Solid Works Flow Simulation 2009 Tutorial
Solid Works Flow Simulation 2009 Tutorial
idWor
ksFl
ow Si
mulat
ion 2009
Tut
ori
al
Contents
Rotating Impeller
Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-1
SolidWorks Model Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-2
Project Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-2
Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-3
Specifying Stationary Walls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-4
Impeller’s Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-5
Specifying Project Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-5
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-8
CPU Cooler
Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-1
SolidWorks Model Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-1
Project Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-2
Computational Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-2
Rotating Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-3
Specifying Stationary Walls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-5
Solid Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-6
Heat Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-6
Initial Mesh Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-6
Specifying Project Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-9
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13-11
9 9 9 9
Determination of Hydraulic Loss
9
Cylinder Drag Coefficient
Mesh Optimization
Textile Machine
Rotating Impeller
in the tutorial examples. To learn more about the usage of a particular feature, read the
1
9 9 9 9 9 9 9 9
Below is the list of the physical and interface features of Flow Simulation as they appear
CPU Cooler
Features List
Non-Newtonian Flow in a Channel with Cylinders
Mesh Optimization
Rotating Impeller
Textile Machine
CPU Cooler
ANALYSIS TYPE
External analysis 9 9 9
Internal analysis 9 9 9 9 9 9 9 9 9 9
PHYSICAL FEATURES
Liquids 9 9 9 9 9
Gases 9 9 9 9 9 9 9 9
Non-Newtonian liquids 9
Multi-species flows (or more than 9 9
one fluid in the analysis)
Fluid Subdomains 9
Heat conduction in solids 9 9 9 9 9
Heat conduction in solids only 9
Gravitational effects 9
2
Rotation
Radiation
Symmetry
Roughness
Porous media
Dependency
CONDITIONS
Laminar only flow
particles or droplets)
Velocity parameters
Computational domain
Local rotating regions
9
First Steps - Porous Media
9
9
9
Cylinder Drag Coefficient
9 9
Heat Exchanger Efficiency
9
Mesh Optimization
9
Application of EFD Zooming
9
9
Textile Machine
9
9
9
3
9
9 CPU Cooler
4
Wall
Real wall
Concentration
Flow openings
Inlet velocity
Static pressure
Solid parameters
Pressure openings
Boundary conditions
Environment pressure
Thermodynamic parameters
9
9
First Steps - Ball Valve Design
9
First Steps - Conjugate Heat Transfer
9 9
9 9
Determination of Hydraulic Loss
9
Cylinder Drag Coefficient
9
9
9
Heat Exchanger Efficiency
9
9
9
Mesh Optimization
9 9 9
Application of EFD Zooming
9
9
Textile Machine
9 9
9
Non-Newtonian Flow in a Channel with Cylinders
9
9
Rotating Impeller
9 9
9
9
CPU Cooler
Fans
Heat sources
Solid material
Dependency
Surface sources
Porous medium
Volume conditions
Initial conditions
Fluid Subdomain
Solid parameters
Velocity parameters
9
9
First Steps - Conjugate Heat Transfer
9
First Steps - Porous Media
9
9
Heat Exchanger Efficiency
9
9 9
Mesh Optimization
9
9
9
9
9
9
Textile Machine
9 9
9
9
9
Heated Ball with a Reflector and a Screen
9
Rotating Impeller
5
9
CPU Cooler
6
GOALS
Global goal
Surface goal
Clone project
From template
General settings
Temperature
Blackbody wall
Volume sources
Whitebody wall
Radiative surfaces
PROJECT DEFINITION
9
First Steps - Ball Valve Design
9
9
9
First Steps - Conjugate Heat Transfer
9 9 9 9
Determination of Hydraulic Loss
9
9
9 9 9
9
Cylinder Drag Coefficient
Mesh Optimization
9
9
Application of EFD Zooming
9
Textile Machine
9
Non-Newtonian Flow in a Channel with Cylinders
9 9 9 9
9
9
Rotating Impeller
9 9 9 9 9 9
9 9
9 9 9 9 9 9 9 9 9 9 9 9 9
9
CPU Cooler
Point goal
Initial mesh
Volume goal
Equation goal
Automatic settings
Narrow channels
Minimum gap size
Manual adjustments
Manual adjustments
Solid/fluid interface
Level of initial mesh
9 9
9
First Steps - Conjugate Heat Transfer
9
First Steps - Porous Media
9
Determination of Hydraulic Loss
9
Cylinder Drag Coefficient
9
9
9
9
Mesh Optimization
9
Application of EFD Zooming
9 9
Textile Machine
9 9 9 9
9
9
9
9 9
Rotating Impeller
7
9
9
9
9 9
9
9 9
9
CPU Cooler
8
TOOLS
Toolbars
Filter faces
Dependency
Custom units
Check geometry
Refining cells
Component control
Narrow channels
User-defined items
Engineering database
9 9
First Steps - Porous Media
9
9
Determination of Hydraulic Loss
9
9
Cylinder Drag Coefficient
9
Heat Exchanger Efficiency
9
Mesh Optimization
9 9
9
9
9 9
9 9
9 9
9
Textile Machine
9
Non-Newtonian Flow in a Channel with Cylinders
9
Heated Ball with a Reflector and a Screen
9
Rotating Impeller
9 9 9
9
CPU Cooler
XY plot
Cut plot
Preview
Goal plot
Goal plot
Batch run
Isosurfaces
Surface plot
Particle study
Flow trajectories
Surface parameters
Display parameters
GETTING RESULTS
MONITORING CALCULATION
9
9
9
9
9
First Steps - Ball Valve Design
9 9
9 9
First Steps - Conjugate Heat Transfer
9 9 9
First Steps - Porous Media
9
9
Determination of Hydraulic Loss
9 9 9 9 9
9
9
Heat Exchanger Efficiency
Mesh Optimization
9
9
9
9
9 9 9 9
Textile Machine
9
Non-Newtonian Flow in a Channel with Cylinders
9
Rotating Impeller
9
9
CPU Cooler
10
Vectors
Contours
OPTIONS
Isosurfaces
View settings
Display mesh
Display mode
Transparency
Apply lighting
Flow trajectories
9
9
9
First Steps - Ball Valve Design
9
9
First Steps - Conjugate Heat Transfer
9
First Steps - Porous Media
9
9 9 9
9
Determination of Hydraulic Loss
9
9
Heat Exchanger Efficiency
9
9
9 9
Mesh Optimization
Textile Machine
Rotating Impeller
CPU Cooler
1
First Steps - Ball Valve Design
This First Steps tutorial covers the flow of water through a ball valve assembly before and
after some design changes. The objective is to show how easy fluid flow simulation can be
using Flow Simulation and how simple it is to analyze design variations. These two factors
make Flow Simulation the perfect tool for engineers who want to test the impact of their
design changes.
1 Copy the First Steps - Ball Valve folder into your working directory and ensure that
the files are not read-only since Flow Simulation will save input data to these files.
Run Flow Simulation.
2 Click File, Open. In the Open dialog box, browse to the
Ball Valve.SLDASM assembly located in the
First Steps - Ball Valve folder and click Open (or
double-click the assembly). Alternatively, you can drag
and drop the Ball Valve.SLDASM file to an empty
area of SolidWorks window. Make sure, that the default
configuration is the active one.
We utilize this model for the Flow Simulation simulation without many significant
changes. The user simply closes the interior volume using extrusions we call lids. In
this example the lids are made semi-transparent so one may look into the valve.
Click Next .
3 Choose the system of units (SI for this
project). Please keep in mind that after
finishing the Wizard you may change
the unit system at any time by clicking
Flow Simulation, Units.
Not only will Flow Simulation calculate the fluid flow, but can also take into account
heat conduction within the solid(s) including surface-to-surface radiation. Transient
(time dependent) analyses are also possible. Gravitational effects can be included for
natural convection cases. Analysis of rotating equipment is one more option available.
We skip all these features, as none of them is needed in this simple example.
Click Next .
1-2
5 In the Fluids tree expand the Liquids item
and choose Water as the fluid. You can
either double-click Water or select the
item in the tree and click Add.
Flow Simulation has an integrated database containing several liquids, gases and
solids. Solids are used for conduction in conjugate heat conduction analyses. You can
easily create your own materials. Up to ten liquids or gases can be chosen for each
analysis run.
Flow Simulation can calculate analyses with any flow type: Turbulent only, Laminar
only or Laminar and Turbulent. The turbulent equations can be disregarded if the flow
is entirely laminar. Flow Simulation can also handle low and high Mach number
compressible flows for gases. For this demonstration we will perform a fluid flow
simulation using a liquid and will keep the default flow characteristics.
Click Next.
6 Click Next accepting the default wall
conditions.
You can also specify the desired wall roughness value applied by default to all model
walls. To set the roughness value for a specific wall, you can define a Real Wall
boundary condition. The specified roughness value is the Rz value.
Result Resolution is a measure of the desired level of accuracy of the results. It controls
not only the resolution of the mesh, but also sets many parameters for the solver, e.g.
the convergence criteria. The higher the Result Resolution, the finer the mesh will be
and the stricter the convergence criteria will be set. Thus, Result Resolution determines
the balance between results precision and computation time. Entering values for the
minimum gap size and minimum wall thickness is important when you have small
features. Setting these values accurately ensures your small features are not “passed
over” by the mesh. For our model we type the value of the minimum flow passage as the
minimum gap size.
Click the Manual specification of the minimum gap size box. Enter the value
0.0093 m for the minimum flow passage.
Click Finish .
Now Flow Simulation creates a new configuration with the Flow Simulation data
attached.
Click on the Configuration Manager to show the new configuration.
1-4
Go to the Flow Simulation Analysis Tree and open all the icons.
Boundary Conditions
A boundary condition is required anywhere fluid enters or exits the system and can be
set as a Pressure, Mass Flow, Volume Flow or Velocity.
1 In the Flow Simulation Analysis Tree,
right-click the Boundary Conditions icon
and select Insert Boundary Condition.
With the definition just made, we told Flow Simulation that at this opening 0.5
kilogram of water per second is flowing into the valve. Within this dialog box we can
also specify a swirl to the flow, a non-uniform profile and time dependent properties to
the flow. The mass flow at the outlet does not need to be specified due to the
conservation of mass; mass flow in equals mass flow out. Therefore another different
condition must be specified. An outlet pressure should be used to identify this
condition.
6 Select the inner face of the Lid <2> part as
shown. (To access the inner face, right-click
the Lid <2> in the graphics area and choose
Select Other , hover the pointer over items
in the list of items until the inner face is
highlighted, then click the left mouse button).
7 In the Flow Simulation Analysis Tree, right-
click the Boundary Conditions icon and
select Insert Boundary Condition.
1-6
8 Select Pressure Openings and Static Pressure.
With the definition just made, we told Flow Simulation that at this opening the fluid
exits the model to an area of static atmospheric pressure. Within this dialog box we can
also set time dependent properties to the pressure.
Engineering goals are the parameters which the user is interested in. Setting goals is in
essence a way of conveying to Flow Simulation what you are trying to get out of the
analysis, as well as a way to reduce the time Flow Simulation needs to reach a
solution. By setting a variable as a project goal you give Flow Simulation information
about variables that are important to converge upon (the variables selected as goals)
and variables that can be less accurate (the variables not selected as goals) in the
interest of time. Goals can be set throughout the entire domain (Global Goals), within
a selected volume (Volume Goals), in a selected surface area (Surface Goals), or at
given point (Point Goals). Furthermore, Flow Simulation can consider the average
value, the minimum value or the maximum value for goal settings. You can also define
1-8
an Equation Goal that is a goal defined by an equation involving basic mathematical
functions with existing goals as variables. The equation goal allows you to calculate
the parameter of interest (i.e., pressure drop) and keeps this information in the project
for later reference.
Click File, Save.
Solution
1 After the calculation has started and several first iterations has passed (keep your eye
on the Iterations line in the Info window), click the Suspend button on the
Solver toolbar.
We employ the Suspend option only due to extreme simplicity of the current example,
which otherwise could be calculated too fast, leaving you not enough time to perform
the subsequent steps of result monitoring. Normally you may use the monitoring tools
without suspending the calculation.
2 Click Insert Goal Plot on the Solver toolbar. The Add/Remove Goals dialog
box appears.
3 Select the SG Average Static Pressure 1 in the
Select goals list and click OK.
1-10
5 This is the Preview Settings dialog box.
Selecting any SolidWorks plane from the
Plane name list and pressing OK will
create a preview plot of the solution in
that plane. For this model Plane2 is a
good choice to use as the preview plane.
Cut Plots
3 Click OK .
1-12
5 Clear Contours and select Vectors .
6 Click OK .
Surface Plots
Isosurface Plots
1-14
You should see something similar to this image.
For this plot we selected the outlet lid (any flat face or
sketch can be selected) and therefore every trajectory crosses that selected face. The
trajectories can also be colored by values of whatever variable chosen in the View
Settings dialog box. Notice the trajectories that are entering and exiting through the
1-16
exit lid. This is the reason for the warning we received during the calculation. Flow
Simulation warns us of inappropriate analysis conditions so that we do not need to be
CFD experts. When flow both enters and exits the same opening, the accuracy of the
results will worsen. In a case like this, one would typically add the next component to
the model (say, a pipe extending the computational domain) so that the vortex does not
occur at opening.
XY Plots
4
charts is shown below. You
will need to toggle between 3
view any result along 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
-1
sketched lines. The data is Curve Le ngth (m)
1-18
Surface Parameters
Surface Parameters is a feature used to determine pressures, forces, heat fluxes as well as
many other variables on any face within your model contacting the fluid. For this type of
analysis, a calculation of the average static pressure drop from the valve inlet to outlet
would probably be of some interest.
1 Right-click the Surface Parameters icon and select
Insert.
