,.;.=.-,-.-
[i A TrT+.-- T+i-i
rl ! i
I • -d I I a
) Ill
ill
.IJli i i Solid-Solidlnterface I
.1\1\`,,+,
/,;,(/ `
-` ``\`
,`.
I•("t`;,'i,+ -\'1''
i.\\., ` %
i
_ Wall, Fluld-Solid Interface
II( .`\
-
i II
T_TT_ I
-+---.i-
i;
)
i
ir . ) ---
++`;.-i ri T-- ! i ( ,n,et I
Page 16 CTO9101_CM12_1
a a continua
- H Parts Meshes
A physics continuum is a collection of models
=- , Physics 1 that represent a single medium (fluid or solid).
3 a Models © Model properties, methods, equations and their
'i) Constant Density
3 a Gas coefficients can be edited in continua
3 a Air © Each physics continuum contains Reference
=r a Material Properties
`3 * Density Values and Initial Conditions.
• Constant
r3 * Dynamic VISCosity
• Constant
- jJ Gradients I
z i Reference Values
t,{,|,;# ;rm,.I __ ( i---i-.-.`.-`.-
3--i Initial Conditlon5
*` --:. fr; -
3*, . Physics Steel i
ex
Ei , Solid
)
i
-
? a Reperlg
a Eff({!enc`/_percent
3 lnletTdea)Pressure
a iriputpowor
a outjetMa5SPlowR@t9
a outletTotdlpre$3ure
a out!?V\/olumeFlowpate
a Outler`,'olumeF)owRateln
a OutputpQyver
a Pres3urefiati`]
-a Torque
a Tott]l Pressure Bige
? ii Monttorg
- B condnufty
* a EfficieHey
a lteration
giv a. Mass Flow
a Phy3ical Time
ri~ a, pressure rat,a
ae B X-momentum
What are the major steps required to simulate the aerodynamic behavior and
compute drag on the car shown in the picture?
I... `]
Click OK.
lrfuso
:=j ue I Power scs5rm I;' Poweron.Denrend
mprtg mty M S
@ sdv?
Mesh. E SaveAs,.
a SaveAll
a Autosave"
`.I Browse to the folder 02_Workflowcar,
select both parasolid files and click Open.
t jii Parts
+® car
ch a wheel
a All parts appear under the Geometry > Parts folder after tr 6 wheel 2
a- ® wheel 3
i_mport. a- ® wheel 4
op t sbt I M
The Air part is now just the wind tunnel air around the
car, and it has inherited the part and surface names
from the original parts.
;?`' ffi Pan
•£ ® gr .
:?; ffi tr .
•£- iE ines
¥;. iE b-Parts
p tprt d M
Open a new Empty Scene using the Create/Open Scenes icon in the toolbar.
Drag and drop the Air part into the new scene.
I Click the Make Scene Transparent icon in the toolbar.
a VIOw
:i- J* Update
H Anlmatlon'
EL Axes
i., il Lights
i, aB cllp planes
3) a Annotations
LE e> v Edit„.
Df b dytyp M
wl nd+unnol.Symmetry . Proportlos
Note the icons changed to indicate the boundary type. I-'Properttes
a Geomtry
ln the Geometry > Parts node, right-click I 3D-GAD Modets
J I Parts
1 ® car
•:;.
Select the meshers -Co Detoch
.®OEas
` Surface Remesher, Boctlean
:`'fiiiii'...``9\-¥`t;rxptmutmesE(aaeBbvLf`.,-
aady for 2D Mashing
Trimmed Cell Mesher, Autorwhed Mech (20)
V®luma E;I tludct
Dltec[ed Mesh
\'\ Prism Layer Mesher.
Click OK.
E Cancel Help
Df lt t I M
il operations
Set the default mesh settings for the automated 1 © subtract
-, qb Automated Mesh
meshers. tr iii Meshers
't Expand the Operations > Automated Mesh > Ei Default Controls
• I;EEENFH
• CAD Projectlon
Default Controls node. • Target Surface Slze
• Mlnlmum Surface Size
+ Select the Base Size and in the Properties • Surface Curvature
• Surface Proxlmlty
window set Value to 0.25 in. • Surface Growth Rate
• Number of Prism Layers
L~ Set the Prism Layer Total Thickness value to • Prism Layer Stretching
• Prism Layer Total Thickness
10 (% of the Base Size). • Maxlmum Core/Prlsm Transition Ratio
• Volume Gro\^th Rate
• Maxlmum Cell Slze
Value 0.25 in
E Geometry
- ii-3D-CADModeis
The exclamation mark disappears and
a new Continua > Parts Meshes node appears.
Page 35 CTO9101_CM12_1
mesh on the Symmetry Plane, click the Save-Restore-Select LqD Qpen All scenes|
Views icon in the toolbar and select Views > -X > Up +Y.
' _-
I
Proj€`t!ori Mode
+z
-Z
`:
,? up-y
Lz
Page 36 CTO9101_CM12_1
phy mdl sM
a dy dt M
= iB Regions
The speed of air in the tunnel is set at the inlet '=. H Air
M ethod Consta nt
Value 50.0 mph
Dimensions
a ii Regions
a Derived Parts
E3 a solvers
¥ a Stoppin Criteria
h Stop File
DtA ly Rpt SM S
i` gB Geometry
Create a report to compute drag on the car. +. iii Continua
A plot opens to view the value of the drag during the simulation, plotted versus
iteration.
=a.gReff=¥____.__.`_~_
±E Mo. RunReport
I -Drag Mormon I
:.IE plo. Highlight
f a sce Edit„.
ia sur SetNewTag.„
.+3 il Ref
jl ii TO(
Create Monitor from Report
Create An.nc)tation from Repc)rt
Copy
Paste
Delete
Rename...
Dependeiicles
ErrorLocj
DtA ,y S M
=T tip Reglons
:-, . Alr
•3; i Boiindaries
iE . wind tunnel,Floor
i * wind-tilnnel,Inlet
i. 11111 wind funnel.Outlet
t I wlnd-tunnel.Sldewall
I .diThffl -.
£i . wind_tunnel.Top
fij i Feature Curves
(a a Physics Condltion$
33 a Physlcs Values
ei Derived Parts
L± 5i Regions
E I Air
E} ii Boundaries
5j . car,Bedy
`fl I wheel 2.Faces
3 I wheel.Faces
z! I wind tunnel,Floor
zj * wind tunnel,Inlet
¥i 11111 wind tunnel.outlet
z * wind tunnel.Sidewall
a . wind_tunnel,Top
3 ii Feature Curves
a a Physics Condltions
T±-: a Physics Values veloctry{m/5}
Ei Derived Parts
The goal of our simulation was to compute the drag on the car. Check the final
predicted value.
i Check the Drag Monitor Plot if the value has
Right-click the Drag report, then select Run Report. The value is printed to
the outD.ut window.
I___
iiiiiiiiiiii
I io!d § fife
open the starting file. ESTAR-CCM+
Type. Simu-fatlon
New,.. Ctrl+ N
Reload 5imulation...
Launch STAR-CCM+.
-\ From the File menu, select Load.
\ Choose to run Parallel on Local Host with the
default 2 Compute Processes.
- Browse to the folder 02 WorkflowTube and load
CoolingTube_Start.sim. ~
' Accept default settings and click OK.
a ti] Regions
Create a new Geometry scene and review the ¥ . CoollngTube
€] a Boundaries
tube geometry- i InletEnd
' Expand Regions > CoolingTube > Boundaries. - OutletEnd
I Wall
A: Select lnletEnd and change type to Velocity & ii Feature Curves
a Derived Parts
Inlet.
Change OutletEnd to Pressure Outlet. OutlotErid - Properties x
=Properfies
Index
lmerfaces
Wall is kept as boundary type Wall. Pa rt siirfaces [Coo ll n gTube. Outl etEnd]
Atmtdm hp I SM S
BB-8
CoolingTube to Input Parts. % il Derutconbds
i Custom Gonbds
.._: Expand the Default Controls node. a Contfro
t,\ Select Base Size and in Properties set Value to a Redoma
\=i I cgivThrfe
(3 iD Boundaries
0-001 in. d ZrbetErd
` Generate Mesh by right-clicking Automated Mesh lw OudetEnd
1wal
¥ iB FeatLrecurus
and selecting Execute or clicking the Generate ii oenved pats
a §tx~ cntEria
Volume Mesh icon in the toolbar.
•=.Jptopun
per.Pert lt®shhg
(coongmde]
Mesh. Report
uriked vlow
AJ1'mat/i,r)
7afordert
BOGpedoutdur
e. Turbulent, ffruthaftyrm ECchstan.Oenatv
HwidCied EZG.afro
HTgbu"ycos]tyl.s®scefro E5stw
9. Segregated Fluid Temperature. dctlQuaftyRen"un EaThaDinmut
dR-tin
4. Click Close.
a_Ei
I-i-
a dy dt M S
a ii Physics Values
Expand the Physics Conditions node for a3 lil Feature Curves
Heat Flux.
In the Physics Values node select Heat Flux
and enter the value 30,000 W/m^2.
stppg I sM S
Enabled E
Maximum Ste s250
Logical Rule Or
Criterion Satisfied
Tags
„ Wall
region.`
* Rename the report to Maximum
Temperature.
t Right-click the Maximum Temperature
Simulation Sccoe/Pf
Adjust plot font size and line width. € Maxlmilm Temperfure Monitor Plct
I Switch to the Scene/Plot tab and in the = Ei Data Series
i_L±ratL,r,ezM`,:n,tor
Properties window, click in the text field Title • Symbol style I.r,
3 ii Axes
Font and replace Plain 14 with Bold 28. a. ii. Bottom Axis
R th mlt M
` Switch back to the Simulation tab, drag Maximum Temperature plot from the
`'``d{
-Maxi in u in Temperature
Temperalure (C)
-¥.1# 33.7S4 41.8QD
R wq I 1 SM S
After the simulation reaches its Stopping Criteria, study the Maximum
Temperature plot and Wall Temperature scene. Describe below what you think
has happened physically in this model.
At the beginning of the simulation, what is the wall temperature associated with a
heat flux of 30,000 W/m^2? Where in the simulation can this information be
found?
What are some ways in which accuracy for this simulation could be assessed?
I..---....---.-
Page 68 CTO9101_CM12_1
lt d I M S
i::--i-:i_-:-_I--_--:ir-ffi ¥T--
` `,-;:-.i-+= ,------- :--` '= .i----
Sklltb1 d dp I d M S
a Connect to server..
: lGBT.x_b, Receiit Files
Mac,o
mptg mty M
Select OK.
\ Save the simulation file as Geometry.sim.
Restore the saved view View 1 from the Save-Restore- + @ a ill g8 a ts RE I-' EL S \
ch kt Ill SIvl
:::::i-p.'.`On! Ea;:".
iii Contacts
GiE Operations
;'^ Ensure both Geometry Parts are included and select OK.
~\ Run the diagnostics by selecting
-399-
____ _______
•-.-+
EN. q # EN
Ea Reset DBpbyed
ifenage-.
Discussion Point
\/Vhat is causing the pierced faces?
R wdg I it SM S
lmprintl
a i Coritact5
9 Exp€n
g a Operations
er a ,.Ftjri`*`S
iii Continua
E Regions j Edit..
ai Derived Parts ;, Set N€wTag< ,
giv a stopping CmerE=.-~~~ €1,I-i-\=
Ga Solution Histor;
gr i SalutlDn
6k a Monitors
t* a Plots E=.i_I
The fluid part can be extracted from the solid geometry using
operations.
a Fill the holes.
® Extract the internal volume.
Extract Volume operation requires input part(s) that bound the volume to be
extracted .
Upon Creation.
`,I Click OK.
[=i-i-i FIir_i
Page81 CTO9101 CM121
optM
Check the Filled Holes part in the scene by selecting the part in the tree. The
new surfaces are highlighted.
Scene/Plo( (
qE| Geom€tr/_start
S a Cfometry
6ii 3D-CAD M®dels
f a Parts
a- e3 ch\\If 'pof ee'<
a- a H€atsink
ar a ,CBT
giv GE De5Griptic]n§
ct. E] Cantacts
Ch I Operations
i Cominua
Ei P-egion5
Gil Derived parts
tr ia Stopping Criteria
a §oliltion Histories
tr a solutionvieus
a. a Monitors
i> i Plots
fr a `ScenE3,.
ii Summaries
qu a Representatioris
fr ii Tools
H.r,:cii+(=Eir
op I M S
giv®fflLRE
OL a F,ii Holes
dr a Heat5ink
E==
a. ii Descripticns
gr a Comacts
giv ii operalions
a =ontinuai
i Regions
i Deriyed parts
oj` a S`opping Critgria
a Solutiori His`ories
tr i Salu`iorivlev\c
th ii I,'onitors
a- i Plots
- a Scenes
G] Summaries
a a Pepre3en{attoris
gr GB Tools
' i coma[t5
a Gil Fill Holes/Fluid
Discussion Point \ tr i Heatslnk/Fluid
ch kt Ill M
Click OK.