This section is intended to show you how easy it is to analyze design variations. The
variations can be different geometric dimensions, new features, new parts in an
assembly – whatever! This is the heart of Flow Simulation and this allows design
engineers to quickly and easily determine which designs have promise, and which
designs are unlikely to be successful. For this example, we will see how filleting two
sharp edges will influence the pressure drop through the valve. If there is no
improvement, it will not be worth the extra manufacturing costs.
Create a new configuration using the SolidWorks Configuration Manager Tree.
3 Click OK .
1-20
1 Right-click the root item in the SolidWorks
Configuration Manager and select Add
Configuration.
3 Click OK .
4 Add a 1,5 mm
fillet to the
shown face.
1-22
Clone the Project
Now the Flow Simulation project we have chosen is added to the SolidWorks project
which contains the geometry that has been changed. All our input data are copied, so
we do not need to define our openings or goals again. The Boundary Conditions can be
changed, deleted or added. All changes to the geometry will only be applied to this new
configuration, so the old results are still saved.
Please follow the previously described steps for solving and for viewing the results.
In the previous sections we examined how you could compare results from different
geometries. You may also want to run the same geometry over a range of flow rates.
This section shows how quick and easy it can be to do that kind of parametric study.
Here we are going to change the mass flow to 0.75 kg/s.
Activate the Project 1 configuration.
1 Create a copy of the Project 1 project by clicking
Flow Simulation, Project, Clone Project.
2 Type Project 3 for the new project name and click
OK.
Flow Simulation now creates a new configuration. All our input data are copied, so we do
not need to define our openings or goals again. The Boundary Conditions can be changed,
deleted or added. All changes to the geometry will only be applied to this new
configuration, so the old results remain valid. After changing the inlet flow rate value to
0.75 kg/s you would be ready to run again. Please follow the previously described steps
for solving and for viewing the results.
Imagine being the designer of this ball valve. How would you make decisions concerning
your design? If you had to determine whether the benefit of modifying the design as we
have just done outweighted the extra costs, how would you do this? Engineers have to
make decisions such as this every day, and Flow Simulation is a tool to help them make
those decisions. Every engineer who is required to make design decisions involving fluid
and heat transfer should use Flow Simulation to test their ideas, allowing for fewer
prototypes and quicker design cycles.
1-24
2
First Steps - Conjugate Heat Transfer
This First Steps - Conjugate Heat Transfer tutorial covers the basic steps to set up a flow
analysis problem including heat conduction in solids. This example is particularly
pertinent to users interested in analyzing flow and heat conduction within electronics
packages although the basic principles are applicable to all thermal problems. It is
assumed that you have already completed the First Steps - Ball Valve Design tutorial
since it teaches the basic principles of using Flow Simulation in greater detail.
1 Copy the First Steps - Electronics Cooling folder into your working directory and
ensure that the files are not read-only since Flow Simulation will save input data to
these files. Click File, Open.
2 In the Open dialog box, browse to the Enclosure Assembly.SLDASM assembly
located in the First Steps - Electronics Cooling folder and click Open (or
double-click the assembly). Alternatively, you can drag and drop the
Enclosure Assembly.SLDASM file to an empty area of SolidWorks window.
Heat Sink
Inlet Fan
PCBs
Small Chips
Main Chip
Capacitors
Mother Board
Power Supply
In a typical assembly there may be many features, parts or sub-assemblies that are not
necessary for the analysis. Prior to creating an Flow Simulation project, it is a good
practice to check the model to find components that can be removed from the analysis.
Excluding these components reduces the computer resources and calculation time
required for the analysis.
The assembly consists of the following components: enclosure, motherboard and two
smaller PCBs, capacitors, power supply, heat sink, chips, fan, screws, fan housing, and
lids. You can highlight these components by clicking on the them in the FeatureManager
design tree. In this tutorial we will simulate the fan by specifying a Fan boundary
condition on the inner face of the inlet lid. The fan has a very complex geometry that may
cause delays while rebuilding the model. Since it is outside the enclosure, we can exclude
it by suppressing it.
1 In the FeatureManager design tree, select
the Fan-412, and all Screws components
(to select more than one component, hold
down the Ctrl key while you select).
2 Right-click any of the selected
components and select Suppress ..
2-2
Suppressing fan and its screws leaves open five holes in the enclosure. We are going to
perform an internal analysis, so all holes must be closed with lids. It can be done with
the lid creation tool avalilable under Flow Simulation, Tools, Create Lids.
To save your time, we created the lids and included them to the model. You just need to
unsupress them..
3 In the FeatureManager design tree,
select the Inlet Lid, Outlet Lid and
Screwhole Lid components and
patterns DerivedLPattern1 and
LocalLPattern1 (these patterns contain
cloned copies of the outlet and
screwhole lids).
4 Right-click any of the selected
components and select
Unsuppress .
Click Next.
Now we will create a new system of units
named USA Electronics that is better
suited for our analysis.
3 In the Unit system list select the USA
system of units. Select Create new to
add a new system of units to the
Engineering Database and name it
USA Electronics.
custom unit system. Both pre-defined and custom unit systems are stored in the
Engineering Database. You can create the desired system of units in the Engineering
Database or in the Wizard.
By scrolling through the different groups in the Parameter tree you can see the units
selected for all the parameters. Although most of the parameters have convenient units
such as ft/s for velocity and CFM (cubic feet per minute) for volume flow rate we will
change a couple units that are more convenient for this model. Since the physical size
of the model is relatively small it is more convenient to choose inches instead of feet as
the length unit.
4 For the Length entry, double-click its cell
in the Units column and select Inch.
Click Next .
6 Set the analysis type to Internal. Under
Physical Features select the Heat
conduction in solids check box.
2-4
7 Expand the Gases folder and double-click
Air row. Keep default Flow
Characteristics.
Click Next.
to set the initial temperature close to the anticipated final solution to speed up
convergence. In this case we will set the initial air temperature and the initial
temperature of the stainless steel (which represents the material of enclosure) to 50°F
because the box is located in an air-conditioned room.
10 Set the initial fluid Temperature and the
Initial solid temperature to 50°F.
Click Next .
We will use the Flow Simulation Analysis tree to define our analysis, just as the
FeatureManager design tree is used to design your models.
Right-click the Computational Domain icon and select
Hide to hide the wireframe box.
A Fan is a type of flow boundary condition. You can specify Fans at selected solid
surfaces, free of Boundary Conditions and Sources. At model openings closed by lids
you can specify Inlet or Outlet Fans. You can also specify fans on any faces within the
2-6
flow region as Internal Fans. A Fan is considered as an ideal device creating a flow
with a certain volume (or mass) flow rate, which depends on the difference between the
inlet and outlet pressures on the selected faces.
If you analyze a model with a fan, you sholud know the fan's characteristics. In this
example we use one of the pre-defined fans available in the Engineering Database. If you
cannot find an appropriate fan in the Engineering Database, you can create your own fan in
accordance with the fan’s specification.
1 Click Flow Simulation, Insert, Fan . The Fan dialog box appears.
2 Select the inner face of the Inlet
Lid part as shown. (To access
the inner face, right-click the
Inlet Lid in the graphics area
and choose Select Other, hover
the pointer over items in the list
of features until the inner face is
highlighted, then click the left
mouse button).
3 Select External Inlet Fan as fan
Type.
Face coordinate system is created automatically in the center of a planar face when
you select this face as the face to apply the boundary condition or fan. The X axis of
this coordinate system is normal to the face. The Face coordinate system is created
when only one planar face is selected.
7 Accept X as the Reference axis.
Now you can edit the External Inlet Fan1 item or add a new fan
using Flow Simulation Analysis tree. This folder remains visible
until the last feature of this type is deleted. You can also make a
feature’s folder to be initially available in the tree. Right-click the
project name item and select Customize Tree to add or remove
folders.
Since the outlet lids of the enclosure are at ambient atmospheric pressure, the pressure
rise produced by the fan is equal to the pressure drop through the electronics
enclosure.
A boundary condition is required at any place where fluid enters or exits the model,
excluding openings where a fan is specified. A boundary condition can be set in form of
Pressure, Mass Flow, Volume Flow or Velocity. You can also use the Boundary
Condition dialog for specifying an Ideal Wall condition that is an adiabatic, frictionless
wall or a Real Wall condition to set the wall roughness and/or temperature and/or heat
conduction coefficient at the model surfaces. For internal analyses with Heat conduction
in solids enabled, you can also set thermal wall condition on outer model walls by
specifying an Outer Wall condition.
1 In the Flow Simulation analysis tree, right-click the
Boundary Conditions icon and select Insert
Boundary Condition.
2-8
2 Select the inner faces of all outlet
lids as shown.
5 Click OK .
6 In the Flow Simulation Analysis tree click-pause-click the
new VS Heat Generation Rate 1 item and rename it to
Main Chip.
Volume Heat Sources allow you to specify the heat generation rate (in Watts) or the
volumetric heat generation rate (in Watts per volume) or a constant temperature
boundary condition for the volume. It is also possible to specify Surface Heat Sources
in terms of heat transfer rate (in Watts) or heat flux (in Watts per area).
Click anywhere in the graphic area to clear the selection.
1 In the Flow Simulation analysis tree, right-click the Heat Sources icon and select
Insert Volume Source.
2 In the flyout FeatureManager
design tree select all Capacitor
components.
4 Click OK .
5 Click-pause-click the new VS
Temperature 1 item and rename
it to Capacitors.
2-10
6 Following the same procedure as
above, set the following volume
heat sources: all chips on PCBs
(Small Chip) with the total heat
generation rate of 4 W,
Power Supply with
the temperature of 120 °F.
The real PCBs are made of laminate materials consisting of several layers of thin metal
conductor interleaved with layers of epoxy resin dielectric. As for most laminate
materials, the properties of a typical PCB material can vary greatly depending on the
direction - along or across the layers, i.e. they are anisotropic. The Engineering Database
contains some predefined PCB materials with anisotropic thermal conductivity.
In this tutorial example anisotropic thermal conductivity of PCBs does not affect the
overall cooling performance much, so we will create a PCB material having the same
thermal conductivity in all directions to learn how to add a new material to Engineering
Database and assign it to a part.
1 Click Flow Simulation, Tools, Engineering Database .
5 Switch to the Items tab and click New Item on the toolbar.
6 Specify the properties
of the chips material:
Name = Tutorial
component package,
Comments =
Component package,
Density = 2000 kg/m^3,
Specific heat = 120 J/(kg*K),
Conductivity type = Isotropic
Thermal conductivity = 0.4 W/(m*K),
Melting temperature = 1688.2 K.
7 Click Save .
8 Click File, Exit to exit the database.
2-12
You can enter the material properties in any unit system you want by typing the unit
name after the value and Flow Simulation will automatically convert the entered value
to the SI system of units. You can also specify temperature dependent material
properties using the Tables and Curves tab.
Solid Materials are used to specify the materials for solid parts in the assembly.
1 Right-click the Solid Materials icon and select Insert Solid Material.
2 In the flyout FeatureManager
design tree select
MotherBoard, PCB<1> and
PCB<2> components.
3 In the Solid list expand User
Defined and select Tutorial
PCB.
4 Click OK .
5 Following the same procedure, specify solid materials for other components:
• for the main chip and all small chips assign the new Tutorial component package
material (available under User Defined);
• the heat sink is made of Aluminum (available under Pre-Defined, Metals);
• the lids (Inlet Lid, Outlet Lid, Screwhole Lid and all lids in both the
DerivedLPattern1 and LocalLPattern1 patterns) are made of the Insulator
material (available under Pre-Defined , Glasses and Minerals).
To select a part, click it in the FeatureManager design tree or SolidWorks graphics
area.
6 Change the name of each assigned solid material. The new,
descriptive names should be:
PCB - Tutorial PCB,
Chips - Tutorial component package,
2-14
6 Change the name of the new item to
VG Small Chips Max Temperature. You
can also change the name of the item using the
Feature Properties dialog that appears if you
right-click the item and select Properties .
10 Click OK .
11 Rename the new VG Max
Temperature of Solid 1 item to
VG Chip Max Temperature.
2-16
7 Right-click the Goals icon and select Insert
Surface Goals.
8 Click the Flow Simulation Analysis Tree tab
and click the Environment Pressure 1 item to
select the faces where the goal is going to be
applied.
9 In the Parameter table select the first check
box in the Mass Flow Rate row.
10 Accept selected Use for Conv. (Use for
Convergence Control) check box to use this
goal for convergence control.
In this tutorial the engineering goals are set to determine the maximum temperature of the
heat generating components, the temperature rise in air and the pressure drop and mass
flow rate through the enclosure.
Click File, Save.
Next let us check the automatically defined geometry resolution settings for this project.
2-18
Changing the Geometry Resolution
Solution
2-20
An Excel spreadsheet with the goal results will open. The first sheet will show a table
summarizing the goals.
Enclosure Assem bly.SLDASM [Inlet Fan (original)]
Goal Name Unit Value Averaged Value Minimum Value Maximum Value Progress [%] Use In Convergence
GG Av Static Pressure [lbf/in^2] 14.69678696 14.69678549 14.69678314 14.69678772 100 Yes
SG Inlet Av Static Pressure [lbf/in^2] 14.69641185 14.69641047 14.69640709 14.69641418 100 Yes
GG Av Temperature of Fluid [°F] 61.7814683 61.76016724 61.5252449 61.86764155 100 Yes
SG Outlet Mass Flow Rate [lb/s] -0.007306292 -0.007306111 -0.007306913 -0.007303663 100 Yes
VG Small Chips Max Temp [°F] 91.5523903 90.97688632 90.09851988 91.5523903 100 Yes
VG Chip Max Temperature [°F] 88.51909612 88.43365626 88.29145322 88.57515562 100 Yes
You can see that the maximum temperature in the main chip is about 88 °F, and the
maximum temperature over the small chips is about 91 °F.