9 ii Parts
: i i,:!!u;rj',ensk
3 Of 4 sethed
Bouridary-node fnteife6;Froriconfaa;A ~ ~ -y
_irpivm E .inHS -
HeatsirikFaceg|Fluid/H€atsirik]
Mass Flow Inlet. tr ii Feature Curtye5
What steps were compl'eted before the fluid volume could be extracted?
mp tf, typ M
When to import . , . (,
3D.CAD parts
• Removedetail. • CAD modifiication is not
• Add missing surfaces or build iequired.
missing parts. • Using a surface tessellation
• Automate design changes. data format.~
If auto.repair cannot
I.emedy the issue, use:
• 3D-CAD
• SurfaceRepair
rfugS j& rfu A
• Source CAD
package
F8ceg 0 ?
@.\¥/a ffiF S /
EH RE,
Ch kgf Sf p M
Face's edges pierce a different face. Measure of a face. Perfect face = The ratio of the distance to the
equilateral = 1.0. nearest neighbor face to the size of
the face. j\ ^{\
Quality = A,/Al
Close proximity
faces
/' Self-intersecting
faces
Query
God rco8
Organize
and #.J!® ca @)®
F- Bgiv
&-i,_±:JL:,€_¥_
9, ,€'€tt¢d A\'.` ,
D,,i,,,'.a _®_-i
J`"J
Set Control:
Defines collections of faces, edges, or vertices.
E.=Ee_±:I:==±-=[:EE±
Action Store \
Part Edit
Assigns faces, edges, and vertices to
'f£' Part surface r=i pait curve
part surfaces, contacts, curves, or points. C`:` Pert surf@ce contact t...r:I part point
Edx OptoJ)s
New partsurface T
pt,pt f g I M
Organization and clear labeling needed for efficiency and control of the Mesh,
Physics and Post-processing.
c t I M S
Operations are defined actions for part and part surface manipulations.
updated or replaced,
© Geometry processing and meshing steps available.
OP t T fin SM S
A single transform operation can have multiple transformations which will all
-tit .y--Lan,cecoou,mo
System
Surface
Extruder
OP I FIIHI Et tvim SM S
op I ,mp I M
R , g SM S
If you see. . .
± Surface
Fcafure
Pierced faces Non-manifold edges Free edges Active Model -
._ []-
J I-£alL|..u
±ai,' :i.TJ
¥ iiiiiEiiE
-? EiiiEiiiiii
rrd « ,Hill
-I Reset Vlew
8 Try imprinting
Try extracting the Try filling the i- Resct Displayed
OP I Sf Wpp SM S
Limitations:
® Loses CAD connectivity.
© Defeatures geometry smaller than specified size.
A gprtt g M
=; L*
-Fhr;:-A-r|iT
.rfu _,
r+EE±isiri;
T±EELITi_I
Defines how to create region level objects from part level objects.
ee One-to-One.
® Many parts level objects in a single region level object.
@ Assign part level object to existing region level object.
( .y,~ i , (-y
R wq I SM S
•r-
r±| F-
The starting point is the sim file saved at the end of the geometry workshop.
If it is not open, load simulation Geometry.sin from the 04 Meshing folder and
open Geometry Scene 1.
C I tmtdm hp I M
Select Mesher9
:`p#:::]a::Tuc#®#::poafr::,#ea:eM=9::rsh.;
Optional Volume Mo8her8 :
representation.
Turn on t`he mesh for all parts displayed in
the scene: ill RE + B + se S fa
H Efl ill
§::::: tch'eeaMre:::::aE::taMnecsehteoso,i:atrh]:::?'bar. i ng H
Discussion point
\/Vhat is an `appropriate base size?
© Consider the smallest geometry features,
© Compare to the overall size of the geometry.
Df lt I I M
Discussion Points
a The base size is 10 mm. Are the surface sizes (face edge lengths) 10 mm
everywhere?
® \/Vhere are the surface sizes larger or smaller than the base size?
Use the measuring tool and estimate a thickness from the pipe diameter.
D = 7 mm
0.15*D = 1 mm
Discussion point
Why define the prism layer total thickness as an absolute value?
p I ,mm h M S
Open the mesh scene and add two section planes to show the mesh interior.
I Open Mesh Scene 1.
~ry , Create a Section > Plane from the Derived Parts node menu.
a ia Geometry
a continua
td iE Regions
`£' ii Interfaces
Cell Surface...
as Iso5iirface...
Paste Ctrl + V
S-BS
Edit...
iiR
Refresh
faM Cylinder`..
-@BP
New Group Constrained Streamline... Sphere...
EE- D S Group By Vortex Core... Constrained Pla`ne...
Gii9 Separation/Attachment Line„. Arbitrary...
I P t lmm h SM S
Coordinate System
Display
a No Dlsp!ayer
•r-.rd..#*...-.;-,...,,^,-,I.-:-..--.-.A-``:+-//,.lf##:!
® How do the surface mesh sizes influence the volume mesh sizes?
a Which boundaries in the model do not have prism layer cells? Why?
p h k M S
~ fosition Vector
Vertices -position vector in space.
Faces -ordered collection of vertices that define a
surface.
Cells -ordered collection of faces that defies a
closed volume in space. 'ceFace
Meshing divides a single surface or volume into
smaller di;crete surfaces (faces) or volumes
(cells) on which discretized equations can be solved.
CADIN h M S
??- ii G-try
Base Size ii soee arfuife
reference size for other parameters. &t iE Parts
`.&i H Gesgifro
Target Surface Size ii entaRE
edge length desired given no other constraints. ri{ iH giv8be
.=u.i # futrmrdun
Minimum Surface Size ¥4 il Wtebe
smallest edge length allowed. • surhaR±
:y£`# iB H€anentrds
Surface Curvature - . aaeeife
number of `points around a circle. * Cenprq-
a Target Surfime 5iae
Surface PFox'jmity * aArfro surfro sise
number of faces across gaps. • 5rfecgivadre
• sufrfe*®ffi
Surface Growth Rate / . SurifeGrmthifete
aB -Emtrds
growth rate of triangles on surface. ¥ 5H Cchfro
Srf in h I S Ivl s
Other parameters may force edge length to be smaller, down to minimum size.
Influencing parameters:
© Surface proximity,
a Curvature refinement,
® Custom controls.
A high quality volume mesh quality is essential for an accurate, stable, and
converged solution.
Vlmm h I I M S
Polvhedral Mesh
Preferred for flows with high swirl or rotation.
Numerically superior to equivalent tetrahedral mesh.
Preferred wh.en conforma[ mesh desired (multi-
domain).
Tetrahedral Mesh
Solid mechanics or comparisons with legacy
tetrahedra[ models.
i i n ul ;( , i v,,
p mLy M
Tbl tb dryly M
a Gtaetry
Number of ca soeeb4edefa
prjsmLayers EEE5H;a;; \ffi a Par ts
`€i a Desongtxma
a Cenife
gal a apt#atus
a ife Ardmatirdun
Prism Layer •vg,`; iE -
Stretching al fifeerfuer
• Baeesde
• hLde ofprcoley-s
• Prtm laytr soetdrg
Prism Layer • PrmLarer TotH!Th.dmees
i i i i i i a ifeBer"
Total Thickness !!i
iB Cb5fro entrds
/,,,,,
!ii!!E
ib Conde
Advancing Layer
Name the three basic unstructured volume mesh types in STAR-CCM+ and
describe an application suited to each.
J II
•;( \\
:j' ,\.i..:!!
I,{` 1- `-`<_-
I,ii;y; =
SIEMENS
lntroduction
This workshop covers analyzing the volume mesh and making adjustments to
models and input values as necessary.
The starting point is the sim file saved at the end of the Meshing I Workshop.
If it is not open, load simulation Meshl.sim from the O4_Meshing folder and
Open All Scenes.
lv
q,ailagefi*ni s(}ttwn/s in
Page 183 CT09101_CM12_1
SIEMENS
lnspect volume mesh
Reduce the target surface size and observe the effects of 4't.,
t.
'
t! " ',,
the change. .L . ,,.,<
:r \1" .l
' ''t"
i\-
Pase186 CT0e1o1_CM12_1
ffi ffi
SIEMENS
lnspect volume mesh
Page187 CT09101_CM12_1
Page188 CT09101_CM12-1
lnspect volume mesh l;=ffi]
Page189 CT09101_CM12_1
gelect Mesh.B
Page190 CT09101_CM'12_1
SI EM ENS
Fixing problem 3 cont'd: Create custom control
.*:,.)"
'li "''
* hi^ frrthtt
4 .\,.,'.-:, ..
I
SI EM ENS
lnspect volume mesh
Generate the volume mesh again with the new settings and examine.
Save the simulation as Mesh2.sim.
This mesh is adequate for the purposes of this training. However, here are some
additional notes that would be relevant for production runs.
Page193 CT09101_CM12_1
Review questions
Why did we implement the custom Part Control for the Thin Mesher?
Page194 CT09101_CM12_1
Process check
Page195 CT09101_CM12_1
Page196 CT09101_CM12_1
SIEMENS
laVer,.r{y {.. {^{"-
>!/
'!
SIEMENS
Thoughts to consider
What are some benefits to inspecting the surface mesh before proceeding to
creating the volume mesh?
\-
Page198 CT09101_CM12_1
SI EM ENS
Meshing workflow
No
Page199 CT09101_CM12_1
''.'.
What is mesh refinement?
-..1
l
,&jj*lrri4d;li:," l.j-:.,..-,; . . .
:, :.t
ll
]
---l
Page200 CT09101_CM12_1
a
SIEMENS
Mesh refinement options
SIEMENS
Local refinement - Custom controls
Curve Part
Surface
\*
Page202 CT09101_CM12_1
SIEMENS
Local refinement - Curve controls
Page203 CT09101_CM12_1
SIEMENS
Local refinement - Surface controls
Page204 CT09101_CM12_1
SIEMENS
Local refinement - Part controls
Page205 CT091o'l_CM12_1
SIEMENS
Local refinement - Volumetric controls
Volumetric controls allow modification of:
- Surface mesh size within the volume,
. Volume mesh size (within the volume),
. Volume mesh growth rates,
' Prism layer parameters.
\-
Page206 CT09101_CM12_1
r
SIEMENS
Local refinement - Wake refinement I
E
Wake refinement projects from a curve/surface into the volume mesh.
Volume mesh sizes and growth rates controlled in the refinement zone.
Page207 CT09101_CM12_1
SIEMENS
Local refinement - Controls hierarchy
' fBEilMB
, & {fl
Hsn,$
n Seqqrs
!citware
Page209 CT09101_CM12_1
SIEMENS
Surface mesh refinement L__
xx
Recall what makes a good surface mesh from the previous discussion.
, Resolves all relevant geometric features,
, Smooth growth rate of face area,
. Cells have adequate face quality.
-.-.-ljaE&&-
t qrtu*o'*'- I
l g f rt-ogrt *. I
I arrra.or*o I
| . irtttery I
I a *rachcr*ra I
t Orercd**
a cofrE
) ;t!: ira:.lirii .rr.)*!:i a;irafare ,1;1r,t;;t,, I tl !*!iw,
Page2'10 CT09101_CM12_1
SIEMENS
Surface Remesher- Target and minimum size
Page211 CT09101_CM12_1
SIEMENS
Surface Remesher - Growth rate and proximity
Page212 CT09101_CM12_1
SIEMENS
\- Volume mesh refinement
I
Recall that the volume mesh is highly dependent on the surface mesh.
\-
\*
Page213 CT09101_CM12_1
t
Each volume mesher also has parameters to control.
Page214 CT09101_CM12_1
SIEMENS
Trimmed Cell mesher - Volume growth rate
.t
l.
.:aetd* ,-.ffi
I
-l A#{qtdi
I Cei{*t
a.'r!, a.'.:- .n
a,',.,. rlr'r.^:, 1:}*
SIEMENS
Trimmed Cell mesher - Volume
Default Values
il
rl
:t I
i,frE
I
M-l:=ririi.' .
--4-'-i ;
--1 'd
ffi
Define both the alignment and the origin of the
template to avoid thin cells at outer boundaries.
Page217 CT09101_CM12_1
SIEMENS
Trimmed Cell mesher - Recommendations
Size parameters should be specified in factors of two of the base size.
@@@ @
Page219 CT09101_CM12_1
SIEMENS
Polyhedra! mesher - Recommendations
Use similar surface sizes for adjacent surfaces.
Avoid sharp cell volume transitions when using volumetric controls.
Page220 CT09101_CM12_1
SIEMENS
Prism Layer mesher
Prism layers should properly resolve the physics of the boundary layer for the
simulated flow regime and chosen models.
. Consider physics models, surface sizes, acceptable mesh size for hardware,
simulation goals, etc.
S-
. Detailed discussion included in Effective Simulations section.
SIEMENS
Prism Layer mesher - Gaps and corners
\-
Thickness of the prism layer is automatically reduced or layers retracted to
maintain cell quality.
Behavior in gaps and corners is controlled by three parameters:
" Gap Fill Percentage,
. Minimum Thickness Percentage,
. Layer Reduction Percentage.
Page222 CT09101_CM12_1
t-
SIEMENS
Prism Layer mesher - Gaps and corners
sl glvl Hsd$
sher - Recommendations
Ensure reasonable groMh rate between outer prism layer and first core cetl.
BAN! G'O&I}{
Avoid: i
r High stretching ratios and excessive cell aspect ratios near wall.
. Prism layer retraction where it is not expected / desired.
Page224 CT09101_CM12_1
a-
Which custom control might be used to refine a portion of a single part surface?
What prism layer parameter should be modified if prism lpyers have been
retracted in an area that was not intended?
Page225 CT09101_ClVl12_1
Page226 CT0910l_CM12_1
SIEMENS
laXzr^^rly {.. t^k
5I EM ENS
Activity
Which Surface Remesher Control(s) can be modified to refine the mesh below?
:. l',.;,, ,i,lf+c!ciF 1
Page228 CT09101_CM12_1
SIEMENS
Activity
Which Surface Remesher Control(s) was modified to change the mesh below?
Page229 CT09101_CM12_1
SIEMENS
Activity
Name three custom controls that were used to produce this mesh.
See next slide for detailed view.
i.ri,.
Page230 CT0910l_CM12_1
E
SIEMENS
Activity
Name three custom controls that were used to produce this mesh.
Page231 CT09101_CM12_1
SI EM ENS
Activity I
The next model is for internal flow through a muffler with flow baffles.
Page232 CT09101_CM12_1
SIEMENS
Activity
@
053
' oc
Temperature (C)
180
Page233 CT09101_ClV]12_1
SIEMENS
Activity
ritlyre {c)
}U
Page234 CT09101_CM12_1
SIEMENS
Activity
a1 t]
rature (C)
j.]0
Page235 CT09101_ClVl12_1
SIEMENS
Activity
Page236 CT09101_CM12_1
SIEMENS
Activity
What things need to be considered when choosing settings for this model?
Page237 CT09101_CM12_1
Page238 CT09101_CM12-1
SIEMENS
la6cr^-.{y{..t {z
SIEMENS
lntroduction
This workshop covers the selection of the physics models and setting initial and
boundary conditions.
The starting point is the sim file saved at the end of the Meshing Il Workshop.
lf it is not open, load simulation Mesh2.sim from the 05_Physics folder and
Open All Scenes.
Page240 CT09101_CM12_,,i,,-
Skills to be learned and practiced
SIEMENS
I
i ] tu/eic.L
Q.$!tstr.n,tq
l. A esu*to.'rn..
; l) c"a,.n"
: t_d) r-F;rarrr*urm;r
t O r'q,'r
44 42,
*a r*rd$
. +f A!
&r*6
, *I a*.I r.rrd6&ea!*ed
r *.nltk&rg{co*e*
. s t *6**"re
. * lffirkftrd*ffi,
*-I!@M
*'a rhe*ldk
*jL ftne&ia
,* I rar4**
. $Iffir
Page241 CT09'101_CM12_1
SIEMENS
Process planning
\.-
a
SIEMENS
Physics models - Fluid
When a volume mesh is generated an empty physics continuum is created.
Rename the existing continuum to Water.
Select physics models for the fluid physics continuum.
Steady,
Liquid,
irr.l:Ii, rt n'{ |tr]r,','lalll,
Segregated Flow,
Constant Density,
Turbulent,
K-Epsilon Turbulence,
Segregated Fluid Temperature.
. Lrqrr:i,/:.11,'rr fr ij!
SIEMENS
Physics models - Fluid
The model properties of the selected models can be viewed and edited within
the Physics continua, which we have just renamed as Water.
Expand the Water physics continuum node and consider these questions:
What is the default material for liquid? Does it fit our requirements?
Does the initialstatic temperature make sense?
Page244 CT091o1_Clvl12_,
SIEMENS
Physics models - Solid
the fluid all other continua must use the ' Ele€tromagnetism
Il Co-Simulation
segregated (energy) model. 'r Auto-slect rs6mmended modeis
Pase245 croe101_cM12_,t
mffim
$tf;ft'*s'rui';
Physics models - Solid
Expand the Aluminum node and consider these questions:
What is the default material for solid? Does it fit our requirements?
Does the initial static temperature need to be changed?
Page246 CT091O1_CM12_,1
SIEMENS
Physics models - Solid
F:11:
Ti l.
q Expen
Page248 CT09101_CM12_1
Check interface mapping tr EMENS
t.L t?prlr!fllirnt
Check the interface mapping. a ftory
$ a c.oB{ry
Expand the Representations > Volume i. I YoJumr u*rh
i a F,nt.!orq,!R.r,6',
+ t Frurd
Mesh node until you can select the i
i * H.45inr
I a rbl.volum.rolndur.s
HeatSink > Faces boundary. ;sffi
I F&.r(Frurd/ts.irtirk,
I F...s lfi.nsink/lc81J
Note the number of faces in the
boundary.
Right-click the Interfaces folder and Faces 't7445 Faces 7360
select lnitialize.
.Note that the faces in the contacting area
have been moved to the interfaces.
-
t'a Rrtio", --.---
?I Ftuad
I f xcnsinx
* I Somcrrirr Boundarv I
-!I-- ----r
'. r.FII Fadrr lFtutdlHcctirlkl
Fncr(H.rr5rnk/lcGTlG lnte.faces I
. o. I Fcr*urc Cuilct
nent S{)ftdarg
Page 249 CT09101_CM12_1
-,
SIEMENS
Apply physics to boundaries
x
Boundary conditions must accurately represent the problem statement.
' The wall boundaries of the solid regions transmit energy to the surrounding
environment,
' The chip region produces heat.
. The flow boundaries of the fluid region introduce and remove cooling water.
Conditions at interfaces will be solved.
Silibon
Power = 300W
Water HTC=5W/mzK
rit : 0,005 kg / s = 300 K
Tambient
T6so1 = 15 C
Aluminum
HTC=5W/m2K
= 300 K
Tambient
Page250 CT09101_CM12_1
SIEMENS
Boundary conditions
t'oll
5 Fearurr
t 5 PhYritt !ei lleruTag
6- 3 Pnyslrs
Page251 CT09101_CM12_1
SIEMENS
Boundary conditions
I *oo* |
I {fHclTrmfrrcodciam
lffilr.rwri;?nl::r.:
.. --.ll:: l a'r
E r,1
.-.* rr,,,-,
ai 5.
Page253 CT09101_CM12_'l
SIEMENS
Boundary conditions
Expand the Physics Values node of the lnletEnd boundary in the fluid region.
2 Set Mass Flow Rate value to 0.005 kg/s.
$
"J
Set Total Temperature value to 15 C.
4 Outlet boundary values can be left at defaults.
* A F.eqisnt
?t Fruid
f .l goundarie,
tX-t- 1 Ff i; -r:i,1 ffi I r ".r,- r1.1,:..i ),t t":r1
. fi ri,r :-. -. ,.,.1t',i- ' {* tfr; i: :iil,..rr:
I ra rFr,-i rii,r & ;t\ . :: \i1j 11-
,,,
:,- Mnsr FlowRate u.-, lt, , r_,,
;-''r r'rr':.r,'r .' rorir leffpar!lurr
Turbulence lnlenrily 'r. a 1.ar) rt :a \.r. . I
Turbulent Vircorky Ratio
v
SI EM ENS
Process planning - Status check
Page255 CT09101_CM12_1
I
t'=*=*r
Stopping criteria I I
t
SIEMENS
Review questions I
I
t
I
I
SIEMENS
la6*r.^,{7 {..t^k
SIEMENS
Thought questions
For a simple pipe flow, what fluid behaviors are important to consider?
When mixing two flows of differing speeds together, what happens to each?
Page261 CT09101_CM12_1
SIEMENS
Physics models - Time
Page262 CTo9101_CM12_1
,
SIEMENS
Physics models - Time: Unsteady solver
lmplicit Explicit
Stable with large time-steps with Less stable, small time-steps
the consequence of smoothing tiny required which preserve tiny
oscillations oscillations
Appropriate for aeroacoustics
More CPU time/iteration, Less CPU time/iteration
More memory Less memory
lmplicit unsteady solver generally preferred because most problems don't require
high temporal resolution.
SIEMENS
Physics models - Materials
Material models manage thermodynamic and transport properties.
Single component
A single solid, liquid or gas.
Multi-component:
Single-phase, fully miscible mixture of single components.
Properties weighed by composition in a given cell.
Multiphase
Two or more immiscible phases.
Page264 CTo9101_CM12_1
SIEMENS
Physics models - Equation of state
Pase265 croelol-clvl12-t
ffiffi
SIEMENS
Physics models - Solver
Segregated Coupled
, Compressible flows
- lncompressible flows . Natural convection
Best for... , Mildly compressible (large body forces
flows ' Flows with large energy
sources.
Page267 CT09101_CM12_1
I
I
Physics models
i SIEMENS
- Viscous effects j
i
i_-,....,..,. .
Page268 CT09'101_CM12_1
SIEMENS
Physics models - Viscous flow categories
Laminar: Flow features are directly resolved (using a continuum assumption).
Turbulent: Small, high frequency features are typically modeled instead of
resolved.
Dramatically reduces mesh size and computational expenses.
\--
Inertial
Laminar
--:=---- = Rg
Viscous Smooth flow
lnertial Turbulent
\- Yiscous Chaotic flow --*!;-_,
-=Re
:. ?l1 l Sf,;:'er: P.ildu{t l-ai*i
Page269 CT09101_CM12_1
Physics models
Gharacteristics
- Turbulence @l
. Significant, irregular fluctuations in (3D)
space and time.
. Rapid mixing, and energy dissipation.
Modeling
Many models available, resolving
turbulence in more or less details.
. RANS: time-average of mean and
\- fluctuating components.
. LES / DES: model small-scale
motions.
Page2TO CT0910l_CM'12_1
SIEMENS
Page272 CT09'101_CM12_1
a-
SIEMENS
Boundary conditions
Page273 CT09101_CM12_1
SIEMENS
Boundary types
Open boundaries allow flow and energy transfer. lnlets and Outlets.
>>{ Velocity Inlet ilU Pressure Outlet
X Mass Flow Inlet $" Flow-Split Outlet
B Stagnation Inlet
(8 Free Stream
' Define flow direction, temperature, turbulence, , Pressure Outlet allows backflow.
and phase conditions. . Flow-Split Outlet: ratio of mass flows for a set
' No backflow. of outlets.
' Stagnation lnlet and Free Stream have
additional options.
Closed boundaries allow only energy transfer. walls and Symmetry planes.
wall x Symmetry
',
. Prescribe zeto ot non-zero near-wall velocity. " Velocity across boundary is zero (slip).
' Several energy conditions available. heat flux, , Temperature and pressure are zero gradient.
temperature, heat transfer coefficient, ambient . All physical quantities have zero-derivatives in
temperature. boundary-normal direction.
" Allows for phase-specific outflow
Page274 CTO91O1_CM12_1 m
ftffi*
SIEMENS ]
E - Regions : n Regions
a r Region i * Region
t B Boundaries ; iB Boundari6
* tt Wall r, Wail
n - ElilE&E&@G r n Physics Cbnditiong
r 'F Referenee Frame SpecificaUon ,,* EEIEIIEIEEI
:- Shear Stress Specification {rBlended Wall Function
r" Tangential Velocity SF
. EllH7*rlIitk
.' ur-#di--[h;,*Jir*_*u , ,n
e@E
Feature Curves
i' User Wall Heat Flux Coefficient Speaification
i' Wall Surface Specification
,e Physics Values
'I Featura Curves
Page275 CT09101_ClV112_1
tt SIEMENS
lnterfaces i
it I
Topology
ln-Place Periodic Repeating
in-place
rnlertace
Type
. lnternal: Mass, momentum, and energy transfer.