To analyze the results in more detail let us use the various Flow Simulation
post-processing tools. The best method for the visualization of how the fluid flows inside
the enclosure is to create flow trajectories.
Flow Trajectories
2-22
This setting defines how trajectories are
colored. If Use from contours is selected
then the trajectories are colored with the
distribution of the parameter specified on
the Contours tab (Velocity in our case). If
you select Use fixed color then all flow
trajectories have the same color that you
specify on the Settings tab of the Flow
Trajectories dialog box.
9 Click OK to save the changes and exit the View Settings dialog box.
10 In the Flow Trajectories dialog click OK . The new Flow Trajectories 1 item
appears in the Flow Simulation Analysis tree.
This is the picture you should see.
Cut Plots
2-24
Let us now look at the fluid temperature.
9 Double-click the palette bar in the upper left corner of the graphics area. The
View Settings dialog appears.
10 Change the Parameter from Velocity to Fluid Temperature.
By specifying the custom Min and Max values you can control the vector length. The
vectors whose velocity exceeds the specified Max value will have the same length as
the vectors whose velocity is equal to Max. Likewise, the vectors whose velocity is less
than the specified Min value will have the same length as the vectors whose velocity is
equal to Min. We have set 1 ft/s to display areas of low velocity.
14 Click OK .
15 Right-click the Cut Plot 1 item and select Edit
Definition.
16 Click Vectors
18 Expand the Vectors group box. Using the slider set the
Vector Spacing to ~ 0.18 in.
19 Click OK .
2-26
Right-click the Cut Plot 1 item and select Hide. Let us now display solid temperature.
Surface Plots
7 In the Surface Plot dialog box click OK . The creation of the surface plot may
take some time because many faces need to be colored.
8 Repeat steps 1 and 2 and select the Power Supply and all Capacitors components,
then click OK .
2-28
You can view and analyze the results further with the post-processing tools that were
shown in the First Steps - Ball Valve Design tutorial. Flow Simulation allows you to
quickly and easily investigate your design both quantitatively and qualitatively.
Quantitative results such as the maximum temperature in the component, pressure drop
through the cabinet, and air temperature rise will allow you to determine whether the
design is acceptable or not. By viewing qualitative results such as air flow patterns, and
heat conduction patterns in the solid, Flow Simulation gives you the necessary insight to
locate problem areas or weaknesses in your design and provides guidance on how to
improve or optimize the design.
2-30
3
First Steps - Porous Media
In this tutorial we consider flow in a section of an automobile exhaust pipe, whose exhaust
flow is resisted by two porous bodies serving as catalysts for transforming harmful carbon
monoxide into carbon dioxide. When designing an automobile catalytic converter, the
engineer faces a compromise between minimizing the catalyst's resistance to the exhaust
flow while maximizing the catalyst's internal surface area and duration that the exhaust
gases are in contact with that surface area. Therefore, a more uniform distribution of the
exhaust mass flow rate over the catalyst's cross sections favors its serviceability. The
porous media capabilities of Flow Simulation are used to simulate each catalyst, which
allows you to model the volume that the catalyst occupies as a distributed resistance
instead of discretely modeling all of the individual passages within the catalyst, which
would be impractical or even impossible. Here, as a Flow Simulation tutorial example we
consider the influence of the catalysts' porous medium permeability type (isotropic and
unidirectional media of the same resistance to flow) on the exhaust mass flow rate
distribution over the catalysts' cross sections. We will observe the latter through the
behavior of the exhaust gas flow trajectories distributed uniformly over the model's inlet
and passing through the porous catalysts. Additionally, by coloring the flow trajectories
by the flow velocity the exhaust gas residence time in the porous catalysts can be
estimated, which is also important from the catalyst effectiveness viewpoint.
The project Wizard guides you through the definition of the project’s properties
step-by-step. Except for two steps (steps to define the project fluids and default solid),
each step has some pre-defined values, so you can either accept these values (skipping
the step by clicking Next) or modify them to your needs.
These pre-defined settings are:
unit system – SI,
analysis type – internal, no additional physical capabilities are considered,
wall condition – adiabatic wall
initial conditions – pressure - 1 atm, temperature - 293.2 K.
result and geometry resolution – level 3,
For this project these default settings suit perfectly and all what we need to do is just to
select Air as the project fluid. To avoid passing through all steps we will use Navigator
pane that provides a quick access to the Wizard’s pages.
3-2
3 In the Navigator pane click
Fluids.
5 Since we do not need to change other properties we can close the Wizard.
Click Finish in the Navigator pane.
You can click Finish at any moment, but if you attempt to close Wizard without
specifying all obligatory properties (such as project fluds), the Wizard will not close
and the page where you need to define a missing property will be marked by the
exclamation icon .
Now Flow Simulation creates a new configuration with the Flow Simulation data
attached.
In the Flow Simulation Analysis tree, right-click the Computational Domain icon and
select Hide to hide the black wireframe box.
5 Click OK .
3-4
6 Select the inner face of the outlet lid as shown.
7 Right-click the Boundary Conditions icon and
select Insert Boundary Condition .
9 Click OK .
The material you are going to create is already defined in the Engineering Database under
the Pre-Defined folder. You can skip the definition of porous material and select the
pre-defined "Isotropic" material from the Engineering database when you will assign the
porous material to a component later in this tutorial.
1 Click Flow Simulation, Tools, Engineering Database .
2 In the Database tree select Porous Media,
User Defined.
3 Click New Item on the toolbar. The blank Item Properties tab
appears. Double-click the empty cells to set the corresponding
property values.
4 Name the new porous medium Isotropic.
5 Under Comment, click the button and type the desired comments for this porous
medium. The Comment property is optional, you can leave this field blank.
Porosity is the effective porosity of the porous medium, defined as the volume fraction
of the interconnected pores with respect to the total porous medium volume; here, the
porosity is equal to 0.5. The porosity will govern the exhaust flow velocity in the porous
medium channels, which, in turn, governs the exhaust gas residence in the porous
catalyst and, therefore, the catalyst efficiency.
7 Choose Isotropic as the Permeability type.
First of all let us consider an Isotropic permeability, i.e, a medium with permeability
not depending on the direction within the medium. Then, as an alternative, we will
consider a Unidirectional permeability, i.e., the medium permeable in one direction
only.
8 Choose Pressure drop, Flowrate, Dimensions as the Resistance calculation
formula.
For our media we select the Pressure Drop, Flowrate, Dimensions medium
resistance to flow, i.e., specify the porous medium resistance as k = ∆P×S /(m×L) (in
units of s-1), where the right-side parameters are referred to a tested parallelepiped
sample of the porous medium, having the S cross-sectional area and the L length in the
selected sample direction, in which the mass flow rate through the sample is equal to m
under the pressure difference of ∆P between the sample opposite sides in this
direction.
In this project we will specify ∆P = 20 Pa at m = 0.01 kg/s (and ∆P = 0 Pa at
m=0 kg/s), S = 0.01 m2, L = 0.1m. Therefore, k = 200 s-1.
Knowing S and L of the catalyst inserted into the model and m of the flow through it,
you can approximately estimate the pressure loss at the model catalyst from
∆P = k×m×L/S.
9 For the Pressure drop vs.
flowrate choose Mass Flow
Rate. Click the button to
switch to the Tables and
Curves tab.
3-6
10 In the Property table specify the
linear dependency of pressure drop
vs. mass flow rate as shown.
11 Go back to the Item Properties tab.
12 Set Length to 0.1 m and Area to
0.01 m2.
13 Click Save .
14 Click File, Exit to exit the database.
3-8
2 Click the Flow Simulation Analysis Tree tab
and click the Inlet Velocity 1 item to select the
inner face of the inlet lid.
3 In the Parameter table select the Av check box
in the Total Pressure row.
4 Accept the selected Use for Conv. (Use for
Convergence Control) check box to use this
goal for convergence control.
5 Under Name Template, located at the bottom
of the PropertyManager, click Inlet .
Equation Goal is a goal defined by an analytical function of the existing goals and/or
parameters of input data conditions. This goal can be viewed as equation goal during
the calculation and while displaying results in the same way as the other goals. As
variables, you can use any of the specified goals, including another equation goals,
except for goals that are dependent on other equation goals, and parameters of the
specified project’s input data features (general initial or ambient conditions, boundary
conditions, fans, heat sources, local initial conditions). You can also use constants in
the definition of the equation goal.
3-10
1 Right-click the Goals icon and select Insert Equation
Goal.
You can use goals (including previously specified Equation Goals), parameters of
input data conditions and constants in the expression defining an Equation Goal. If the
constants in the expression represent some physical parameters (i.e. length, area, etc.),
make sure that they are specified in the project’s system of units. Flow Simulation has
no information about the physical meaning of the constants you use, so you need to
specify the Equation Goal dimensionality by yourself.
5 Keep the default Pressure & Stress in the Dimensionality list.
6 Click OK. The new Equation Goal 1
item appears in the tree.
Solution
You can see that the total pressure drop is about 120 Pa.
Catalyst.SLDASM [Isotropic]
Goa l Na me Unit Va lue Ave ra ged Va lue Minimum Va lue Ma x imum Va lue P rogre ss [%] Use In Conve rgence
E quation Goal 1 [Pa] 120.0326909 121.774802 120.0326909 124.432896 100 Yes
To see the non-uniformity of the mass flow rate distribution over a catalyst’s cross section,
we will display flow trajectories with start points distributed uniformly across the inlet.
3-12
Flow Trajectories
To see trajectories inside the porous media we will apply some transparency to the model.
9 Click Flow Simulation, Results, Display, Transparency
and set the model transparency to 0.75.
3-14
2 Enter Unidirectional as the Configuration name.
3 Click OK.
The material you are going to create is already defined in the Engineering Database under
the Pre-Defined folder. You can skip the definition of porous material and select the
pre-defined "Unidirectional" material from the Engineering database when you will assign
the porous material to a component later in this tutorial.
1 Click Flow Simulation, Tools, Engineering Database .
2 In the Database tree select Porous Media, User Defined.
3 On the Items tab select the Isotropic item.
4 Click Copy .
5 Click Paste . The new Copy of Isotropic (1) item appears in the list.
6 Select the Copy of
Isotropic (1) item and click the
Item Properties tab.
7 Rename the item to
Unidirectional.
8 Change the Permeability type
to Unidirectional.
9 Save the database and exit.
4 Click OK .
Since all other conditions and goals remain the same, we can
start the calculation immediately
When the calculation is finished, create the goal plot for the Equation Goal 1.
C atalyst.SLD ASM [U nidirectional]
Goa l Na m e Unit Va lue Ave ra ged Va lue M inim um Va lue Ma x im um Va lue P rogre ss [%] Use In Conve rgence
E quation Goal 1 [Pa] 117.0848512 118.6235708 117.0761518 121.5639633 100 Yes
Comparing the trajectories passing through the isotropic and unidirectional porous
catalysts installed in the tube, we can summarize:
3-16
Due to the asymmetric position of the inlet tube with respect to the larger tube in which
the catalyst bodies are installed, the incoming flow is non-uniform. Since the incoming
flow is non-uniform, the flow inside the first catalyst body is non-uniform also. It is seen
that the catalyst type (isotropic or unidirectional) affects both the incoming flow
non-uniformity (slightly) and, more substantially, the flow within the catalysts (especially
the first catalyst body). In both the cases the gas stream mainly enters the first catalyst
body-closer to the wall opposite to the inlet tube. For the isotropic case, the gas flows into
the first body nearer to the wall than for the case of the unidirectional catalyst. As a result,
the flow in the initial (about one-third of the body length) portion of the first catalyst body
is noticeably more non-uniform in the isotropic catalyst. Nevertheless, due to the isotropic
permeability, the main gas stream expands in the isotropic catalyst and occupies a larger
volume in the next part of the body than in the unidirectional catalyst, which, due to its
unidirectional permeability, prevents the stream from expanding. So, the flow in the last
two-thirds of the first catalyst body is less non-uniform in the isotropic catalyst. Since the
distance between the two porous bodies installed in the tube is rather small, the gas stream
has no time to become more uniform in the volume between the catalyst bodies, although
in the unidirectional case a certain motion towards uniformity is perceptible. As a result,
the flow non-uniformity occurring at the first catalyst body's exit passes to the second
catalyst body. Then, it is seen that the flow non-uniformity does not change within the
second catalyst body.
Let us now consider the flow velocity inside the catalyst. This is easy to do since the flow
trajectories' colors indicate the flow velocity value in accordance with the specified
palette. To create the same conditions for comparing the flow velocities in the isotropic
and unidirectional catalysts, we have to specify the same velocity range for the palette in
both the cases, since the maximum flow velocity governing the value range for the palette
by default is somewhat different in these cases. It is seen that, considering the catalyst on
the whole, the flow velocities in the isotropic and unidirectional catalysts are practically
the same. Therefore, from the viewpoint of gas residence in the catalyst, there is no
difference between the isotropic and unidirectional catalysts.
We can conclude that the isotropic catalyst is more effective than the unidirectional
catalyst (of the same resistance to uniform flows), since the flow in it, as a whole, is more
uniform. In spite of specifying the same resistance of the catalysts to flow, the overall
pressure loss is lower by about 2% in the case of employing the unidirectional catalyst.
This difference is due to the different flow non-uniformity both in the catalyst bodies and
out of them.
3-18
4
Determination of Hydraulic Loss
In engineering practice the hydraulic loss of pressure head in any piping system is
traditionally split into two components: the loss due to friction along straight pipe sections
and the local loss due to local pipe features, such as bends, T-pipes, various cocks, valves,
throttles, etc. Being determined, these losses are summed to form the total hydraulic loss.