" Contact. Only energy transfer (solid-solid or fluid-solid).
. Baffle: lnfinitesimally thin solid barrier (porous or impermeable).
Additional types exist for specific applications
Page?76 CT09101_CM12_1
SIEMENS
la6u,-rly{..l^k
T
SIEMENS
lntroduction
This workshop covers the creation of reports, plots, and scenes, running the
simulation, and analyzing the solution results.
The starting point is the sim file saved at the end of the Physics workshop.
lf it is not open, load simulation Physics.sim from the 06_Data Analysis folder.
Pase278 CT09101_C|V12_1 ;:rf ':.' ', "ilr. t,:\.1 * 'u\\ ..'h . '., -.%*a!l!-!SH(!FL
tru-gy
SIEMENS
Skills to be learned and practiced
*
Create reports, monitors, and
plots to jud.ge convergence and
extract .desired engineering
values.
Data
AnalYsis
Br
X
1l
.- .-
re
.r".)
Page279 CT09101_CM'12_1
SIEMENS
Process planning
SIEM ENS
Greate reports
B !m tit r: rrnl
Create a report of the water outlet }E NewRcporl
E
se
temperature. tE
FJrrl
l;rit a:,trtf,:irr11
Select new Mass Flow Averaged report E
*E E'JiI Heir Er aiirrjqer il-rrrl, ir:"ari )
from the Reports right-click menu. sE F.rlrr: n Hiat E.rclatrger ii n!lr ilfeifll
l).\n t:roup Heal fr aBriar
Rename the report to Outlet irenrfJf:it ef ir|?r|:9
Lri! 1r,r!qrll
Temperature.
Set the Field Function to Temperature.
Change the Units to C. t aiirr,rln i,1rilo./ Report
Select Parts: Regions > Fluid > Fill
, .&
Holes. HeatSink.OutletEnd. ?..a':tt1.
a Expen
Page281 CT0910l_Clvl12_1
Create reports
t Exprn
Page283 CT09101_CM12_1
SIEMENS
Create monitors and plot
t+E Mont
r t' E Ptott
+B
e E tool5
t:r.11n lJi'nittr f, ail P.p.n
i4,riirplt rep.fr: r:(le!:1r11 l-lre !itrglr plor, L1r (rea1. m!lilple plqta:
Page284 CT09101_CM12_1
r
SIEMENS
Plot options
In the menu, select Tools > Options. ;lle E'rlli f/ie:h Soilitirn Tools !Vinri.rr./ Hel|)
t1e-:-- - )C2 tdrl-ons
Select Plotting.
F Rt:iel Options
Firqirs
Servics Prr
Seryiccs Anli
E En,Jiroameni Lin€
? P:niJci s/ry
i..!censing
uaj
Page285 CT0910l_CM12_1
SIEMENS
Plot options
Tick Count: 2.
Close the Options dialog.
ln the Plot node, right-click Reports
Plot and select Apply PIot Options.
Page286 CT09101_CM12_1
SIEMENS
Scenes - Temperature Sections
SIEMENS
Scenes - Temperature Sections
[,
-r;-FllrtrE
Create a scene of temperature inside the domain aa '-.::n.lr Jr: :a,:"..i'.-
.) &&.:l.,tf.i,:
(continued). t" -' ...,,:,r,. :
I . ,:1i,1rl
i, *ii* -"r:.:
"1 !n the scalar displayer, select the Scalar Field a ,{ali.r !i.ld
Page288 CT09101_CM12_1
a
\-
SIEMENS
Scenes - Temperature Sections
T
Create a scene of temperature inside the domain 1g irrng,1r1" ).a "a',il.ri
l gk :: .:! '1t.'
(continued). + ._- .-,,;rtltit 1
i .ri,1r.i
''" &* 11' -
Page289 CT09101_CM12_1
SIEMENS
Scenes - Temperature Sections
The default Outline 1 displayer shows the region outlines. We want to see the
outlines of the different components on our plane sections.
\-
OK Cancel Help
fi eurtt
+ Faces
+ Heatsink.In 2 of 33 *lcctd
.a UaalCinlr Tn
Page291 CTO91O1-CM12-1
Page292 CT091o1_C|VI12_1
rI srerueus i
Run I
I
Save the file. The simulation takes a couple
Run the simulation fu minutes to reach the stopping
criteria of 350 iterations.
L
ha{ion
Page293 CT09101_CM12_1
SIEMENS
Analyze results - Temperature Sections
Discussion points
, What were the simulation objectives? Were they achieved?
" Recall the design requirements. Were they met?
* E Anribrres
..ffi1 Modett#indow
l. 3 'Eirkgreurd calor v
i&Yiro
?- J$ uptare
. ffiX Ar'inali6n
v'
.-j- Axe=
* E Light, \/
q- I CliF Planeg
* f Annstatisflt
.T
Eril
\,
Edil qnn.idlion.
\/
Page295 CT09101_CM12_1 \/
\_
Iv
I Analyze results
Discussion points
- IGBT
SIEMENS
:'-'-*
vhv\
r. rt .:
4*. //5
3S 099
a-
-
3il$ Produ(a Lifecysle llla*aqement Softwa
Page297 CT09101_CN412_1
SIEMENS
Scenes - Temperature Threshold
Create a scene showing the volume with temperatures
above 45 C (continued). ; i 1.:'
Page 298
SIEMENS
Analyze results
The portion of the chip above the temperature limit is quite large.
Discussion points
' Assuming the boundary conditions cannot be changed, how can the design
be modified to reduce the maximum chip temperature?
" What determines the amount of heat transfer to the water?
Page299 CT09101_CM12_1
i SIEMENS l
Scenes - Heat Flux t
!r
Page300 CT09101_CM12_1
SIEMENS
Scenes - Thoughts
* Regions
' * Fluid
' x Solid
lnterface boundary contains , *a BOUndarieS
solution data at the interface. . tt Default
\ , i* n Physics Velues
rE@l*= Interfaces
'
ir,io t,i ,r,"*" sotid tide inteiete
ffi rs1I
Interfaces are two-sided, so it is possible for
the solution to vary depending on which
side of the intErface boundary is selected.
tlail Yt
Wbfi Y+
t&, T
Page3O1 CTO91O1_CM12_,1
SIEMENS
Analyze results
IEIT'
Discussion points
. Where is the heat flux the highest?
. How can the design be modified to have more areas of high heat flux?
,*
!* *l1
etralsa. iaai
i -r::'-. -:'j,.1i
*;*:,.. ii.
Page302 CT09101_Cl\112_1
SIEMENS
Design change
Page303 CT09101_CM12_1
r-
] SIEMENS
Design change: Replace geometry part
I
Replace the baseline HeatSink in the Geometry node with & .:: .:..:,'.. jax :
i Navigate to HeatSink_5bend.x_b. *
- & 'r.'at.1.,4.:h.:2.ft\rn
:r'
-
!_fn.,.tlt:
,
l.a.:t:),t t I
& c*qr€t'Y
& O-.AD!"tadeh
& pa!
" r:i. F* a.i6
,qs.lM
' Q| eae<*
' *; res'
' *8t t"sD;6e
' & crta<ts
. i& Cpaabos
' &! r"o''
:'-i.F*#d
. @ arraaro*
; $, Agtrut*rtae*
& csrg
Page305 CT09101_Cl\r12_1
SIEMENS
Design change: Part organization L___
The new HeatSink part surfaces have replaced the existing ones.
The new part only has a single named curve, just like the first HeatSink part did.
The empty curyes can be removed.
Delete all curves except Edges.
Confirm deletion.
if a ems
,' b.l Ftu'd
F €f rrrnore,
fO Hedslnt
': * - iufieat
..t ,'tcuoas
-.. :1 tt:
lnletEnd
SIEMENS
Design change: Update operations
Page308 CT09101_CM12_1
a-
SIEMENS
Design change: Update operations
Construct the dynamic query predicate: Select the predicate inputs: Name,
contains, end.
. Results: lnletEnd, OutletEnd.
Select OK.
Page309 CT09101_ClVl1'_1
I
SIEMENS
Design change: Execute pipeline
Page310 CT0910'l_CM12_1
I Design change: Analyze results
Discussion points
" Has the modified design simulation reached an equilibrium state?
t--
SIEMENS
Repon5 Plot
? '':: i j::: -:
Page311 CT09101_CM12_1
SIEM ENS
Design change: Analyze results
ffi'"'
':. ':
_
.' i:i
Page312 CT091o1_Clvl12_1
SIEMENS
Review questions
s- How was it confirmed that the design criterion was not met with the original
design?
SIEMENS
Process check
\-
:le M611r{re{}q
Page314 CT09101_CM12_1
Review questions - entire coldplate workshop
What are some assumptions that were made in this analysis that might affect
prediction accuracy?
lf you wanted to explore the design, what things could be investigated with a
design sweep or optimization study?
SIEMENS
Thought questions
Page318 CT09101_CM12_1
SIEMENS
Overview I
Solution data: values of each conserved variable on each volume mesh cell and
boundary face.
,i ., ---'
I.'.r ro!
Data analysis: evaluate solution data to gain insight, draw conclusions, and
support engineering decisions.
Page319 CT09101_CM12_1
SIEMENS
Data analysis workflow
Page320 CT09101_CM12_1
SIEMENS
Data analysis objects
lnput Part(s)
Volume, surface or derived
part
Page321 CT09101_CM12_1
'v
lnput parts F"=-l
Various simulation objects are available as input Parts for analysis objects.
Parts / Part Surfaces Regions / Boundaries Derived Parts
'::-:;bt* -, '".-- .= ,ii tua - n .'
! .! r*nk
I i i*'r*s*
*tm:
, I .!9*.i
S r (1 aar:e r Il r,
a1 aa8@ I
. A ii-,!i 7rr.
-', tua *(ton
, q)a4(Er 1r i -lrwtiM :
| ! ,' ,. -;b$*,*2 '!'.J!x'
Page323 CT09101_CM12_1
Derived parts are geometry objects created only for .< Derived Pafts
analysis. ' r, Arbitrary Section
. Not part of model geometry or mesh. * Cell Sur{ace
. Do not influence the solution. ' "' Constrained Streamline
' ,:',, Cylinder Section
" Must exist within the discretized domain. ' ,:j' Isosurface
. lnput part(s) define what part of the volume mesh ,'.' Line Probe
, .:, Plane Section
to intersect. '., Point
. Data are interpolated from mesh if derived part "',' Presentation Grid
not coincident with solution data. ':, Resampled Volume
,': Scaiar Warp
, 'i, Streamline
Many different types of derived parts are available. '. Threshold
For example...
Page325 CT091o'l_CM'12_1
SIEMENS
Derived parts: Examples
Page326 CT09'101_CM12_1
Reports retrieve or calculate solution data and return a 's Reports
Page327 CTo9101_CM12_1
Monitors
statistically sample data (Mean, Maximum, Variance, etc.) for a selected part.
Page328 CT09101_CM12_1
t-
Plots
'ti;
r?i " i 6\tl
eR(,!ahlr r. .' .i\Le: r.ct!
Cell Q.ality
frsnaqpmerll Software lr
Page329 CT091o'l_CM12_1
SIEMENS
Discussion
Reports, monitors, and plots can be used separately but are usually used
together. Which one action best describes the function of each analysis object?
Monitor
Locate
Track Display
Animate Sort
Calculate
Page330 CT09101_CM12_1
SIEMENS
Discussion
Reports, monitors, and plots can be used separately but are usually used
together. Which one action best describes the function of each analysis object?
Locate
Track
Animate Sort
STEMENS j
Scenes
@ffi*ffi' ;"i:rY
Page332 CT09101_CN]12_1
SIEMENS
Displayers
eA Sshr t
g n ParS
Y:i aPressure tPa)
:..:l& evt )'i'
i.-- r Smlar field ', {:, ,?
;i::::
.;/)!)
i
, 'i r F$nt Se*
.
.t' -
-
Page333 CT09101_CM12_1
i
I
SIEMENS
Displayers - Layering information t
i
I
L- ,.- __--_,--. __ -, - -
1500
Vetocity: Magnitude (m s)
400 054 1.o 1.5 / 20
re iJ
Page334 CT09101_CM12_1
T SIEMENS
Displayers - Layering information I
I
i
Displayer
Geometry lnput Part Solid Valve
@ Representation
Properties
Geometry
reduced opacity
&
Vel..ty: M)sniude aht, il
Streamline
Scalar
lnput Part
Representation
Properties
lnput Part
Representation
Streamline derived part
Volume Mesh
Veldcity Magnitude
Page335 CT091o1_Cl\il'12_1
SIEMENS
Exporti ng visualizations
Page336 CT09101-CM12_'l
I Exporting data
SIEMENS
Auto Save saves the full sim file at specified intervals or checkpoints.
Export Data tables and plot data can be extracted and exported to .csv file.
Simulation History file (.simh) saves solution data from many simulation states
for later comparison.
Page337 CT09101_CM12_1
SIEMENS
Running and data analysis
Additional benefits:
. View solution progress interactively during run.
. Export hardcopies of plots/scenes.
.
ffi
ExtracUexport data.
Rendering scenes during a run can
E-
slow down the solver depending on
hardware, simulation setup, and
complexity of visualization.
Page338
W
CT09101_CM12_1
il
I
Triggers
! SIEMENS .
ii
I
I
,-
u'l1 -, 45 ^oov-
-^oooc
o F
U
vo
3 mi --F dO
:40 c
El .4
33S F
'I
,E lteration . i
FJ
30
a o Time Step
i.
i
)ct-. "-*E*-6''--:t-=5'-=;"--66"--r6'--6-*&"-f
"4 Ar
IteEalnn
Triggers [il;-j
Maximum Chip Temperature (C) Maximum Chip Temperature (C)
Time (s) Time (s)
soo l-q 20 o 40 to 6_q 7g 80 90 lg0 --o l0 lo 40 50 60 70 80 90 100
j
j
4tj
i
'r-rr .15
a
I U
9oi 40
!l
si
5a'.
= a 35
Fi F
30i
1
I
. Delta Time = 10 s
10
AIUpdate Event = Every 2 G
!
20 30 40 50 60 70 EO 90 250 5 lo
100 is 4
lErifoE Iteration
Page340 CT09101_CM12_1
I
SIEMENS
Review questions
.t-
Page341 CT09101_CM12_1
Review questions
SIEMENS
I
What objects can be used as input parts for analysis objects?
flanaag;re;lt Soitwar*
Page342 CT09101_CM12_1
t-
SIEMENS
la6zr.-;{7 {.. {^k
t-
I:
-
l-
SIEMENS
lntroduction
This workshop is an extension of the previous cold plate heat transfer analysis.
The steady state model provided equilibrium results of the system. We will now
address the transient behavior of the system as it starts from initial conditions
and it moves towards equilibrium.
15
re::
oOO at 193 la ]-:
, ,,j4it
,a t66
Page344 CT09101_CM12_1
i-
--_--*** :
Skills to be learned and practiced srEMENs
L j
6' ady-Statersetufi'.to'iun' r.
.Set,times@pal1d
r$loppi$,eiiteria'r''
Page345 CT09101_CM12_1
Process planning
' achieve convergence for each time-step and stop entire simulation (Stopping
criteria).
" update with different triggers and save solution states (Post-processing).
Page346 CT091ol_CM12_1
T
SIEMENS
lntroduction I
Silicon
Water Power = 300W
tit = 0.005 kg / s HTC=5W/m2K
Ts67a1 = 1,5 C Tambient = 23 C
Aluminum
HTC=5W/m2K
= 23 C
Tambient
Page347 CT09101_CM12_1
f-
SIEMENS
lntroduction I
The starting point is a modified version of the sim file saved at the end of the
Data Analysis workshop.
Page348 CT09101_CM12_1
SIEMENS
Physics models
a: Electromagn.tism
i Pa*ive SGlar -...-:. a :
The initial temperature has been set to '- Turb!lenca SupprGion
-'Constant
Densiv
i:: f'1csh Oeformation r:Gradients
23 C (room temperature) for all physics : Di;perscd Muitlphase aa.j at:.'a:: : '..,
a lvlultlpha* Interacton
continua. -Virtual Disk ' Thre! Dimansionai
t, ., ...
Also the ambient temperature in the -- E4anc. ac.lad rr+ 3 €xFodl@rb.d !t u€
i.:--.rr Slr(cr Liate.
solids has been updated to 23 C. a;luminw,Siic6,V/at
fag5
. Shhc T.mFr&r.
ar.rari -4..::.or -arisb.!
Page349 CT09101_CM12_1
SIEMENS
lnitial conditions
; 271
Page351 CT09101_CN412_1
SIEMENS
Apply model settings
l
The simulation starts at time t = 0 seconds.
l-]: l,larrmum tnnrr llerarroni
Page352 CT09101_Cl\il12_1
Stopping criteria
lnner lterations
Once the stopping criteria for the inner iterations have been satisfied, the solution
is stepped fonryard in time.
Default criterion: Maximum lnner lterations.
Outer Iterations
The model stops running when the outer iterations stopping criteria are satisfied.
Default criterion: Maximum Physical Time.
Page353 CT09101_CM12_1
Iteration
. Displays iteration history of the parameter.
. Used to judge time-step convergence.
Time
. The curve includes only the converged time step value of the parameter.
. Used for time history.
Page354 CT09101_CM12_1
SIEMENS
Create monitors and plots
Create a plot to monitor chip maximum temperature over time in addition to the
existing plot over iteration.
i Create another monitor and plot from the Maximum Ghip Temperature
report.
? Verifu the monitor Trigger is Time Step.
? Expand/Conhd Trce
Comparison of the two monitors: Rep{ lttraxinum CiiD Tempe€turel - Itlarif;um Chip Temperailre]
EBbl.d ?a ..
Clo* Help
SIEMENS
Setup scene and run
r l,lcdel Window
properties window select Solution Time. , & Background cclor
6 View
"f update
. Delta Time Frequmcy
il Clear the existing solution. r Hardcopy
i+ Animation
- " Axes
I Piot Image 2
d Run the simulation.
I neralicn
. T Logo
I Plcl lmage 'l
.f Plot lmage 2
Page356 CT09101_CM12_1
;
!
Analyze results
! sreurus
!
!
r,Fpe.al|./P,:i
:3 tl5 :r:r,
Solution Time O.3 (s)
Page357 CT09101_CM12_1
SIEMENS
Process planning
\- The time-step size and number of inner iterations can now be modified to reduce
total run time.
The goal is to use a greater time-step size with less inner iterations while
maintaining solution stability and time-step convergence.
Page359 Crootot_Cn/ItZ_t W
SIEMENS
Additional stopping criteria
When the monitor satisfies the given criterion the solver moves to the next time-
step, or can be set to stop the simulation (outer iterations).
r SIEMENS
Analyze results
femperalurc tC)
Page362 CT09101_Cl\iI12_1
SIEMENS
Review questions
I
Which solver setting added by a transient time model should ALWAYS be
considered prior to running the model?
How are the stopping criteria different for steady models and transient models?
What data sampling control needs to be chosen correctly for monitors and
plots to appear as expected?
Page363 CT091o1_C|VI12_1
Workflow steps
Physicr
Data
AnalYsis We have finished the entire cold plate
workshop. Now we try to optimize
Page364 CT09101_CM12_1
SIEMENS
lay*,-r{y {.. t;{e-
SIEMENS
Introduction
Before we can do this, we need to make a few adjustments to our simutation file.
The starting point is the sim file saved at the end of the Data Analysis workshop.
Load simulation BaselineRun.sim file from the folder 07_DesignManager
folder.
g.--'-*--l
Parameterize the chip power IsrErvrENs j
Type
Value 300 0 \tJ
Oimensions Power
Tags il
Page369 CT09101_CM12_1
SIEMENS
Sim u lation adjustments
Auto tu'lax
:{ere/Pbt
frrq
From the root node of the simulation tree, Create Srnulation
SmYi{%ffis
*6i
Design Manager Project. e L ?i1i...
.slt
Basel ineRu n-ParameterSweep.sim fi le. @ )'
cas;teD*,qnM;nacerProid
, *, (?rriqitdirri
t I S;Snulatiss
il D6ignstudix
f s#pdrots
'L Plots
ri L Teb
Page371 CT09'101_ClV]12_1
SIEMENS
Setup the design study
E0*igf,Slsdy-lnputPdrimtu X i
. cb= , Help
Page372 CT09101_Cl\i112_1
a-
SIEMENS
Setup the design study
Select the Responses node and in the properties window select Maximum
Chip Temperature.
Select the Scenes node and select the Temperature IGBT scene.
E--
'* Scan€6
saen.s-Proparties X E;
Responses 6 Scen€s g
Stud:,, rE:F,rn:? nlandg*r 5-oleat refiren(e r{€nes
Page373 CT09101_CM12_1
tisrEMENs
Setup the design study Iti i
,:,:fftr'.jla P, \caaaa a
Simultaneous Jobs g
l.)\mb.r ){ f,irir!n: ti .!eiute :iflr!lta.}e$(.1,'
Page374 CT09101_CM12_1
SIEMENS
Create a scene snapshot
Partfiretat i eeiF
On the Snapshots node, create a new ,- sA tirrlatiii,:
Scene Snapshot. : .8 Oz';iar'',,t..tiiz'.
_.r
.ri lLi-,r il
Select the Temperature IGBT scene. ; B :r,,:i:;!r;,t:-.
r ;S s."n" Snapshat
t{ ltrrnrri rr'
r*.lnriitilir[:
M ""
' u ,\) -
D: :iitn 'itL;C..
ij.:i,lrr:it lii{l
Scenc lTamperature IGBTI
Tin:, tl
? Expert
L:t --'t D < :t grt
arn;ign
S(ene
:
I
,.ere iit i:t 4t:cla'f7
Page375 CT09101_CM12_1
T*-*-* --*l
Create a plot of design study results I srerurrus
it i
DesQn Studies
iiiDesign stuc,y
1 of 1 sehcted
I OK Cancei Help
Page376 CT091o1_Clvl12_1
SIEMENS
Design study series
change the Type property of the Bottom Axis Data , . :'. Line5ryb
. . 'tsymbol5byk
: GnAx€s
node to Parameter. E BottomAx6
. : &e n
ln the new Parameter sub node, change the , a LfrAxis
' e Legend
Parameter property to ChipPower. ;6'ile"-- - --
'nwwP*i'.
For the Left Axis Data node, change the Response tam*
*:_ Deqrnstrdy
to Maximum Chip Temperature. * a BotbmArbDBta
..@
3- L* A)(b Dae
Change the Line Style to a solid line. ,
.
.
B€* D€qO
hs€lre tteEn
' r l"|1e style
' . Symbsqy'e
Prr:mrtar. Propenres ,
Propert€s
Pr.rnari.,, ittrrFFrrcl
Page377 CT09101_CM12_1
SIEMENS
Save and run
ireate Plot )
Op;n All Fl*t:
t ,: p-.
lJ l l8tr
EdIIdIIIE
Iependenrie:
f,3iaa*!11e
Page378 CT0910l_Ctul12_1
L
SIEMENS
Review results
Rearrange the Plot, Scene Snapshot, and Output Table so that you can
interact with them all.
Review the results, selecting each design in the design table and reviewing the
Scalar Scene.
Page379 CT09101_CM12_1
Page380 CT09101_CM12_1
SIEMENS
lan*r^-r{y{..t (e-
lntroduction
The planning phase will be done in small groups for discussion, then you will
individually implement your workflow plan.
At the end of the activity, we will discuss the problem as a group and walk
through the simulation.
,
Pressure Drop
(lnlet - Outlet)
Outlet 1
Outlet 2
Page383 CT09101_Cl\41r_1
SIEMENS
Process planning
Page384 CT09101_CM12_1
SIEMENS
Partner activity
T
1. Do you need to generate any additional geometry in STAR-CCM+ to analyze
the fluid flow? lf so, what operations/workflow steps are needed?
? What will you choose for the base size for this model? How did you determine
that value?
6. How will you determine convergence? What data analysis objects are needed?
ent S{]&ware
Page385 CT09101_CM12_1
SIEMENS
lndividual activity
I
lndividually carry out the analysis plan your group worked out.
. Complete as much as you can in the time allotted.
' Yon will not be "graded" on your progress - there should be no pressure!
After the allotted time, we will discuss the answers to the worksheet questions
and go through the workflow as a group.
Page386 CT09101_CM12_1
SIEMENS
la6c*,-r{y{..t {z
lntroduction SIEMENS
l.:;M
Page388 CT09101_Cl\i112_1
lri,
,M
SIEMENS
Workflow steps
Page389 CT09101_CM12_1
SIEMENS
Simulation overview
(atmospheric P)
Which physics modeling strategy should be followed to model the fan rotation?