Generally, there are no problems in engineering practice to determine the friction loss in a
piping system since relatively simple formulae based on theoretical and experimental
investigations exist. The other matter is the local hydraulic loss (or so-called local drag).
Here usually only experimental data are available, which are always restricted due to their
nature, especially taking into account the wide variety of pipe shapes (not only existing,
but also advanced) and devices, as well as the substantially complicated flow patterns in
them. Flow Simulation presents an alternative approach to the traditional problems
associated with determining this kind of local drag, allowing you to predict
computationally almost any local drag in a piping system within good accuracy.
Click File, Open. In the Open dialog box, browse to the Valve.SLDPRT model located in
the Tutorial 1 - Hydraulic Loss folder and click Open (or double-click the part).
Alternatively, you can drag and drop the Valve.SLDPRT file to an empty area of the
SolidWorks window.
Model Description
D ball
θ = arc sin 2 -------------
D pi pe
The standard engineering convention for determining local drag is by calculating the
difference between the fluid dynamic heads measured upstream of the local pipe feature
(ball valve in our case) and far downstream of it, where the flow has become uniform
(undisturbed) again. In order to extract the pure local drag the hydraulic friction loss in the
straight pipe of the same length must be subtracted from the measured dynamic head loss.
In this example we will obtain pressure loss (local drag) in the ball valve whose handle is
turned by an angle of 40o. The Valve analysis represents a typical Flow Simulation internal
analysis.
Internal flow analyses deal with flows inside pipes, tanks, HVAC systems, etc. The fluid
enters a model at the inlets and exits the model through outlets.
To perform an internal analysis all the model openings must be closed with lids, which are
needed to specify inlet and outlet flow boundary conditions on them. In any case, the
internal model space filled with a fluid must be fully closed. You simply create lids as
additional extrusions covering the openings. In this example the lids are semi-transparent
allowing a view into the valve.
4-2
To ensure the model is fully closed click Flow Simulation, Tools,
Check Geometry. Then click Check to calculate the fluid and
solid volumes of the model. If the fluid volume is equal to zero, the
model is not closed.
Click Fluid Volume to see the volume that will be occupied by
fluid in the analysis.
Uncheck Fluid Volume. Close the Check Geometry dialog box.
The Check Geometry tool allows you to calculate the total fluid
and solid volumes, check bodies for possible geometry
problems (i.e. invalid contact) and visualize the fluid area and
solid body as separate models.
The first step is to create a new Flow Simulation project.
Creating a Project
1 Click Flow Simulation, Project, Wizard . The project wizard guides you through the
definition of a new Flow Simulation project.
2 In the Project Configuration dialog box,
click Use current . Each Flow Simulation
project is associated with a SolidWorks
configuration. You can attach the project
either to the current SolidWorks
configuration or create a new SolidWorks
configuration based on the current one.
Click Next.
4-4
6 Since we do not intend to calculate heat
conduction in solids, in the Wall
Conditions dialog box you can specify
the thermal wall boundary conditions
applied by default to all the model walls
contacting with the fluid.
For this project accept the default
Adiabatic wall feature denoting that all
the model walls are heat-insulated.
In this project we will not consider rough
walls.
Click Next.
7 In the Initial Conditions dialog box specify
initial values of the flow parameters. For
steady internal problems, the specification of
these values closer to the expected flow field
will reduce the analysis convergence time.
Result Resolution governs the solution accuracy via mesh settings and conditions of
finishing the calculation that can be interpreted as resolution of calculation results.
The higher the Result Resolution, the finer the mesh and the stricter the convergence
criteria. Naturally, higher Result Resolution requires more computer resources (CPU
time and memory).
Geometry Resolution (specified through the minimum gap size and the minimum wall
thickness) governs proper resolution of geometrical model features by the
computational mesh. Naturally, finer Geometry Resolution requires more computer
resources.
Select the Manual specification of the minimum
gap size check box and enter 0.04 m for the
Minimum gap size.
The next step is specifying Boundary Conditions. Boundary Conditions are used to
specify fluid characteristics at the model inlets and outlets in an internal flow analysis or
on model surfaces in an external flow analysis.
4-6
Specifying Boundary Conditions
specifying boundary conditions, we recommend that you specify at least one Pressure
opening condition since the mass flow rate on a Pressure opening is automatically
calculated to satisfy the law of conservation of mass.
10 Click OK .
By specifying this condition we define that at the ball valve pipe exit the water has a static
pressure of 1 atm.
The hydraulic losses are calculated through the outlet and inlet total pressure difference
∆P from the following formula:
∆P
ξ = ----------------
2
-
ρV ⁄ 2
where ρ is the water density, and V is water velocity. Since we already know the water
velocity (specified by us as 1 m/s) and the water density (998.1934 kg/m3 for the specified
temperature of 293.2 K), then our goal is to determine the total pressure value at the
valve’s inlet and outlet. The easiest and fastest way to find the parameter of interest is to
specify the corresponding engineering goal.
4-8
3 Select Create a separate goal for each surface
check box to create two separate goals, i.e. one
for each of the selected faces.
4 In the Parameter table select the Av check box in
the Total Pressure row.
5 Accept selected Use for Conv. check box to use
the goals being created for convergence control.
1 Click Flow Simulation, Solve, Run. The Run dialog box appears.
2 Click Run to start the calculation.
You can easily extend the ball valve inlet and outlet sections by changing the offset
distance for the Inlet Plane and Outlet Plane features. Instead, we shall clone the project
to the pre-defined 40 degrees - long valve configuration.
4-10
In the Flow Simulation analysis tree, right-click the root
40 degrees - long valve item and select Run. Then click
Run to start the calculation.
The Cut Plot displays results of a selected parameter in a selected view section. To
define the view section, you can use SolidWorks planes or model planar faces (with the
additional shift if necessary). The parameter values can be represented as a contour
plot, as isolines, as vectors, or in a combination (e.g. contours with overlaid vectors).
2 In the flyout FeatureManager design
tree, expand the Valve item and select
Plane2. Its name appears in the Section
Plane or Planar Face list.
3 In the Cut Plot dialog box, in addition to
Vectors .
5 Click View Settings to specify the parameter which values to show at the contour plot.
The settings made in the View Settings dialog box refer to all Cut Plots, Surface
Plots, Isosurfaces, and Flow Trajectories features. These settings are applied only for
the active pane of the SolidWorks graphics area. For example, the contours in all cut
and surface plots will show the same physical parameter selected in the View Settings
dialog box. So, in the View Settings dialog box you specify the displayed physical
parameter and the settings required for displaying it for each of the displaying options
(contours, isolines, vectors, flow trajectories, isosurfaces) . The Contours settings can
also be applied to Isolines, Vectors, Flow Trajectories, and Isosurfaces. If the Use
from contours option is selected on the corresponding feature tab, the isolines,
vectors, trajectories, isosurfaces are colored in accordance with values of the
parameter selected on the Contours tab (in this case the color settings made in the
specific dialog boxes are not used).
6 On the Contours tab, in the Parameter box,
select X-velocity.
7 Click OK to save changes and exit the View
Settings dialog box.
click OK .
4-12
Now you can see a contour plot of the velocity and the velocity vectors projected on the plot.
For better visualization of the vortex you can scale small vectors:
11 In the Flow Simulation Analysis tree, right-click
the Results icon and select View Settings.
By specifying the custom Min we change the vector length range so that the vectors in
areas where velocity is less than the specified Min value will appear as if it is equal to
Min. This allows us to visualize the low velocity area in more detail.
14 Click OK to save the changes and exit the View Settings dialog box. Immediately the
cut plot is updated.
You can easily visualize the vortex by displaying the flow relative to the X axis. For that,
you can display the X-velocity component in a two-color palette and set the value,
separating two colors, at zero.
15 In the Flow Simulation Analysis tree, right-click the Results icon and select View
Settings.
16 Using the slider set Number of colors to 2.
17 In the Min box type -1.
18 In the Max box type 1.
19 Click OK .
Next, we will display the distribution of total pressure within the valve.
By default the total pressure is not included in the list of parameters available to display.
To enable or disable a physical parameter for displaying you can use Parameter List.
1 In the Analysis tree, right-click the Results
icon and select Parameter List. Select the
Total Pressure check box or select
parameter’s row and click Enable.
2 Click OK to close the Display Parameters
dialog box.
4-14
1 Double-click the palette bar in the upper left corner of the graphics area to
open the View Settings dialog box.
The cut plot shows you the flow pattern. To obtain the exact value of the total pressure
which is required to calculate the loss, we will use the surface goal plot.
In fact, to obtain the pressure loss it would be easier to specify an Equation goal with the
difference between the inlet and outlet pressures as the equation goal’s expression.
However, to demonstrate the wide capabilities of Flow Simulation, we will calculate the
pressure loss with the Flow Simulation gasdynamic Calculator.
The Calculator contains various formulae from fluid dynamics which can be useful for
engineering calculations. The calculator is a very useful tool for rough estimations of
the expected results, as well as for calculations of important characteristic and
reference values. All calculations in the Calculator are performed only in the
International system of units SI, so no parameter units should be entered, and Flow
Simulation Units settings do not apply in the Calculator.
4-16
4 Click OK. The total pressure loss formula
appears in the Calculator sheet.
In the Result (or A) column you see the formula
name, in the next columns (B, C, etc.) you see
names of the formula arguments (variables and
constants). You can either type all the formula
arguments’ values in cells under their names in
the SI units, or copy and paste them from the
goals Excel worksheet table obtained through the
Goals dialog box. The result value appears in the
Result column cell immediately when you enter
all the arguments and click another cell.
5 Specify the values in the cells as follows:
Density = 998.1934 (the water density for the
specified temperature of 293.2 K),
Velocity = 1.
6 Open the goals1 Excel workbook and copy the Value of SG Av Total Pressure 1 into
the Clipboard.
7 Go to the Calculator, click the B2 cell and press Ctrl+V to paste the goal value from
the Clipboard.
8 Return to Excel, copy the Value of SG Av Total Pressure 2. Go to the Calculator,
click the C2 cell and press Ctrl+V. Click any free cell. Immediately the Total pressure
loss value appears in the Result column.
To obtain the pure local drag, it is necessary to subtract from the obtained value the total
pressure loss due to friction in a straight pipe of the same length and diameter. To do that,
we perform the same calculations in the ball valve model with the handle in the 0o angle
position. You can do this with the 00 degrees - long valve configuration.
Since the specified conditions are the same for both 40 degrees - long valve and 00
degrees - long valve configurations, it is useful to attach the existing Flow Simulation
project to the 00 degrees - long valve configuration.
Clone the current project to the 00 degrees - long valve
configuration.
Since at zero angle the ball valve becomes a simple straight pipe, there is no need to set the
Minimum gap size value smaller than the default gap size which, in our case, is
automatically set equal to the pipe’s diameter (the automatic minimum gap size depends
on the characteristic size of the faces on which the boundary conditions are set). Note that
using a smaller gap size will result in a finer mesh and, in turn, more computer time and
memory will be required for calculation. To solve your task in the most effective way you
should choose the optimal settings for the task.
Check to see that the 00 degrees - long valve is the active configuration.
1 Click Flow Simulation, Initial Mesh .
2 Clear the Manual specification of
the minimum gap size check box.
3 Click OK .
4-18
Now you can calculate the total pressure loss in a straight pipe.
1 Click Flow Simulation, Tools, Calculator.
2 In the Calculator menu, click File, Open. Browse to the folder where you saved the
calculator file earlier in this tutorial and select the ball valve.fwc file. Click Open.
3 Click the B4 cell and in the Calculator toolbar click to paste data from the
Clipboard.
4 Save the existing value of the total pressure loss: click the A2 cell, click , then click
the A7 cell and finally click .
5 Double-click the Name7
cell and type 40 degrees.
9 Click the B5 cell. The value of total pressure is now taken from the
B5 cell. Immediately the total pressure value is recalculated.
Now you can calculate the local drag in the ball valve whose handle is set at 40o.
Total Pressure loss (40 deg) Total Pressure loss (0 deg) Local Drag
4-20
5
Cylinder Drag Coefficient
Flow Simulation can be used to study flow around objects and to determine the resulting
lift and drag forces on the objects due to the flow. In this example we use Flow Simulation
to determine the drag coefficient of a circular cylinder immersed in a uniform fluid
stream. The cylinder axis is oriented perpendicular to the stream.
The computations are performed for a range of Reynolds numbers (1,1000,105), where
ρ UD
Re = ---------------- , D is the cylinder diameter, U is the velocity of the fluid stream, ρ is the
µ
density, and µ is the dynamic viscosity. The drag coefficient for the cylinder is defined as:
FD
C D = -----------------------
1--- 2
ρ U DL
2
where FD is the total force in the flow direction (i.e. drag) acting on a cylinder of diameter
D and length L.
The goal of the simulation is to obtain the drag coefficient predicted by Flow Simulation
and to compare it to the experimental data presented in Ref.1 .
Click File, Open. In the Open dialog box, browse to the Cylinder 0.01m.SLDPRT part
located in the Tutorial 2 - Drag Coefficient\cylinder 0.01m folder and click Open (or
double-click the part). Alternatively, you can drag and drop the cylinder 0.01m.SLDPRT
file to an empty area of SolidWorks window.
The Cylinder analysis represents a typical Flow Simulation External analysis.
External flows analyses deal with flows over or around a model such as flows over
aircrafts, automobiles, buildings, etc. For external flow analyses the far-field
boundaries are the Computational Domain boundaries. You can also solve a combined
external and internal flow problem in a Flow Simulation project (for example flow
around and through a building). If the analysis includes a combination of internal and
external flows, you must specify External type for the analysis.