Which areas of the model might need a higher level of mesh refinement than
others?
il
Page392 CT09101_CM12_1
SIEMENS
Organizing parts
I
Two parts exist in the Parts folder. Duct and
Fan. Duct represents the air volume inside
Questions
Are any additional surfaces needed to account for
boundary conditions or mesh refinements?
Page393 CT09101_CM12_1
SIEMENS
Assign Parts to Regions of the extracted part Air would create the region and
boundaries shown in the picture.
ry'5 *s#,:B
Are all of the boundaries necessary? ie-ile
i *,L&t&
Could we merge some of them? : : *-gf ***xrr
i , i"lffi rxrce*:e
i ,,: .*|; &#,{r*
| i,; rrxr***r&
: : r"-t; err,&&*
t' : i'-* tuLlrs
; ry-f Mrc€txrux
i ry- * *rds*afrsslx
i e'* ryG*H
SIEMENS
Create regions and set boundary types
*-
Assign parts to regions manually. More efficient if a ; {Cesion;*"-
large number of surfaces are used for meshing
ux@
;* n Boundaries
purposes but only a few are needed physically. , .t;Fan
I Inlet
. I Outlet
: lI Wall
Create an empty region with four boundaries by i-n Feafure Curves
right-clicking. Use the same names as for the
surfaces if possible: lnlet, Outlet, Wall, Fan. Region 1 - Propertieg
Propertres
Link the Air part to the region (properties window). Indei
. Surfaces with matching names are assigned Mesh Contrnuum None
Page396 CT09101_ClVl12_1
SIEMENS
Create regions and set boundary types
Fan -
Type !e lnlet
Taos + Cutlet
Page397 CT09.101_CM12_1
SIEMENS
Mesh setup
A polyhedral mesh with a prismatic boundary layer is suitable for this simulation.
1 Generate a new Automated Mesh Operation with the parts that are assigned
to the region.
2 Select these meshing models:
a" Surface Remesher (for improving triangulation quality).
*. Polyhedral Mesher (to resolve the swirling flow).
,,;, Prism Layer Mesher (to resolve the boundary layer).
Page398 CT09101_Cl\i112_1
W
Question
Is the surface mesh acceptable?
Check the mesh size on critical
parts, growth rates, etc.
SIEMENS
Refine mesh: Surface control
ffi
Create a new Surface Control with the fan surfaces as lnput Parts.
change the Minimum and rarget sizes to 5% and 25o/o, respectively.
Re-execute the surface mesher.
SIEMENS
Mesh assessment
Discussion points
' How many cells are in the volume mesh?
. Are any refinements needed to improve
the mesh quality? Pay special attention to
the mesh in the area around the blades.
- {& kF16
& ;f@rJ
' & i*€l;trrt
1 I ?@€t;
, bm
IIr.rjJI
, @ :3i3
402
Pase cr0e101_cM12_'
lt$Tilffift,
Refine mesh - Create cylinder F,'"*"l I
We will use a single Volumetric Control to reduce the
prism layer size and improve growth rates in the entire ; ;;* )r- :::-, . '-.1, .
Sli ai
Page403 CT09'101_CM12_1
Create a new Volumetric Control under the Automated Mesh operation and
set the Gylinder as the input part.
I Enable customization for Surface Remesher and Prism Layer Total
Thickness in the Controls folder.
Custom Size: 50o/o of Base Size.
Prism Layer Total Thickness: 1 mm.
Re-execute the operation.
Page404 CT09101_CM12_1
SI EM ENS
Mesh assessment t_
L
Discussion points
. ls the new mesh appropriate for the simulation?
' How many cells were added when the refinement zone was included?
SIEMENS
t__
Select models for the Physics 1 continuum:
-
,: Steady,
,... Gas,
Segregated Flow
,; ldeal Gas,
* Segregated Fluid Temperature,
r Turbulent,
,.l. K-Epsilon Turbulence.
Discussion point
Are the physics models chosen here reasonable for simulating the bulk air flow
in this fan duct?
Page406 CTo91o1_CiI12_1
a
SIEMENS
Background information MRF approach
E approach
Page407 CT09101_Cr\/112_1
_l
Reference frame setup
SI EM IN5
6 Rqqian5
Create a new Rotating reference frame under Tools > * 1aq!r)n i
s StrNainriE!
Reference Frames. * Featiir? a*aei
. * Phyi,.s r:ondili.rnt
Set the properties of the Rotating reference frame. * ,hysra5 Values
n:?qffi
Axis Direction: [0.0, 0.0, -1.0].
Rotation Rate: 2000 rpm. Motioh Spacillcldon - Proprrti6. .
Set the Motion Specification of the region to use the [4?1,on 5tair.iont]
F.aaeta-Lefiaie Ralntrr!)
Rotating reference frame.
Set the duct wall boundary to stationary (non-rotating):
Under Wall > Physics Conditions, change
Reference Frame Specification to Lab Frame.
* rhrr;,-ilE\
.
Page408 CT09101_ClVl12_,1
SI EM ENS
Reports, monitors, and plots
\* Questions
. Wh,"h engineering quantities might be of interest in this simulation?
. Can you"ther
create a report to measure Power required? (Power = Torque "
Rotation Rate)
. Which monitors would be appropriate for judging convergence of the
simulation?
Page409 CT09101_CM12_1
Scenes
re.,. ,....$
Torque converger.€
Questions
' Why might Relative Velocity be recommended instead of Velocity?
. Could another Derived Part better display velocity than the current planes?
" What other Scenes might be of interest in this simulation?
Page4'10 CT09101-CM12_1
T
Question
' How will we judge if the model is converged at the end of the run? ls there a
way that could be incorporated into the Stopping Criteria?
. Does this design meet our design requirements for flow rate?
ti Silib,Yart
Page 41 1 CT09101_CM12_,1
SIEMENS
Scenes
Create a new Geometry Scene in which to display streamlines and set the
Opacity for the geometry displayer to 0.2.
Question
. Why do many of the streamlines
appear to end at the fan?
Design evaluation
I
Can you improve the current design to reduce the torque required while still
meeting the design requirement of a200 CFM (0.11 kg/s) flow rate?
The design has been parameterized to allow you to modify the design. To modify
the design:
, Design Parameters are found in 3D-CAD Models > 3D-CAD Model 1 >
Design Parameters.
. The number of blades and hub radius of the fan can be modified subject to the
following constraints:
. NumberOfBlades:2-8
. HubRadius: 0.01 - 0.09 m
Mesh and stopping criteria need to be updated before running the simulation.
Page413 CT09101_Cl\il12_1
L-
!-
Page414 CT09101_CM12_1
SI EMENS
MRF Approach: Background lnformation
t SIEMENS
MRF Approach : Background lnformation t
w
- -t.
Page416 CT09101_CM12_1
SIEMENS
lNnz*,-rly {.. ti{z
Effective simulations
run efficiently to convergence and produce the desired fidelity and accuracy.
I
\
\
Page419 CT09101_CM12_1
SIEMENS
L
The STAR-CCM+ solver attempts to
balance the conservation equations i
Page420 CT09101_CM12_1
-___==:I____-
_-` -` ,_ .--,
Recall: The discrete solution is obtained
in an iterative manner.
a An initial solution is established.
® The solver updates the solution each
iteration by attempting to satisfy the
discrete equations being solved.
IPL±iL=
i -'--i-i
Page421 CTO9101 CM121
system of equations.
© Steady state simulations only involve
this step and end when converged.
ri v,in ri
ln a discrete simulation, however, there can be differences between the two
fluxes. The`se are referred to as Residuals:
r=Z/!uxesm,simu!afion-I/Zttxesout,Simu!ation
The residual in each cell indicates how well the discretized equation is satisfied.
Residual values are limited by machine round off.
Residuals are a measure of the overall solver progress and should decrease
as a simulation progresses.
Residuals
i 2 '3 4 5 6 8 9 10
lteration
AND
Page425 CTO9101_CM12_1
C g St dy mit SM
Reports Plot
1e-06l
i i ; .i . ` ~\ (
Cylinder
temperature
appears to be
converged.
Stopping criteria based on monitors use the monitor data as well as logical
triggers to stop the simulation or advance the solver to the next time step (for
transient models).
EEREERE
converge the new flow conditions.
HRE © Once converged, the Outer Loop
HEEEERE steps forward again, and the
qE T process repeats until the simulation
Lrm-ffiRE
3unn;
is complete.
¥
-g|ngrf-RESTRE;apRE
•^J.
'teration
Proper inner loop stopping criteria are critical for effective transient simulations.
© Advancing too early (before convergence) compounds system error.
® Advancing too late wastes solver time after the time state js converged.
r-----.\E-
lte,atlon
ltel.atlon
I .) ,-,, A--
E mpl Ad g tnt it M
Page434 CTO9101_CM12_1
The solver outer loop advancement interval is determined by Time-Step.
© Time-step should be chosen to resolve flow phenomena and maintain stability.
a Courant Number is a measure of how far the fluid travels in one time-step.
a For time accurate simulations (and all Explicit simulations), choose time-step to
maintain Courant number below 1.0 for all cells, passing no more than one cell
length per tjmestep.
ve,oc,ty)A£¥step
size
Courant number = ®
A%\ L ttA£
characteristic
cell length
LAx-
Page435 CTO9101 CM121
The segregated solver uses under-relaxation factors (URFs) to control the solver
progression at each iterati'on.
When a new solution is calculated, it can be blended with the previous solution to
yield a more conservative answer.
© URF = 1 uses the new solution only.
© URF = 0.5 uses 50% each of the previous and current solutions..
•h= F=_=====
- jB Soivers
Coupled froplleit , Properties X
:-,ngREdiit-
Page441 CTO9101 CM121
If this error continues to grow, the simulation can become unstable and diverge.
© Irregular, increasing residuals signal the solution is diverging.
© When this happens STAR-CCM+ will output an error message and stop.
Residuals
lteration
ft Wall B`istance
a-+#^ S€gregmed Flow
dr `fr Segregated Energy
th ¢ rL-Epsilon Turt)ul€nc€
Activate Retain Temporary Storage for the solvers to gain access to additional
terms computed during update.
a Residual field functions, Ap Coefficients, Corrections, Reconstruction
Gradients, Gradients.
a Can be used to locate sources of errors.
D g L tg SM S
11`
Page446 CTO9101_CM12_1
Mesh quality affects the accuracy of cell-to-cell fluxes and solver stability.
Mesh metrics are available as scalar functions for thresholds, reports, and
scenes.
gej{ •`1*.S;
Face
dg
Centroid 1
Plotting mesh metrics helps identify where the mesh should be improved.
a-8 1.0
I_ ___
mesh as well.
? a rtl€shers
Available options for improving mesh quality for the • SurfaceRfmesher t
Optimizer cycles/threshold
® Increase the optimization cycles and quality threshold f Expert
Field Furic{ion Refinfm€m Table None +
to improve the polyhedral mesh quality.
® This can greatly increase the time to generate the
Optimization cr/cl8s 3
)
how the solution is reached
•<^. `i..
Make mesh size changes gradual + improves solver stability and convergence.
¥___i_
Page455 CTO9101_CM12_1
E
Page456 CTO9101_CM12_1
All viscous fluid flows have large gradients in velocity, temperature, etc. near
walls, called boundary layers.
I
I-i)
r,//,///////////////,////,/,,//„//`
/,,/////,,///////////,
Laminar boundary Turbulent boundary
layer layer
=-
y+=5 y+=sO
=-;=
f
lf Wall Y+ < 5, the near-wall region is lf Wall Y+ > 30, the single near-wall cell
resolved enough to model explicitly. has an assumed velocity profile
Low y+ Wall Treatment is appropriate - 8- determined by wall functions.
30+ prism layers typical. High y+ Wall Treatment is appropriate -2-
4 prism layers typical.
IImEE
Page459
Wall Y+ values between 5 and 30 are not recommended.
CTO9101 CM121
Page460 CTO9101_CM12_1
Ensure boundary conditions and values accurately represent the problem
statement.
© Boundary values are imposed on the cell face.
© If incorrect values are applied the wrong problem is being solved.
® Example: Wrong boundary type or value is applied for inlets/outlets.
Inflow / outflow boundaries placed too close to areas of high flow gradients may
result in improper modeling of the physical problem.
Reference Density -997 kg/m^3 (water) Reference Density -1.18 kg/m^3 (air at 25 C)
Total Pressure - 0 Pa Total Pressure - 0 Pa
Page463 CTO9101 CM121
What model details should be inspected before running to verify the model will
produce the desired solution and accuracy?
A simulation of a rocket nozzle will provide the platform for refining simulation
techniques and gradient management ideas. The simulation will be set up to
converge efficiently and in a stable manner.
Page468 CTO9101_CM12_1
Problem Statement: A rocket nozzle is operating under subsonic test
conditions. A large inlet pressure drives the flow towards the low pressure
downstream outlet.
Ambient Pressure
Ambient Pressure
__-:____--_ - ____-___-:-i
Total Pressure: I
25'000 Pa
Design Requirement: Flow must remain subsonic (Mach < 1) in the nozzle.
Page470 CTO9101_CM12_1
Prepare the Rocket Nozzle geometry part for 2D meshing.
Create and execute a Badge for 2D meshing operation (Operations > New >
Mesh) for the Rocket Nozzle part.
Requires a separate face coincident with the z=O plane (already present
here).
The axis of symmetry will be defined in the physics models.
Assign Parts to Regions using the Create One Region for All Parts and
Create A Boundary for Each Part Surface options.
Axis
The domain includes space for the jet to expand radially and axially.
? Ei Regions
a * RocketNozzle
Assign boundary types before meshing. S Gi Boundaries
id Centerline
; Stagnation lnletto the Inlet patch.
I Defau,t
H ,n,et
Axis boundary condition to centerline. 11,,I Out,et
Feature Curves
er a Ph.¢sics Conditions
gr GE Phvslcsvalues
a Denvedparts
Discussion Point
How should the mesh be refined?
Page474 CTO9101_CM12_1
Geometry
Create parts for volumetric controls by right-clicking the P G]
F= E 3D-CADModels
Geometry > Parts node and selecting New Shape Part. # E Parts
gr ire Jet Refinement
Cylinder (Nozzle Refinement). th fi Middle RefinE!ment
Start Coordinate: [0,0,0]. tr a Nozzle P`efinement
tr ffi.Rocket ND=zle
Radius: 0.6 in.
End Coordinate: [3,0,0].
Cylinder (Middle Refinement).
Start Coordinate: [1.6,0,0].
^ End Radius: 1.0 in.
End Coordinate: [10,0,0].
Cone (J;t Refinement).
Start Coordinate: [3,0,0].
Start Radius: 0.5.
` End Coordinate: [10,0.8,0].
End Radius: 1.0 in.
Add Custom Controls to further refine the nozzle, middle, r=' Gi operations
@ Badge for 2D Meshing
L=j fa Automated Mesh (2D)
jet mesh regions. i3i ii Meshers
Discussion Point
How can we quantify if the mesh is valid?
Metric Value
Face Validity =1
Skewness <850
Angle
a Three Dimensional
Axisymmetric is its own 2D model, deselect a Two Dlmemsional
qu a Feature Curves
ls there anything challenging about these conditions? a-ii Physlcs condltions
gh a Phys)csvalues
Primary outlet
P - 0 PcL
User Field functions are algebraic expressions that can change value during
the simulation.
a User-definable math with scalars, vectors, higher order tensors.
a Can be derived from other quantities (velocity, pressure, etc.) or stand alone.
a Used to perform internal calculations / implement equations.
Field functions can be used for intelligent initialization and boundary conditions.
? Properties
Function Name UsErFieldFimctiQn_1
Open the Definition editor by clicking
Type Scalir
Dimensjons Dimensionless H for more advanced editing.
Deflnition 0
Tags []
S Expen
Inverse Distarice weight I
Assembly Code
ignore Boundary valLles I
Double-clicking an
existing field function in
the list adds it to the
definition box
EL
givFSJu-rT}|-
Existing field functions are chosen from here
S{TEfflpEraturE3}
Vectors use double S, with [brackets] added to call a single component direction.
S S [VEl oEi ty i [1 ]
x = [0]
y = [1]
z = [2]
Page484 CTO9101_CM12_1
System Functions
Trigonometric Eni3{TffiTVTITo+cTtT#)I
Exponentia`l
Page485 CTO9101_CM12_1
_-eT=e_|=. _`_.______I
FEa=Fai:-_a=;=c:J=TgTafr,,-ai,-ie-atE,i;-i:::vee:,s::::iir==fe-
i
i
lntemalwall Heat Flux cci
Internal wall Heat Flux co
ln{ernalwallHea!Flux.. Potuer/Leng`nA2-Te
lntermlwallHeatFlux Powcer/Lerigth^\2-Te
EEEEEi
I,I,I. xl `,`1
Page486 CTO9101_CM12_1
SIEMENS
Page487 CTO9101_CM12_1
P H Bodyllri{et
a-ii my!Ics C.rtdrlior3!
g lil Ph,`sicsvalucs
Page488 CTO9101_CM12_1
SIEMENS
Centroidpr] (in)
0.0000 1.9200 3.8400 5.7600 7.6800 9.6000
9 ® Phy5i[Sl
dr i Models
a-a Referencevalue5
p a Initial conditlons
• ifeedha;
Assign the field function to the continuum's initial • StaticTemperatur€
• Turbulence Intensity
conditions. • Turbulence specificatlDn
t Change the Initial Conditions for Pressure Method • Turbulent velocity scale
• Tiirbulent viscosity Ratio
S F.ropertiE5
!Method Field Function T
5talar Function
: D i in i 11 S i a n ± Fi r ,= L= L" r t
Initialpressure
0.0000 1000.0 2000.0 3000.0 4000.0 5000.0
For monitoring convergence, we will look at the inlet mass flow rate in the nozzle
and overall mass imbalance on all inlets/outlets.
Confirm that velocity does not exceed the Speed of Sound on the centerline. To
do so, create an XY Plot Of the centerline Mach Number.
' Create a new XY Plot and choose the Centerline boundary for the Part.
. Change the X Type to Scalar and select Centroid[X] for the scalar value.
\ Choose Mach Number (in the Laboratory Frame) for the Scalar Function
under Y Type > Y Type 1.
Under Axes > Left Axis, set the Maximum to 1.0.
Optional:
Change the Plot Title to Centerline Mach Number and the Bottom Axis Title to
Axial Position, the Left Axis Title is already Mach Number. I
Questions
® Has the simulation reached convergence?
a \/Vhat is the .maximum Mach number on the centerline?
i:REillRE_
Question
a Was the model converged at 1000 iterations? If not, what has changed?
® Did we achieve the design requirement of maintaining subsonic flow in the
nozzle (Mach < 1.0)?
The domain \should include all areas that affect regions of interest
~tf3fjj/Ffrfff.»gyssgr_fu,-Zj4:::iarff$5K3if:aror§sef?;ffffi:§se;{:s/if!£esi±sffi::/f]f8::::fgi;::I_~_rsIfi:±!s#f:/!¢r!r:fi_srw_iJJSSK^:±aeii§:psRTr^fRE_J!3r:isf»ENffREr=_.___
Page508 CTO9101_CM12_1
SIEMENS
Passive Scalar
Page512 CTO9101_CM12_1
STAR-CCM+ offers multiple methods for simplifying, repeating, and automating
processes.
® Leveraging existing .sim files.
® Custom Simulation Trees.
® Java macros.
© Simulation Assistant.
© Design Manager.
Parts based meshing allows users to build and manipulate complex, automated
workflows to prepare and hiesh any input model.
g ii Geometry
Ei
ch ii Parts
3D-CAD Models
a Add operations to create a pipeline.
oL ii Descrlpt!ons
a Operations can be reordered.
6ii Contacts
:t:i::i:n:g:::rEL
Objects and/or properties can be copied and pasted across sim files.
Files can be opened in versions newer than in which they are saved.
Macros can:
a Automate routine tasks to increase speed and reliability of workflows
© Run complex tasks and simulations in batch mode.
® Be recorded interactively using the build tools in theToolbar: + ®
The Java API Documentation lists available classes and methods and is
accessed from the Help menu:
Overview
I am customer using STAR-CCM+. I work at BetterDesignsFaster Aero Systems, in the ECS
(Environmental Controls System) group. My job is to design the bleed air distribution system for a large
commercial aircraft. A summary of the system is shown in the figure:
Page522 CTO9101_CM12_1
Model Issue Summary
I setup and ran the analysis of the static mixer in STAR-COM+. After some initial problems with model
divergence, I was able to obtain a converged solution. I followed your wall y+ recommendations,
avoiding the 5 < y+ < 30 area. I found that I needed to use the "Enable Enhanced Stability Treatment"
under the Segregated Flow Solver or the code was highly unstable.
The results of the simulation and the final product physical tests are in the table:
The product was sent to production because the simulation results met the design specifications, The
system failed.in the final validation test, as outlined in the above table.
Can you help me -understand why the CFD predicted results were largely different than the physical
validation test?
My file StaticMixer_Finalsolution.sim is provided in the folder 12_StaticMixer.
Kind regards,
BDF AERO
Page523 CTO9101_CM12_1
_. What will you check first when you open the sim file?
Design a test plan of items you will check in the simulation to diagnose errors.
How will you assess the quality and suitability of the mesh?
Page524 CTO9101_CM12_1
You are a CFD analysts at BDF Motors (BetterDesignsFaster Motors). For the
new model engine, you need to determine the solid temperature distribution of
the exhaust manifold shown here to ensure the design meets our required
specifications.
Simulation Obiectives:
Predict the flow and thermal performance of the exhaust manifold at the
given design point.
Page526 CTO9101_CM12_1
Runner 1
Mass Flow Rate = 0 05 kg/s
Tc)tat Temperature = 514 5 C
Turbulence Intensity = 10%
Turbulent Viscosity Ratio = 10
Outlet
Absolute Pressure = 1 atm
Static Temperature = 400 C
External Surfaces
Static Temperature = 75 C
Convective Heat Transfer Coefficient = 50 W/m2~K
Our structural analysis team has set the following manifold requirements:
Material: UNSG 101000 (Carbon Steel) and Air.
Maximum Temperature in solid must remain below 420 °C.
suitable?
re inc.
Page528 CTO9101 CM121
Run the simulation in STAR-CCM+.
Once the model has finished running, review the design requirements. Have we
achieved our objectives?
Which simplifying assumptions regarding the physics have been made? How do
these affect the accuracy of the predictions?
After you've run the baseline simulation, assess the design to determine if
the design requirements have been met. If they have not, here are some
ideas for modifications that might improve performance:
How do you get help when you aren't able to solve a simulation issue on your
own?
Page532 CTO9101_CM12_1
The documentation can be accessed in three locations:
HEEDS ,
Page533 CTO9101_CM12_1
H bag-€lihffiun
tr il C€ometrv
ur ffi €ominua
giv iB REgion5
gr i| Interfaces
gr iB Berwedparl5
th iB S®l'€r5
> -ii.;p.6,`ng ci,eL.` -.-----.-----------.
a Solu{i®n Hi%torifg
as i 5SILliiofi \¢'j€w£
* EL B€poil5
fr H Man!t®rs
rfu il Pl®t5
giv E 5£EnEs
i Summarle5
er a Repre5emalions The user guide section on Stopping Criteria
dr iB TQcls
is opened in a browser window.
Page534 CTO9101_CM12_1
SIEMENS
Iun-
l:±A±_I
-`fro.TcomAi]co.meFlee
1pet-in
1rtk-T-
10-
Im
I-nefh„
1a-Fha
I w.- inm.in o`4 rmtgiv
I-Fb„
1-
- oeou. Efroedi Armed
Icortyfty
I,qu3un
--'r.
- I --_. i
Page535 CTO9101_CM12_1
r-
ontact in orma Ion or
your Siemens PLM Recently opened support cases
representatives
I
" == ~ ryy"®=zi=acL7,
Newl xp= _-- rm Company news
and tweets
®®®,,®,,®,,
Reconl Artldes
EmmplG ol a Masonry Best praotlc®D loJ Full
H®-Co.SinSlm
Toclinlcal Forum
links to Technical New and featured videos
Forum and Macro Hut MacroHut tryo and articles
', -Ov-'»'IT,-
RemaininoHoursonboD
Page536 CTO9101_CM12_1
DSE: Dedicated Support Engineer
Your contact for technical assistance with the software such as:
a modeling processes and procedures,
© software features and capabilities,
© simulation issues.
Account Executive
Your contact for business assistance such as:
a license upgrades, server migration & renewal,
a product a`nd training quotes,
a account plahning,
S help achieve simulation goals quickly and effectively,
a on-site support and services,
• burst capacity.
Search fry sLiqeEL ease nLJmter or desengiven a OPEN CASES I ClosED l=ASES
Lri ul
Currently open
support cases
hiul
E] rty Group'8 Cases
Knededge Base
video Library
Page540 CTO9101_CM12_1
General information about STAR-CCM+ and HEEDS.