The first step is to create a new Flow Simulation project.
Creating a Project
1 Click Flow Simulation, Project, Wizard. The project wizard guides you through the
definition of a new Flow Simulation project. In this project we will analyze flow over
the cylinder at the Reynolds number of 1.
2 Select Create new. In the Configuration
name box type Re 1. This is the name of
the SolidWorks configuration that will be
created for the associated Flow
Simulation project.
Click Next .
5-2
3 In the Unit System dialog box you can
select the desired system of units for both
input and output (results).
In this project we will specify the
International System SI by default.
Click Next.
To disregard closed internal spaces within the body you can select Exclude internal
spaces; however no internal spaces exist within the cylinder in this tutorial. The
Reference axis of the global coordinate system (X, Y or Z) is used for specifying data
in a tabular or formula form with respect to a cylindrical coordinate system based on
this axis.
The flow over a cylinder is steady at a Reynolds number Re < 40 (see the cylinder Re
definition above) and unsteady (time-dependent) at Re > 40. Since in this tutorial the
first calculation is performed at Re=1, to accelerate the run, we perform a steady-state
analysis.
Click Next.
5 Since we use water in this project, open
the Liquids folder and double-click the
Water item.
Click Next.
Using Dependency you can specify data in several ways: as a constant, as a tabular or
formula dependency on x, y, z, r, θ, ϕ coordinates and time t (only for time-dependent
analysis). The radius r is the distance from a point to the Reference axis selected from
the reference coordinate system (the Global Coordinate System for all data set in the
Wizard and General Settings dialog boxes), while θ and ϕ are the polar and
azimuthal angles of spherical coordinate system, respectively. Therefore, by
combination of r, θ, and ϕ coordinates you can specify data in cylindrical or spherical
coordinate systems.
9 In the Dependency type list select Formula Definition .
5-4
10 In the Formula box type the formula defining the flow
velocity using the Reynolds number:
1*(0.0010115/0.01/998.19). Here:
1 – the Reynolds number (Re)
0.0010115 (Pa*s) - the water dynamic viscosity (µ) at the
specified temperature of 293.2 K
0.01 (m) - the cylinder diameter (D)
998.19 (kg/m3)- the water density (ρ) at the specified
temperature of 293.2 K
11 Click OK. You will return to the Initial and Ambient
Conditions dialog box.
For most flows it is difficult to have a good estimation of their turbulence a priori, so it
is recommended that the default turbulence parameters be used. The default turbulence
intensity values proposed by Flow Simulation are 0.1% for external analyses and 2%
for internal analyses and these values are appropriate for most cases. In this project we
will specify a turbulence intensity of 1%.
12 Expand the Turbulence parameters item
and in the Turbulence intensity box
type 1.
Click Next.
In this tutorial we are interested in determining the drag coefficient of the cylinder only,
without the accompanying 3D effects. Thus, to reduce the required CPU time and
computer memory, we will perform a two-dimensional (2D) analysis in this tutorial.
1 In the Flow Simulation Analysis tree, expand the Input Data item.
2 Right-click the Computational Domain icon and
select Edit Definition. The Computational Domain
dialog box appears.
3 Click the Boundary Condition tab.
4 In the 2D plane flow list select XY-Plane Flow
(since the Z-axis is the cylinder axis).
Automatically the Symmetry condition is
specified at the Z min and Z max boundaries of
the Computational Domain.
Click the Size tab. You can see that the Z min and
Z max boundaries are set automatically, basing on
the model dimensions.
Thus the reference cylinder length L in the cylinder drag (CD) formula presented above
is equal to L = Z max-Z min = 0.002 m.
For most cases, to study the flow field around an external body and to investigate the
effects of design changes it is recommended to use the default Computational Domain
size as determined by Flow Simulation. However, in this case we will compare the
Flow Simulation results to experimental results and we would like to determine the
drag coefficient with a high degree of accuracy. In order to eliminate any disturbances
of the incoming flow at the Computational Domain boundaries due to the presence of
the cylinder, we will manually set the boundaries farther away from the cylinder. The
accuracy will be increased at the expense of required CPU time and memory due to the
larger size of Computational Domain.
5 Specify the coordinates of the
Computational domain boundaries as
shown on the picture to the right.
6 Click OK .
The X-component of force can be determined easily by specifying the appropriate Flow
Simulation goal. For this case you will specify the X - Component of Force as a Global
Goal. This ensures that the calculation will not be finished until X - Component of Force
in the entire computational domain (i.e. on the cylinder surface) is fully converged.
5-6
Specifying a Global Goal
When the calculation is finished, you will need to manually calculate the drag coefficient
from the obtained force value. Instead, let Flow Simulation to make all the necessary
calculations for you by specifying an Equation Goal.
1 Click Flow Simulation, Insert,
Equation Goal.
2 In the Flow Simulation Analysis tree
select the GG X - Component of
Force 1 goal. It appears in the
Expression box.
4 Select No units in the Dimensionality list and click OK. The new Equation Goal 1 item
appears in the Flow Simulation Analysis tree.
5 Rename the Equation Goal 1 to Drag Coefficient.
To compare the Flow Simulation results with the experimental curve taken from Ref.1, we
will obtain the results at a Reynolds number of 1, 103 and 105. As with Re = 1, the
Cylinder 0.01m.SLDPRT is used to calculate the flow at the Reynolds number of 10 3.
The Cylinder 1m.SLDPRT is used to calculate the flow at the Reynolds number of 105.
Since the new project is a copy of the Re 1 Flow Simulation project, you only need to
change the flow velocity value in accordance with the Reynolds number of 1000. Use
the General Settings dialog box to change the data specified in the Wizard, except the
settings for Units and Result and Geometry Resolution.
The General Settings always presents the current state of the project parameters. You
can change General Settings to correct the settings made in the Wizard or to modify
the project created with the Flow Simulation Template in accordance with the new
project requirements.
5-8
Changing Project Settings
1 Click Flow Simulation, General Settings. The General Settings dialog box appears.
2 As it has been mentioned above, since
the flow over a cylinder is unsteady at
Re > 40, select the Time-dependent
physical feature for this project.
3 In the Navigator click Initial and
ambient conditions.
5 In the Formula box, type the formula for the new Reynolds
number:
1e3*(0.0010115/0.01/998.19).
1 Right-click the Drag Coefficient icon under Goals and select Edit Definition .
2 In the Expression box type the new formula for the new Reynolds number:
{GG X - Component of Force 1}/(0.002*(0.0010115*10^3)^2)*(2*998.19*0.01).
3 Select No units in the Dimensionality list.
4 Click OK to save changes and close the Equation Goal dialog box.
In the experiments performed with one fluid medium, the Reynolds number’s large rise is
usually obtained by increasing both the velocity and the model overall dimension (i.e.
cylinder diameter) since it is difficult to increase only velocity by e.g. 105 times. Since our
simulation is performed with water only, let us increase the cylinder diameter to 1 m to
perform the calculation at a Reynolds number of 105.
Cloning a project is convenient if you want to create similar projects for the same model.
The easiest way to apply the same general project settings to another model is to use the
Flow Simulation Template.
Template contains all of the general project settings that can be used as a basis for a
new project. These settings are: problem type, physical features, fluids, solids, initial
and ambient flow parameters, wall heat condition, geometry and result resolution, and
unit settings. Notice that Boundary Conditions, Fans, Initial Conditions, Goals and
other features accessible from the Flow Simulation, Insert menu, as well as results are
not stored in the template. Initially, only the New Project default template is available,
but you can easily create your own templates.
5-10
Creating a Template
All templates are stored as .fwp files in the <install_dir>/Template folder, so you can
easily apply a template to any previously created models.
4 Save the model.
The newly created project has the same settings as the Re 1000 project with the cylinder
0.01m model. The only exceptions are Geometry Resolution and Computational
Domain size, which are calculated by Flow Simulation in accordance with the new model
geometry.
Notice that the 2D plane flow setting and Global Goal are retained. Next, you can modify
the project in accordance with the new model geometry.
7 Create the Equation Goal for the drag coefficient of the cylinder as it was described before.
In the Expression box enter the formula:
{GG X - Component of Force 1}/(0.2*(0.0010115*10^5)^2)*(2*998.19*1).
8 Select No units in the Dimensionality list.
9 Click OK . Rename the Equation Goal 1 to Drag Coefficient.
Now you can solve all of the projects created for both the cylinders.
5-12
Solving a Set of Projects
Flow Simulation allows you to automatically solve a set of projects that exist in any
currently opened document.
1 Click Flow Simulation, Solve, Batch Run.
2 Select the Solve check
box in the All projects
row to select Solve for
all projects (Re 1,
Re 1000, Re 1e5). Also
select the Close Monitor
check box in the
All projects row. When
the Close Monitor check
box is selected, Flow
Simulation
automatically closes the
Solver Monitor window
when the calculation
finishes.
3 Click Run.
Getting Results
After all calculations are complete, go to the cylinder 0.01m model and activate the Re
1000 configuration. Create Goal Plot to obtain the Drag Coefficient value:
1 Click Flow Simulation, Results, Load\Unload Results.
2 In the Load Results dialog box, keep the default project’s results file (2.fld) and click
Open.
3 In the Flow Simulation Analysis tree, under
Results, right-click the Goals icon and select
Insert. The Goals dialog box appears.
4 Click Add All.
6 Activate the Re 1 configuration and load results. Create the goal plot for both the goals.
cylinder 0.01m .SLDPRT [Re 1]
Goa l Na m e Unit V a lue Ave ra ge d V a lue Minim um V a lue M a x im um V a lue
GG X - Com ponent of Force [N] 1.14448E-09 1.16764E -09 1.12756E-09 1.81674E -09
Drag Coeffic ient [ ] 11.16575499 11.39179479 11.00070462 17.72455528
7 Switch to the cylinder 1m part, activate the Re 1e5 configuration, load results and
create the goal plot for both the goals.
cylinder 1m .SLDPRT [Re 1e5]
Goa l Na m e Unit V a lue Ave ra ge d V a lue Minim um V a lue M a x im um V a lue
GG X - Com ponent of Force [N] 0.482967811 0.478070888 0.465937059 0.491484755
Drag Coeffic ient [] 0.471193865 0.46641632 0.454578294 0.47950318
Even if the calculation is steady, the averaged value is more preferred, since in this case
the oscillation effect is of less perceptibility. We will use the averaged goal value for the
other two cases as well.
You can now compare Flow Simulation results with the experimental curve.
Re
0.1 1 10 100 1000 10000 100000 100000 1E+07
0
Ref. 1 Roland L. Panton, “Incompressible flow” Second edition. John Wiley & sons Inc., 1995
5-14
6
Heat Exchanger Efficiency
Flow Simulation can be used to study the fluid flow and heat transfer for a wide variety of
engineering equipment. In this example we use Flow Simulation to determine the
efficiency of a counterflow heat exchanger and to observe the temperature and flow
patterns inside of it. With Flow Simulation the determination of heat exchanger efficiency
is straightforward and by investigating the flow and temperature patterns, the design
engineer can gain insight into the physical processes involved thus giving guidance for
improvements to the design.
A convenient measure of heat exchanger performance is its “efficiency” in transferring a
given amount of heat from one fluid at higher temperature to another fluid at lower
temperature. The efficiency can be determined if the temperatures at all flow openings are
known. In Flow Simulation the temperatures at the fluid inlets are specified and the
temperatures at the outlets can be easily determined. Heat exchanger efficiency is defined
as follows:
actual heat transfer
ε = ------------------------------------------------------------------------------
maximum possible heat transfer
The actual heat transfer can be calculated as either the energy lost by the hot fluid or the
energy gained by the cold fluid. The maximum possible heat transfer is attained if one of
the fluids was to undergo a temperature change equal to the maximum temperature
difference present in the exchanger, which is the difference in the inlet temperatures of the
hot and cold fluids, respectively: ( T hot
inle t – T i nl et ) . Thus, the efficiency of a counterflow
c ol d
T coutl et
ol d – T cold
inlet
than cold fluid capacity rate or ε = -----------------------------------
inlet – T inl et
- - if hot fluid capacity rate is more than
T hot col d
cold fluid capacity rate, where the capacity rate is the product of the mass flow and the
· c (Ref.2)
specific heat capacity: C= m
The goal of the project is to calculate the efficiency of the counterflow heat exchanger.
Also, we will determine the average temperature of the heat exchanger central tube’s wall.
The obtained wall temperature value can be further used for structural and fatigue
analysis.
Click File, Open. In the Open dialog box, browse to the Heat Exchanger.SLDASM
assembly located in the Tutorial 3 - Heat Exchanger folder and click Open (or double-
click the assembly). Alternatively, you can drag and drop the Heat Exchanger.SLDASM
file to an empty area of SolidWorks window.
Cold water = 0.02 kg/s
Tinlet = 293.2 K
Warm water
Air
Steel
Hot air = 10 m/s
Tinlet = 600 K
Creating a Project
Click Next .
6-2
3 In the Units dialog box select the desired
system of units for both input and output
(results). For this project we will use the
International System SI by default.
Click Next.
Click Next.
6 Since we have selected the Heat conduction in solids option at the Analysis Type
step of the Wizard, the Default Solid dialog box appears. In this dialog you specify the
default solid material applied to all solid components. To assign a different material to
a particular assembly component you need to create a Solid Material condition for this
component.
6-4
9 In the Results and Geometry Resolution
dialog box we accept the default result
resolution level 3 and the default minimum
gap size and minimum wall thickness.
Click Finish.
After finishing the Wizard you will complete the project definition by using the Flow
Simulation Analysis tree. First of all you can take advantage of the symmetry of the heat
exchanger to reduce the CPU time and memory required for the calculation. Since this
model is symmetric, it is possible to “cut” the model in half and use a symmetry boundary
condition at the plane of symmetry. This procedure is not required, but is recommended
for efficient analyses.
Symmetry Condition
Since we have selected Liquids as the Default fluid type and Water as the Default fluid in
the Wizard, we need to specify another fluid type and select another fluid (air) for the fluid
region inside the tube through which the hot air flows. We can do this by creating a Fluid
Subdomain. When defining a Fluid Subdomain parameters we will specify Gas as the
fluid type for the selected region, Air as the fluid and the initial temperature of 600 K and
flow velocity of 10 m/s as the initial conditions in the selected fluid region.
1 Click Flow Simulation, Insert, Fluid Subdomain.
2 Select the Air Inlet Lid inner face (in contact with the
fluid). Immediately the fluid subdomain you are going
to create is displayed in the graphics area as a body of
blue color.
6-6
5 Under Flow Parameters in the Velocity in Z Direction
1 Right-click the Boundary Conditions icon in the Flow Simulation Analysis tree and
select Insert Boundary Condition . The Boundary Condition dialog box appears.
Boundary Condition
list.
3 Accept the default Inlet
Mass Flow condition and
the default Coordinate
System and
Reference axis
.
4 Click the numerical value in the Mass Flow Rate Normal
6-8
7 In the Flow Simulation Analysis tree, right-click
the Boundary Conditions icon and select Insert
Boundary Condition.
8 Select the Water Outlet Lid inner face (in
contact with the fluid). The selected face
appears in the Faces to Apply the Boundary
Condition list.
Next we will specify the boundary conditions for the hot air flow.
13 In the Flow Simulation Analysis tree, right-click the Boundary Conditions icon and
select Insert Boundary Condition .
14 Select the Air Inlet Lid inner face (in contact with
the fluid).
The selected face appears in the Faces to Apply
6-10
21 Select the Air Outlet Lid inner face (in contact
with the fluid).
The selected face appears in the Faces to Apply
Click OK .
24 Rename the new item Environment Pressure 1
to Environment Pressure – Air.
This project involving analysis of heat conduction in solids. Therefore, you must specify
the solid materials for the model’s components and the initial solid temperature.
Notice that the auxiliary lids on the openings are solid. Since the material for the lids is the
default stainless steel, they will have an influence on the heat transfer. You cannot
suppress or disable them in the Component Control dialog box, because boundary
conditions must be specified on solid surfaces in contact with the fluid region. However,
you can exclude the lids from the heat conduction analysis by specifying the lids as
insulators.
The thermal conductivity of the Insulator substance is zero. Hence there is no heat
transferred through an insulator.
5 Rename the Insulator Solid Material 1 item to Insulators.
1 In the Flow Simulation Analysis tree, right-click the Goals icon and select Insert
Volume Goals.
6-12
2 In the Flyout FeatureManager
Design tree select the Tube part.
3 In the Parameter table select the Av
check box in the Temperature of
Solid row.
4 Accept the selected Use for Conv.
check box to use this goal for
convergence control.
5 In the Name template type
VG Av T of Tube.
6 Click OK .
1 Click Flow Simulation, Solve, Run. The Run dialog box appears.
2 Click Run.
After the calculation finishes you can obtain the temperature of interest by creating the
corresponding Goal Plot.
In addition to using the Flow Simulation Analysis tree you can use Flow Simulation
Toolbars and SolidWorks CommandManager to get fast and easy access to the most
frequently used Flow Simulation features. Toolbars and SolidWorks CommandManager
are very convenient for displaying results.
Click View, Toolbars, Flow Simulation Results.
The Flow Simulation Results toolbar appears.
Click View, Toolbars, Flow Simulation
Results Features. The Flow Simulation
Results Features toolbar appears.
If you wish, you may hide the Flow Simulation toolbars to save the space for the graphics
area, since all necessary commands are available in the CommandManager. To hide a
toolbar, click its name again in the View, Toolbars menu.
1 Click Goals on the Results Main toolbar or CommandManager. The Goals dialog
box appears.
2 Click Add All to select all goals of the project
(actually, in our case there is only one goal) .
3 Click OK . The goals1 Excel workbook is
created.
You can view the average temperature of the tube on the Summary sheet.
H eat E xch an g er.S L D A S M [L evel 3]
Goa l Na m e U n it V a lu e Ave ra g e d V a lu e M in im u m V a lu e M a x im u m V a lu e P ro gre ss [%] U se In Co n ve rg e n ce
V G A v T of Tube [K ] 328.4682387 327.4703038 324.7176733 328.4682387 100 Y es
Ite ratio n s : 51
An aly sis in terv al: 2 1
1 Click Cut Plot on the Flow Simulation Results Features toolbar. The Cut Plot
dialog box appears.
6-14
2 In the flyout FeatureManager design tree select Plane3.
3 In the Cut Plot dialog box, in addition to displaying
4 Click View Settings in order to specify the temperature as the parameter for the
contour plot. By default the Pressure is specified.
5 On the Contours tab, in the Parameter
list, select Temperature.
6 Using the slider set the Number of colors
to maximum.
11 Click Geometry on the Flow Simulation Display toolbar to hide the model.
14 Click the bottom pane and select the Isometric view on the Standard
Views toolbar.
6-16
Displaying Flow Trajectories
Notice that in the top pane the temperature contours are still displayed. The different
view settings for each pane allow you to display contour plots for different physical
parameters simultaneously.
Since we are more interested in the temperature distribution let us color the trajectories
with the values of temperature.
1 Right-click in the graphics area of the
bottom pane and select View Settings.
6-18
The water temperature range is less than the default
overall (Global) range (293 – 600), so all of the
trajectories are the same blue color. To get more
information about the temperature distribution in
water you can manually specify the range of interest.
Let us display temperatures in the range of inlet-outlet
water temperature.
The water minimum temperature value is close to 293 K. Let us obtain the values of air
and water temperatures at outlets using Surface Parameters. You will need these values to
calculate the heat exchanger efficiency and determine the appropriate temperature range
for flow trajectories visualization.
8 Click Evaluate.
9 After the parameters are calculated
click the Local tab.
You can see that the average air
temperature at the outlet is about
584 K.
6-20
Calculating the Heat Exchanger Efficiency
The heat exchanger efficiency can be easily calculated, but first we must determine the
fluid with the minimum capacity rate (C= m & c ). In this example the water mass flow rate
is 0.02 kg/s and the air mass flow rate is 0.046 kg/s. The specific heat of water at the
temperature of 300 K is about five times greater than that of air at the temperature of
584 K. Thus, the air capacity rate is less than the water capacity rate. Therefore,
according to Ref.2, the heat exchanger efficiency is calculated as follows:
inle t out let
T hot – Thot
ε = ------------------------------
i nl et
-,
Thot – T cinle ol d
t
inlet outl et
where Thot is the temperature of the air at the inlet, T hot is the temperature of the
inlet
air at the outlet and T cold is the temperature of the water at the inlet.
We already know the air temperature at the inlet (600 K) and the water temperature at the
inlet (293.2 K), so using the obtained values of water and air temperatures at outlets, we
can calculate the heat exchanger efficiency:
i nl et outlet
T hot – T hot 600 – 584-
ε = ------------------------------
i
- = --------------------------- = 0.052
inlet
Thot – T cold nlet 600 – 293.2
If you specify the range, it may be convenient to display the global (calculated over the
Computational Domain) minimum and maximum values of the current contour plot
parameter.
5 Click Display Global Min Max on the Flow Simulation Display toolbar. The
temperature global minimum and maximum values appear at the top. The points where
the parameter value reaches its minimum or maximum will be highlighted in the
graphics area by color dots. The blue dots display locations of the points, where
parameter value is minimum, while the red ones display locations of the maximum
parameter value points. of the active (bottom) pane
As you can see, Flow Simulation is a powerful tool for heat-exchanger design
calculations.
Ref. 2 J.P. Holman. “Heat Transfer” Eighth edition.
6-22
7
Mesh Optimization
The goal of this tutorial example is to demonstrate various meshing capabilities of Flow
Simulation allowing you to better adjust the computational mesh to the problem at hand.
Although the automatically generated mesh is usually appropriate, intricate problems with
thin and/or small, but important, geometrical and physical features can result in extremely
high number of cells, for which the computer memory is too small. In such cases we
recommend that you try the Flow Simulation options allowing you to manually adjust the
computational mesh to the solved problem's features to resolve them better. This tutorial
teaches you how to do this.
The Ejector in Exhaust Hood example aims to:
• Settle the large aspect ratio between the minimum gap size and the model size by
adjusting the initial mesh manually.
• Resolve small features by specifying local mesh settings.
Problem Statement
The ejector model is shown on the picture. Note that the ejector orifice’s diameter is more
than 1000 times smaller than the characteristic model size determined as the
computational domain’s overall dimension.
Exhaust
Opening
Baffles
Ejector
7-2
SolidWorks Model Configuration
Copy the Tutorial 4 – Mesh Optimization folder into your working directory and ensure
that the files are not read-only since Flow Simulation will save input data to these files.
Open the Ejector in Exhaust Hood.SLDASM assembly.
Project Definition
When you enable gravitation, pay attention that the hydrostatic pressure is calculated
with respect to the global coordinate system, as follows:
Phydrostatic = ρ(gx*x + gy*y+ g z*z), where ρ − reference density, gi - component of the
gravitational acceleration vector and x, y, z - coordinates in the global coordinate
system.
Conditions
At first, let us specify all the necessary boundary conditions because they influence the
automatic initial mesh through the automatic minimum gap size, which depends on the
characteristic size of the faces on which the boundary conditions are set.
Flow Simulation calculates the default minimum gap size using information about the
faces where boundary conditions (as well as sources, fans) and goals are specified.
Thus, it is recommended to set all conditions before you start to analyze the mesh.
The first two boundary conditions are imposed on the exhaust hood's inlet and outlet.
If you open the Initial Mesh dialog box (click Flow Simulation, Initial Mesh ) and select
the Manual specification of the minimum gap size option, you can see that the current
automatic minimum gap size is 0.5 ft, which is the width of the outlet opening (if you have
opened the Initial Mesh dialog box, click Cancel to discard changes).
The next inlet volume flow rate condition defines the gas ejected from the bottom of the
Ejector component.
7-4
If you now look at the automatic minimum gap size value (click Flow Simulation, Initial
Mesh, Manual specification of the minimum gap size), you can see that it is close to the
orifice diameter - 0.0044528 ft.
The Minimum gap size is a parameter governing the computational mesh, so that a
certain number of cells per the specified gap should be generated. To satisfy this
condition the corresponding parameters governing the mesh are set by Flow
Simulation (number of basic mesh cells, small solid features refinement level, narrow
channel resolution, etc.). Note that these parameters are applied to the whole
computational domain, resolving all its features of the same geometric characteristics
(not only to a specific gap).
Since the minimum gap size value influences the mesh in the entire computational
domain, the large aspect ratio between the model and the minimum gap size value will
produce a non-optimal mesh: not only will all small gaps be resolved, but there will also
be many small cells in places where they are not necessary. As a result, an extremely large
mesh will be produced, which may result in overly large computer memory requirements
exceeding the computers' available resources. Moreover, if the aspect ratio between the
model and the minimum gap size is more than 1000, Flow Simulation may not adequately
resolve such models with the automatically generated mesh anyway.
Finally, let us create the ejector’s porous media and apply it to the ejector’s top and side
screens.
The material you are going to create is already defined in the Engineering Database under
the Pre-Defined folder. You can skip the definition of the porous material, then when
creating the porous condition, select the pre-defined "Screen Material" from the
Engineering database.
Components to apply:
Top Screen
Side Screen
To see advantages of the local mesh and refinement options better, now let us try to
generate the computational mesh governed by the automatic mesh settings. The resulting
mesh consists of more than 1100000 cells, and cannot be processed by old computers due
to the computer memory restriction (you may get a warning message about insufficient
memory)
7-6
Manual Specification of the Minimum Gap Size
We can distinguish two very different parts of the model: a relatively big cavity having
several thin walls within and no small solid features, and the ejector’s region containing
some very fine geometrical features. Therefore, the mesh required to properly resolve the
ejector and the mesh appropriate for the rest of the model should be also very different.
Since the ejector region is a part of the entire computational domain, we need to specify
such settings for the automatic mesh generation that the model’s geometry outside the
ejector’s region will be resolved without excessive mesh splitting.
The minimum gap size value, automatically defined from the dimensions of the ejector’s
Top Screen and Side Screen components, is too small and results in excessive mesh
splitting.
To define an appropriate minimum gap size we need to examine all narrow flow passages
outside the ejector’s region:
• Boundary conditions;
• The passages connecting the ejector’s internal volume with the model’s cavity;
• The narrow flow passages between the baffles.
After reviewing the model we can accept the width of
the gap between the middle and upper baffles as the
minimum gap size. To avoid excessive mesh splitting,
we will specify the same value for the minimum wall
thickness.
1 Click Flow Simulation, Initial Mesh.
2 Use the slider to set the Level of the initial mesh to
5.
3 Select the Manual specification of the minimum
gap size checkbox and enter 0.067 ft in the
Minimum gap size box.
4 Select the Manual specification of the minimum
wall thickness checkbox and enter 0.067 ft in
the Minimum wall thickness box.
0.067 ft
5 Click OK .
The resulting mesh has significantly less cells than the mesh generated automatically with
the default values of Minimum gap size and Minimum wall thickness. The total number
of cells is less than 200 000.
7-8
Switching off the Automatic Mesh Definition
We have successfully reduced the number of cells, yet using the mesh of the higher level.