Software downloads. products The
StJift-CCH + in meas
Tbankyqutrfrog
submdted, \deo er caTimenled
an 8`e LIO& Cue
gife:'#&¥o%¥iFgrtryty
Boafds
Page543 CTO9101_CM12_1
ancl whose competency can be adequately demonstrated lhrough clooumentable evidence wlll recelve an assessment-based certificate that confirms (he users
demonstrated level of competency as judged by exanis and/or evaliiations developed by Slemens PLM Sotware's Sl-AR-COM+ Tralning Advisory Group. Competency
levels lncliide Blue (beginner), Bronze (lmermedlate I)` Silver (intermedlate 11), Gold (advanced), and Platlnum (expert),
toELurty„
lf you need to contact your DSE, whether through the Steve Portal, phone or e-
mail, providing good information will expedite the process.
What is the issue you are having -clearly state the question.
Any information that will help the DSE diagnose the issue:
a things you h.ave tried and the outcomes,
® software version,
a operating system,
a log files,
a User Guide > Using STAR-CCM+ > Getting Support.
Page547 CTO9101_CM12_1
Page548 CTO9101_CM12_1
SIEMENS
laXz*.-rl7 {'. L;{z
\-
t
I
SIEMENS
Coldplate workflow workshop - review questions
Was the desiqn objective achieved?
No, temperature is too high.
Studv the Maximum Temperature plot and Wall Temperature scene. Describe below what
you think has happened phvsically in this model.
. Maximum temperature converges as the solution reaches a steady state value.
' Wall temperature varies because the fluid is heating up as the water flows through the
pipe and because the heat transfer coefficient varies as the water travels through the
turns in the pipe.
At the beqinnino of the simulation. what is the wall temperature associated with a heat flux
of 30.000 Wmn2? Where in the simulation can this information be found?
The initial condition makes the wall temperature constant at the beginning of the simulation
and is independent of the boundary condition.
Page550 CT09101_Cl\i112_1
SIEMENS
\Mat are some wavs in which accuracv for this simulation could be assesed?
' Do a grid dependency study.
. Check the length of the pipe and do a simple hand calculation to make sure that energy
balance is correct.
. Check residuals.
. Check convergence history of the maximum temperature.
Page551 CT091o1_Clvl12_1
i
SIEMENS
Geometry workshop - discussion points
U/hat is causinq the pierced faces? The pierced faces occur where the two parts touch -
minor intersections.
Are any other contacts required? Yes, Contacts are required between all of the parts that
will be meshed.
Page552 CT09101_CM12_1
SIEMENS
Geometry workshop - review questions
What operation was used to fix pierced faces?
Imprint. Pierced faces are commonly caused by separate parts touching or intersecting
each other. The lmprint operation aligns surfaces that are supposed to touch, eliminating
those pierced faces.
What steps were completed before the fluid volume could be extracted?
Before extracting the volume the holes needed to be filled, and curves needed to be split to
be used as inputs for the fill holes operation. Alternatively surfaces could have been split
for the flll holes operation.
SIEMENS
What are the advantaoes of a pipeline of Operations for: qeometry processing compared to
other methods?
The original parts are preserved and the input parts can be changed and the steps re-
executed.
Page554 CT09101_CM'12_1
T
n
SIEMENS
Meshing workshop - discussion points I
Page555 CT09'l0l_CMl2_1
How do the surface mesh sizes influence the volume mesh sizes?
The volume mesh sizes should mirror the surface mesh sizes. Since the volume mesh is
built from the surface mesh faces, the sizes should correlate near the boundaries.
Which boundaries in the modeldo not have prism layer cells? Wrv?
lnlet and outlet because of their type. PL are generated only on walls,
Page556 CT09101_CM12_1
SIEMENS
Meshing discussion - review questions
What is discretization error and how can it be measured?
the difference between the exact solution of the differential equation (mathematical model)
and the exact solution of the algebraic system of equations obtained through discretization.
Discretization error can be estimated by performing systematic mesh refinement and
comparing the solutions.
\--
Page557 CT09'1o'l_CM12_1
\-
Page558 CT09101_Cl\i112_1
SIEMENS
Meshing ll workshop - review questions
How are mesh sizes adiusted for specific areas onlv?
Custom controls are used to modifly the mesher behavior locally; however, we saw that
changes to the Default Controls (like the reduction in the Target Surface Size) can also
result in local refinements. The difference is that Default Controls apply globally in the
domain, while Custom Controls are only applied to their input entities.
Whv did we implement the custom Part Control for the Thin Mesher?
The Custom Part Control for the Thin Mesher was used to explicitly turn off the Thin
Mesher in areas that we did not wish the mesher to be active. The thin mesher increases
meshing time and can have undesired effects (thin-thin interfaces have no prism, mesh
quality at transitions from thin/bulk can be affected, etc.), so it is a general best practice to
explicitly tell the mesher which regions should be active for the Thin Mesher.
Page559 CT09'101_ClV112_1
t--
I SIEMENS
Meshing ll discussion - review questions !
Which custom control miqht be used to refine a portion of a sinqle part surface?
Volumetric controls are the obvious ones here, though Curve Controls could also work in
some cases.
Which surface remesher parameter should be modified on the far-field boundary (far from
the bodv of interest)?
Target size
What prism laver parameter should be modified if prism layers have been retracted in an
area that was not intended?
Minimum Thickness Percentage (and possible Gap Fill or Layer Reduction) would be a
global setting, but this can be done using the total thickness, distribution, and number of
layers as well.
:l'':i.l '...r.::]iaraii..:-,:ju',,at:r&!trii.rq4nrit;rati,r:,':tlir.:
Page560 CT09'101_CM12_1
Activity - mesh
Which Surface Remesher Control(s) was modified to chanqe the mesh below?
Surface proximity was increased from 2 points to 10 points"
Page561 CT09'101_CM12_1
SIEMENS
Physics workshop - physics models
What is the default material for liquid? Does it fit our requirements?
Water. Yes
What is the default material for solid? Does it fit our requirements?
Aluminium. Yes.
Page562 CT09101_CM12_1
SIEMENS
Physics workshops - review questions I
\Nhen are separate phvsics continua required?
Separate physics continua are required when different physics models need to be applied
to different regions. A single continua can be applied to more than one region if the models
are correct for each region. Multi-part solids and multiphase fluids are advanced models
not discussed here.
Page563 CT09101_CMlr_1
SIEMENS t-
I
Page564 CT09101_CM12_1
av
\--
SI EM ENS
v
\/
U/hy is the hiqhest temperature located where it is?
v The high temperature is there because it is the furthest from the cooling channel (z-
direction) and the cooling is uneven as the water temperature increases along the length of
the cooling channel (y-direction).
Assuminq the boundarv conditions cannot be chanoed. how can the desiqn be modified to
reduce the maximum chip temperature?
More turns.
Page565 CT09101_CM12_1
How can the desiqn be modified to have more areas of hioh heat flux?
The fluid heats as it moves through the pipe so the heat flux progressively decreases for
each subsequent turn.
i
I
r--
I
SIEMENS t
I
What obiects had to be created. and in what order, to monitor temoerature converqence?
To monitor temperature convergence, first create reports of the temperatures of interest,
then monitors and plots from the reports.
How was it confirmed that the desiqn criterion was not met with the oriqinal desiqn?
The maximum chip temperature report confirms the chip is too hot somewhere. A scene of
temperature on sections, a scene of temperature on the chip boundaries, and a scene of a
temperature threshold all confirmed that the chip temperature exceeded 45C and showed
where.
When the oeometrv is chanqed. what needs to be updated in the operations before runnino
the pipeline? ls there a waLto avoid this step?
The operation input parts need to be updated. This can be avoided if dynamic queries are
used to populate the operation inputs.
Page567 CT09101_CM12_1
SIEMENS
Entire coldplate workshop - review questions I
\Nhat are some assumptions that were made in this analvsis that mioht affect prediction
accuracy?
We didn't simulate external environment, did not include effect of adhesive contact
resistance, wally+, geometry was simplified - chip was assumed as pure silicon.
lf you wanted to explore the desion, what thinos could be investioated with a desiqn sweep
or optimization studv?
Sweep water inlet temperatures, optimize geometry of cooling tube, heatsink main body,
location of chip on heatsink.
Page568 CT09101_CM12_1
SIEMENS
Page569 CT09'101_CM12_1
SIEMENS
Transient workshop - review questions
Were desiqn criteria met?
Yes. The maximum chip temperature did not exceed 50 C during the transient heat-up
period.
U/hich solver settinq added by a transient time model should ALWAYS be considered prior
to runninq the model?
Time-step always needs to be chosen intelligently in transient models and should not be
left at the default value.
How are the stopping criteria different for steady models and transient models?
For steady models stopping criteria are only needed to stop the simulation. ln transient
models, stopping criteria are required to stop the time-step as well as the simulation.
data samplinq control needs to be chosen correctlv for monitors and plots to appear
\y'Vhat
as expected?
Triggers control the data sampling.
Page570 CT09101_CM12_1
SIEMENS
Ducted fan workshop - planning
Are any additional surfaces needed to account for boundarv conditions or mesh
refinements?
Yes, inlet and outlet.
We set the boundary types for the inlet and outlet before meshino. Do you recallwhv?
Do not create prism layers on these boundaries.
SIEMENS
Ducted fan workshop - mesh
ls the surface mesh acceptable?
The surface mesh on the leading edge of the blade is choppy (not fine enough), requiring
finer cells to resolve that curvature.
Also, the surface size on all of the fan parts (blades and hub) should be capped to avoid
large cells - look especially at the hub and at the trailing edge of the blades.
Other features that could be noted are the proximity refinement causing smaller cells on
the duct walls near the blade tips, and the excessive growth from those cells to the rest of
the wall.
Page572 CT09101_CM12_'l
Ducted fan workshop - discussion points
How manv cells were added when the refinement zone was included?
-100,000
.\-
U/hich other enqineerino quantities rnioht be of interest in this simulation?
'\- Can you create a report to measure Power required? (Power = Torque * Rotation Rate)
1 Surface average density report on inlet.
Page573 CT09101_ClV112-1
\-.
\- SIEMENS
Ducted fan workshop - discussion points
Could another Derived Part better display velocitv than the current planes?
ln the image shown, I'm using a LIC (line integral convolution) on a cylindrical plane in
order to really zoom in on the flow around the blades. A cylindricql plane section with a
radius of around 0.075 m is perfect for showing this.
Page575 CT09101_CM12_1
srEMENs
Ducted fan workshop - questions I
How will we judqe if the model is converged at the end of the run? ls there a wav that could
be incorporated into the Stoppinq Criteria?
Stopping criterion from monitor.
Does this desion meet our design requirements for flow rate?
Yes, requirements (> 200 CFM) are met: Report returns 208.78 CFM
Page576 CT09101_C|VI12_1
Ducted fan workshop - questions
i sreruerus
I
t
I
I
Can you improve the current desion to reduce the torque required while still meetinq the
desiqn requirement of a 200 CFM (0.11 kq/s) flow rate?
A design with 5 blades and a hub radius of 0.04 m will achieve the design requirement of
200 CFM (202 CFM or 0.1125 kg/s) with a lower Torque than the baseline (0.008 N-m vs
0.009 N-m). Though there are numerous other designs that may also work.
Page577 CT09101_CM12_1
!, SIEMENS
Effective sim u lation discussion
\I/hat model details should be inspected before running to verify the model will produce the
desired solution and accuracv?
Prior to running one should inspect the: boundary conditions, initial conditions, reference
values, mesh resolution near flow structures and walls (including boundary layers), mesh
quality.
Page578 CT09101_Cl\i112_1
SIEMENS
Effective simulation workshop - discussion points
Was the model converqed at 1000 iterations? lf not. what has chanqed?
No, now most don't drop any more although they start to oscillate.
Page579 CT09101_CM12-1
I
StElMEtUS Simcenter
l*Vzt*a;{y{--h.b STAR.GGM+
6 ( -omputatronal Combusttr-rn 2
8 Iiulenan ]Iuluphase 2
9 Lagtanstan Niultrphase 2
1l Ae roacoustics 2
t2 Fltud-Strucrut e In teractions 2
1j \-irrual Tou'Tank 2
1+ ST'-\R-( -D lroundarion 2
15 Rasic cs - ice J
4 Wizard Creation 1
6 STAR-ICE 1
\_-
\__
\-
\-.'
\-.
\-
\.-.
\r-
\__.
L-
\-,
\-
\-
\-
'!-
\-
\,
t-.
\.,
\-
\_-
\_-
\-,
\-
,\-
\_-
'v
!.
\!-