The higher level mesh provides better refinement in the regions with small geometrical
features. However, we actually do not need such a fine mesh in some regions where the
flow field changes slowly. We can further decrease the number of cells by switching off
the automatic definition of the mesh generation settings and adjusting these settings
manually. The decreased number of cells will provide us a computer memory reserve
needed to better resolve fine geometrical features of the ejector.
Click Flow Simulation, Project, Rebuild.
1 Click Flow Simulation, Initial Mesh . Switch off the automatic mesh settings by
clearing the Automatic settings check box. The Initial Mesh dialog box controls the
basic mesh and the initial mesh within the entire computational domain unless local
initial mesh settings are specified.
The mesh is named Initial since it is the mesh the calculation starts from and it could
be further refined during the calculation if the solution-adaptive meshing is enabled.
The initial mesh is constructed from the Basic mesh by refining the basic mesh cells in
accordance with the specified mesh settings. The Basic mesh is formed by dividing the
computational domain into slices by parallel planes which are orthogonal to the
Global Coordinate System’s axes.
The Initial Mesh’s parameters are currently set by Flow Simulation in accordance with the
previously specified automatic mesh settings, including Minimum gap size and Minimum
wall thickness.
2 Go to the Narrow channel tab and set
the Narrow channels refinement level
to 1. This allows us to reduce the
number of cells in the channels
between the baffles and the wall of the
Box.
7-10
Using the Local Initial Mesh Option
The ejector’s geometry is resolved reasonably well. However, if you generate the mesh
and zoom in to the ejector’s orifice region, you will see that the gas inlet face is still
unresolved. The resolution of the boundary condition surface is very important for
correctly imposing the boundary condition. To resolve the gas inlet face properly we will
use the Local Initial Mesh option.
The local initial mesh option allows you to specify an initial mesh in a local region of the
computational domain to better resolve the model geometry and/or flow peculiarities in
this region. The local region can be defined by a component of the assembly, disabled in
the Component Control dialog box, or specified by selecting a face, edge or vertex of the
model. Local mesh settings are applied to all cells intersected by a component, face, edge,
or a cell enclosing the selected vertex.
1 Click Flow Simulation, Insert, Local Initial Mesh.
2 Select the inlet face of the ejector’s orifice or click the
Inlet Volume Flow 1 boundary condition in the Flow
Simulation Analysis tree to select the face on which
this boundary condition is applied.
Now we have specified to refine all cells at the ejector’s orifice inlet face up to the
maximum level. The locally refined mesh is shown below.
The basic mesh in many respects governs the generated computational mesh. The proper
basic mesh is necessary for the most optimal mesh.
You can control the basic mesh in several ways:
• Change number of the basic mesh cells along the X, Y, Z-axes.
• Shift or insert basic mesh planes.
• Stretch or contract the basic mesh cells locally by changing the relative distance
between the basic mesh planes.
The local mesh settings do not influence the basic mesh but are basic mesh sensitive:
all refinement levels are set with respect to the basic mesh cell.
You may notice that the mesh resolving the ejector’s orifice inlet face is not symmetric. It
can has a negative effect on the specified boundary condition. We will add a control plane
to shift the boundary between cells so that it will pass through the center of the inlet face.
1 In the Initial Mesh dialog box, go to the Basic Mesh tab.
2 Click Add Plane . The Create Control Planes dialog box
appears.
3 In the Creating mode list select Reference geometry.
4 Under Parallel to select XY.
5 Zoom in to the ejector’s orifice area and select edge of the
inlet face in the graphics area. The control plane will pass
through the middle of the edge parallel to the Global
Coordinate System plane selected in the Parallel to group.
Please check that the value of offset along the Z axis,
appeared in the Control planes list, is equal to 0.703125 ft. If not, it means that you
have mistakenly selected another geometry feature. In this case, right-click on the
7-12
Control planes list and select Delete All, then try to select the edge of the inlet face
again.
6 Click OK. The Z2 Control Plane appears in the Control planes table.
You can visualize the basic mesh before solving the problem. To see the basic mesh,
click Draw basic mesh in the Initial Mesh dialog box or click Flow Simulation,
Project, Show Basic Mesh.
7 Click OK to save changes and close the Initial Mesh dialog box.
Then, generate the initial mesh to check whether the thin walls and the other geometry are
resolved.
1 Click Flow Simulation, Solve, Run.
2 Clear the Solve check box in order to generate the
mesh only.
3 Clear the Load results check box.
4 Click Run.
With the specified mesh settings the ejector’s geometry will be resolved properly. But we
need to create the mesh successfully resolving not only fine geometrical features, but the
small flow peculiarities as well. In the Ejector Analysis project such peculiarities can be
found within the internal volume of the ejector, where the thin stream of chlorine is
injected from the ejector’s orifice. Therefore the mesh within the ejector’s region must be
split additionally. To refine the mesh only in this region and avoid excessive splitting of
the mesh cells in other parts of the model, we apply a local initial mesh at the component
surrounding this region. The component was created specially to specify the local initial
mesh.
Set to resolved the LocalMesh2 component. Click Close after Flow Simulation shows
you a warning message. Note that this component was created so that there is a small
distance between the boundaries of the component and the solid feature of interest (i.e.,
the ejector). Because the local settings are applied only to the cells whose centers lie
within the selected model component, it is recommended to have the component's
boundaries offset from the solid component's walls.
7-14
After resolving the LocalMesh2 component an error message appears informing you that
the inlet volume flow condition is not in contact with the fluid region. The problem
disappears after disabling the component in the Component Control dialog box to treat it
as a fluid region.
Click Flow Simulation, Component Control and disable the
LocalMesh2 component. Click OK.
Rebuild the project by clicking Flow Simulation, Project,
Rebuild.
Next specify the local mesh settings for the ejector’s region.
1 Select the LocalMesh2 component.
2 Click Flow Simulation, Insert, Local
Initial Mesh.
3 Clear the Automatic settings check box
and switch to the Narrow Channels tab.
4 Specify the Characteristic number of
cells across a narrow channel equal to
15.
5 Use the slider to set the Narrow
channels refinement level to 3.
6 Click OK.
The settings on the Narrow Channels tab controls the mesh refinement in the model’s
flow passages. Characteristic number of cells across a narrow channel box specify
the number of initial mesh cells (including partial cells) that Flow Simulation will try
to set across the model’s flow passages in the direction normal to solid/fluid interface .
If possible, the number of cells across narrow channels will be equal to the specified
characteristic number, otherwise it will be close to the characteristic number. If this
condition is not satisfied, the cells lying in this direction will be split to satisfy the
condition.
Rebuild the project. Create the mesh again (without the following calculation) and load
the 1.cpt file.
Click Flow Simulation, Results, Display, Geometry to hide the model.
Finally, let us compare how the final mesh resolves the solid geometry and the fluid region
within the ejector with only about 100 000 cells in contrast with 1 100 000 cells generated
by the automatic mesh settings.
7-16
Flow Simulation 2009 Tutorial 7-17
Chapter 7 Mesh Optimization
7-18
8
Application of EFD Zooming
Problem Statement
Main Chip
Capacitors
Power Supply
Mother Board
Electronic
enclosure
The problem’s engineering aim is to determine the temperature of the main chip when
using one of two heat sink designs. All other conditions within the enclosure will be
invariable. As a result, we will find out the difference in cooling capability between these
two competing shapes.
No.1
No.2
The heat sink’s competing shapes (No.1 and No.2)
As you can see, all components within the electronic enclosure except the main chip’s heat
sink are specified as coarse shapes without small details, since they do not influence the
main chip’s temperature which is the aim of the analysis (the enclosure model was
preliminary simplified to this level on purpose). On the contrary, the heat sink of each
shape is featured by multiple thin (thickness of 0.1 in) fins with narrow (gaps of 0.1 in)
channels between them.
8-2
Two Ways of Solving the Problem with Flow Simulation
Flow Simulation allows us to simplify the solution of this problem. Two possible
techniques are listed below.
In the first and more direct way, we compute the entire flow inside the whole electronic
enclosure for each heat sink shape with using the Local Initial Mesh option for
constructing a fine computational mesh in the heat sink’s narrow channels and thin fins.
Naturally, the Heat conduction in solids option is enabled in these computations.
In the other, two-stage way (EFD Zooming using the Transferred Boundary Condition
option), we solve the same problem in the following two stages:
1 computing the entire flow inside the whole electronic enclosure at a low result
resolution level without resolving the heat sink’s fine features (so, the parallelepiped
envelope is specified instead of the heat sink’s comb shape) and disabling the Heat
conduction in solids option;
2 computing the flow over the real comb-shaped heat sink in a smaller computational
domain surrounding the main chip, using the Transferred Boundary Condition option
to take the first stage’s computation results as boundary conditions, specifying a fine
computational mesh in the heat sink’s narrow channels and thin fins to resolve them,
and enabling the Heat conduction in solids option.
The first stage’s computation is performed once and then used for the second stage’s
computations performed for each of the heat sink’s shapes.
Let us begin from the second (EFD Zooming) approach employing the Transferred
Boundary Condition option. Then, to validate the results obtained with this approach, we
will solve the problem in the first way by employing the Local Initial Mesh option.
The model simplification at this stage allows us to compute the electronic enclosure’s flow
by employing the automatic initial mesh settings with a lower level of initial mesh (we use
4) and accepting the automatic settings for the minimum gap size and the minimum wall
thickness. Moreover, at this stage it is also not necessary to compute heat conduction in
solids, since we do not compute the main chip temperature at this stage. Instead, we
specify surface heat sources of the same (5W) heat transfer rates at the main chip and heat
sink (parallelepiped) faces and at the small chips’ faces (they are heated also in this
example) to simulate heating of the air flow by the electronic enclosure. This is not
obligatory, but removing the heat conduction in solids at this stage saves computer
resources. As a result, the computer resources (memory and CPU time) required at this
stage are substantially reduced.
8-4
Project Definition
Using the Wizard create a new project as follows:
Fluid Air
Result and Geometry Resolution Result resolution level set to 4, other options are
default
For this project we use the automatic initial mesh and the default computational domain.
Unit System
After passing the Wizard, first we will adjust the system of units. The new custom system
of units is based on the selected USA pre-defined system, but uses Watts for power, and
inches for length.
1 Click Flow Simulation, Units.
2 Specify Inch for the Length and Watt for the
Total heat flow & power.
3 Click Save .
Conditions
We specify External Inlet Fan at the inlet, Environment Pressure at three outlets. For more
detailed explanation of how to set these conditions please refer to the First Steps -
Conjugate Heat Transfer tutorial.
8-6
Outlet Environment Pressure:
Boundary Default thermodynamic
Condition parameters (ambient
pressure of 14.6959 lbf/in 2,
temperature of 68.09 °F)
for the Environment
pressure at the Outlet Lids.
Heat Sources
As mentioned earlier in this chapter, to simulate the flow heating by the electronic
enclosure, we specify surface heat sources of the same (5W) heat transfer rates at the main
chip and the heat sink (parallelepiped) faces and at the small chips’ faces. Since we do not
consider heat conduction in solids in this project, the surface source can be applied only to
faces in contact with fluid. Follow the steps below to create the sources on the necessary
faces:
1 Click Flow Simulation, Insert, Surface Source.
In the Flyout FeatureManager Design Tree, select the Heat Sink and Main Chip
components. Flow Simulation automatically selects all faces of the Heat Sink and
Main Chip components. Faces that are not in contact with fluid must be removed from
the Faces to Apply the Surface Source list.
faces to remove. The Filter allows you to remove unnecessary faces of specified type
from the list of selected faces.
4 Set the value of the source to 5 W.
5 Click OK .
Goals
Specify the surface goals of mass flow rate at the inlet and outlet.
Run the calculation. After the calculation is finished you can start the second stage of EFD
Zooming to focus on the main chip.
Save the model.
8-8
Second Stage of EFD Zooming
At the 2nd stage of EFD Zooming aimed at determining the main chip’s temperature, we
compute the flow over the heat sink in a smaller computational domain surrounding the
main chip, using the Transferred Boundary Condition option to take the first stage’s
computation results as boundary conditions. To compute the solids temperature, we enable
the Heat conduction in solids option. Since at this stage the computational domain is
reduced substantially, a fine computational mesh with an affordable number of cells can
be constructed in the heat sink’s narrow channels and thin fins, even when considering
heat conduction in solids during computation.
Project Definition
Using the Wizard create a new project as follows:
Fluid Air
Here, we use the automatic initial mesh by specifying the Result resolution level (Level
of initial mesh) of 4, but in contrast to the first stage’s computation, we specify manually
the minimum gap size of 0.1 in to resolve the fine features of heat sink.
Next, we will reduce the computational domain to focus on the main chip, i.e. perform
EFD Zooming.
Computational Domain
When reducing the computational domain for EFD Zooming purposes, it is necessary to
take into account that the first stage’s computation results will serve as the boundary
conditions at this domain’s boundaries. Therefore, to obtain reliable results in the second
stage’s computations, we have to specify computational domain boundaries (as planes
parallel to the X-, Y-, Z-planes of the Global Coordinate system) satisfying the following
conditions:
1 the flow and solid parameters at these boundaries, taken from the first stage’s
computation, must be as uniform as possible;
2 the boundaries must not lie too close to the object of interest, since the object’s features
were not resolved at the first stage’s computation. The computational domain must be
large enough not to receive influence from more complex features of the newly added
object;
3 the boundary conditions transferred to or specified at the boundaries must be consistent
with the problem’s statements (e.g., if in the problem under consideration the mother
board is made of a heat-conducting material, then it is incorrect to cut the mother board
with computational domain boundaries, since this will yield an incorrect heat flux from
the chip through the mother board).
In this project we specify the following computational domain boundaries satisfying the
above-mentioned requirements. Click Flow Simulation, Computational Domain to adjust
the computational domain size as follows:
• Xmin = -2.95 in (entirely lies inside the electronic enclosure side wall made of
aluminum, this material does not influence the main chip’s temperature since it is
insulated from the chip by the heat-insulating mother board and the air flow, its
8-10
boundary condition is automatically specified as the 68.09 °F temperature specified
as the initial condition for all solids),
• Xmax = 0.7 in (the boundary conditions in the fluid region of this boundary are
transferred from the first stage’s computation results, the same boundary conditions
as at Xmin = -2.95 in are automatically specified at this boundary’s upper solid part
lying in the electronic enclosure’s aluminum wall, and the same boundary
conditions as at Zmin = -1 in are automatically specified at the lower solid part lying
in the mother board),
• Ymin = -1 in, Ymax = 4 in (the boundary conditions at these boundaries are specified
in the same manner as at Xmax = 0.7 in, as well as at the boundaries’ side parts also
lying in the aluminum wall),
• Zmin = -1.1 in (entirely lies inside the mother board specified as a heat insulator,
therefore the adiabatic wall boundary condition is automatically specified at this
boundary),
• Zmax = 1.2 in (entirely lies inside the electronic enclosure’s aluminum upper wall,
therefore the same boundary condition, as at Xmin = -2.95 in, are automatically
specified at this boundary).
Conditions
First, we specify Transferred Boundary Conditions.
5 In the Browse for Project dialog select the Zoom – Global - L4 configuration and
click OK.
6 Click Next .
7 At Step 3, accept Ambient as the Boundary condition type.
8-12
Heat Sources
Volume Source of 5W heat
generation rate in the main chip;
Solid Materials
a) Main Chip is made of silicon
(Pre-Defined/Semiconductors);
c) all other parts (e.g. the heat sink) are made of aluminum.
Goals
Specify the Volume Goals of maximum and average
temperatures of the main chip and the heat sink.
Run the calculation.
The obtained computational results are presented in tables and pictures below. These
results were obtained with the heat sink’s shape No.1.
If you look at the computational mesh you can see that it has two cells for each of the heat
sink’s channels, and two cells for each of the sink’s fins.
The mesh cut plot obtained for the heat sink No.1 at Y=-0.3 in.
8-14
In fact, the Minimum gap size and Minimum wall thickness influence the same
parameter, namely, the characteristic cell size. By default, Flow Simulation generates
the basic mesh in order to have a minimum of two cells per the specified Minimum gap
size. The number of cells per the Minimum gap size depends non-linearly on the Level
of initial mesh and cannot be less than two. In turn, the Minimum wall thickness
condition induces Flow Simulation to create the basic mesh having two cells (two cells
are enough to resolve a wall) per the specified Minimum wall thickness (regardless of
the specified initial mesh level). That’s why, if the Minimum wall thickness is equal to
or greater than the Minimum gap size, then the former does not influence the resulting
mesh at all.
After cloning the project you can start the calculation immediately.
The obtained results are presented in tables and pictures below. It is seen that due to the
new shape of the heat sink the main chip’s temperature is reduced by about 15 °F. That is
caused by both the increased area of the heat sink’s ribs and streamlining the flow in the
heat sink’s narrow channels between the ribs (in heat sink No.1 about half of the channel is
occupied by a counterflow vortex).
To validate the results obtained with the EFD Zooming approach, let us now solve the
same problems employing the Local Initial Mesh option. To employ this option, we add a
parallelepiped surrounding the main chip to the model assembly and then disable it in the
Component Control dialog box. This volume represents a fluid region in which we can
specify computational mesh settings differing from those in the other computational
domain, using the Local Initial Mesh option.
The electronic enclosure configuration with the additional part for applying the Local Initial Mesh op
Flow Simulation Project for the Local Initial Mesh Approach (Sink No1)
To create the project we clone the Zoom – SinkNo1 - L4 to the existing LocalMesh –
SinkNo1 - N2 configuration, but in contrast to the previous cloning we reset the
computational domain to the default size so the computational domain encloses the entire
model.
Activate Zoom – SinkNo1 - L4 configuration.
8-16
Open the Clone Project dialog, click Add to
existing and, in the Existing configuration list
select the LocalMesh – SinkNo1 - N2 as the
configuration to which Flow Simulation will attach
the cloned project.
After clicking OK, confirm with Yes both the
appearing messages.
Conditions
First remove the inherited transferred boundary
condition. Right-click the Transferred
Boundary Condition1 item in the tree and select
Delete.
Next, copy the boundary conditions from the Zoom – Global - L4 configuration using the
Copy Feature tool.
1 ActivateZoom – Global - L4 configuration.
2 Click Flow Simulation, Tools, Copy Features. The
Copy Features dialog box appears.
3 Switch to the Flow Simulation analysis tree tab, hold
down the Ctrl key and in the Flow Simulation Analysis
tree select Environment Pressure 1 and External Inlet
Fan 1 items. These features appear in the Features to
copy list.
4 Select LocalMesh – SinkNo1 - N2 as the Target
Project.
5 Click OK .
6 Activate LocalMesh – SinkNo1 - N2 configuration.
Heat Sources
To the already existing volume
source of the 5W heat generation rate
in the main chip, add the total 5W
heat generation rate in the small
chips.
Solid Materials
The following material definitions
were inherited from the previous
project so you do not need to create
them again, but you need to edit the
Silicon Solid Material 1 to include
small chips and to edit Insulator Solid Material 1 to include inlet and outlet lids:
a) the Main Chip and small chips are made of silicon;
b) the MotherBoard, the Enclosure, the Inlet Lid and the Outlet Lids are made of
insulator;
c) PCB1 and PCB2 are made of user defined Tutorial PCB material, which is added to
the Engineering Database in the First Steps - Conjugate Heat Transfer tutorial example.
d) all other parts are made of the default aluminum.
Goals
Keep the cloned volume goals of maximum and average temperatures of the main chip
and the heat sink.
8-18
Set the Level of initial mesh to 3. Since heat
conduction in solids is enabled, setting the Level of
initial mesh to 4 together with the local mesh settings
will produce large number of cells resulting in longer
CPU time. To decrease the calculation time for this
tutorial example we decrease the Level of initial mesh
to 3. Note that the Result resolution level is still equal
to 4 as it was specified in the Wizard. To see the value
of the result resolution level, click Flow Simulation, Calculation Control Options, and
then click Reset. To close the Reset dialog box, click Cancel.
Click Flow Simulation, Project, Rebuild.
The Narrow Channels term is conventional and used for the definition of the model’s
flow passages in the normal-to-solid/fluid-interface direction. The procedure of
refinement is applied to each flow passage within the computational domain unless you
specify for Flow Simulation to ignore the passages of a specified height with the
Enable the minimum height of narrow channels and Enable the maximum height of
narrow channels options. The Characteristic number of cells across a narrow
channel (let us denote it as Nc) and Narrow channels refinement level (let us denote
it as L) both influence the mesh in narrow channels in the following way: the basic
mesh in narrow channels will be split to have the specified Nc number per a channel, if
the resulting cells satisfy the specified L. In other words, whatever the specified Nc, a
narrow channel’s cells cannot be smaller in 8L (2L in each direction of the Global
Coordinate System) times than the basic mesh cell. This is necessary to avoid the
undesirable mesh splitting in superfine channels that may cause increasing the number
of cells to an excessive value.
In our case, to ensure the 2 cells across a channel criterion, we increased the Narrow
channels refinement level to 4.
We perform these settings for both of the heat sinks under consideration.
Flow Simulation Project for the Local Initial Mesh Approach (Sink No2)
Clone the active LocalMesh – SinkNo1 - N2 to the
existing LocalMesh – SinkNo2 - N2 configuration.
While cloning confirm the message to rebuild the
mesh.
Results
The computational results obtained for both of the heat sinks are presented below in
comparison with the results obtained with the EFD Zooming approach. It is seen that
computations with the local mesh settings yield practically the same results as the EFD
Zooming approach, therefore validating it.
The computed maximum and average main chip and heat sink temperatures when
employing the different heat sinks.
H e a t sink N o .1 H e a t s ink N o .2
Zo o m - L o c a lM e sh - Zo o m - Lo c a lM e sh -
S ink N o 1 - S ink N o 1 - S ink N o 2 - S ink N o 2 -
P a ra me te r L4 N2 L4 N2
t m a x , °F 1 11.1 114.1 96.4 9 9.4
M a in c hip t a v e r , °F 1 10.8 113.8 96.1 9 9.2
t m a x , °F 111 114.1 96.3 9 9.4
H e a t sink t a v e r , °F 1 10.6 113.7 95.9 99
8-20
EFD Zooming Local Mesh
The temperature cut plots obtained for heat sink No.1 at Y=2.19 in (Top plane) with the EFD Zooming (left) and
Local Mesh (right) approaches.
The temperature cut plots obtained for heat sink No.1 at Z= -0.32 in (Front plane) with the EFD Zooming (left) and
Local Mesh (right) approaches.
The temperature cut plots obtained for heat sink No.1 at X= -1.53 in (Right plane) with the EFD Zooming (left) and
Local Mesh (right) approaches.
The temperature cut plots obtained for heat sink No.2 at Y=2.19 in (Top plane) with the EFD Zooming (left) and
Local Mesh (right) approaches.
The temperature cut plots obtained for heat sink No.2 at Z=-0.32 in (Front plane) with the EFD Zooming (left)
and Local Mesh (right) approaches.
The temperature cut plots obtained for heat sink No.2 at X= -1.53 in (Right plane) with the EFD Zooming (left) an
Local Mesh (right) approaches.
8-22
9
Textile Machine
Problem Statement
The simplified textile machine used by this tutorial is described as a closed hollow cylinder
having a cylindrical stator with a narrow inlet tube. A thin-walled cone rotates at a very high
speed. The air flows over the rotating cone before leaving through the outlet pipe. Due to the
shear stress, the rotating cone swirls the air. The swirling air motion orients the fibers, for the
correct formation of yarn.
In this example1 a hollow cylinder with the following dimensions were used: 32 mm inner
diameter and 20 mm inner height. Air is injected into an inlet tube of 1 mm diameter at a
mass flow rate of 0.0002026 kg/s. The cone thickness is 1 mm and the cone's edge is
spaced at 3 mm from the bottom of the main cylinder. The cone rotates at a speed of
130000 RPM. The static pressure of 96325 Pa is specified at the cylinder's outlet tube exit.
Flow Simulation analyzes the air flow without any fiber particles. The influence of the
fiber particles on the air flow was assumed to be negligible. Small polystyrene particles
were injected into the air stream using the postprocessor Flow Trajectory feature to study
the air flows influence on the fibers. A 40 m/s tangential velocity of air is specified as an
initial condition to speed up convergence and reduce the total CPU time needed to solve
the problem.
Outlet
P = 96325 Pa
Housing
Rotating wall
ω = 130000 RPM
Stator
1mm
Copy the Tutorial Advanced 2 - Rotating Walls folder into your working directory and
ensure that the files are not read-only since Flow Simulation will save input data to these
files. Open the Textile Machine.SLDASM assembly.
9-2
Project Definition
Fluid Air
Conditions
1 In the Flow Simulation Analysis tree, right-click the Boundary Conditions icon and
select Insert Boundary Condition.
2 Select Wall ,
then Real Wall.
3 In the Flyout
FeatureManager
Design Tree select
the Rotor
component. All the
rotor’s faces are
selected. However,
the top face is out of
the computational
domain and must be
excluded.
9-4
6 Select Wall Motion .
12 Click OK .
13 Click-pause-click the new Initial Condition1 item and
rename it to vel = 40 m\s.
Specifying Goals
Since the rotating cone swirls the air, it make sense to specify the air velocity as a goal to
ensure the calculation stops when the velocity is converged. In addition, let us specify the
static pressure surface goal at the inlet and the mass flow rate surface goal at the outlet as
additional criteria for converging the calculation.
Specify the following project goals:
Surface Goal Mass Flow Rate Outlet face(click the outlet static
pressure boundary condition item to
select the outlet face)
Surface Goal Average Static Pressure Inlet face(click the inlet mass flow rate
boundary condition item to select the
inlet face)
9-6
Volume Goal Average Velocity Initial Velocity 2
(select the component in the Flyout
FeatureManager Design Tree)
The calculated flow velocity field and velocity Y-component field at Z = 0 (XY section)
are shown in the pictures below. It can be seen that the maximum flow velocity occurs
near the inlet tube and near the rotating cone's inner surface at the cone's edge.
It is interesting that the vertical (i.e. along the Y axis) velocity in the region close to the
rotating cone's internal and external surfaces is directed to the cylinder bottom. Also, this
velocity component is nearly zero in the gap between the rotating cone and the bottom of
the cylinder, and positive (i.e. directed to the top) in the vicinity of the cylinder's side
walls. As a result, small particles carried by the air into the region between the lower edge
of the rotating cone and the bottom of the cylinder cannot leave this region due to the
small vertical velocity there. On the other hand, larger particles entering this region may
bounce from the cylinder’s bottom wall (in this example the ideal, i.e. full reflection is
considered) and fly back to the region of high vertical velocity. Then they are carried by
the air along the cylinder's side walls to the cylinder's top wall where they remain in this
region's vortex.
1 In the Flow Simulation Analysis tree, right-click the Flow Trajectories icon and select
Insert.
2 Click the Flow Simulation Analysis tree tab and then
click the inlet boundary condition icon
(Inlet Mass Flow = 0.0002026) to select the inlet face
from which the particles are injected.
9-8
5 Under Constraints increase the Maximum Length
of trajectories to 15000 mm.
9-10
18 In the Particle Study Results dialog box,
select Injection 1 and click 3D-View
Options.
19 Select to draw trajectories as Line with
Arrow and click OK.
In the previous calculation zero roughness was used for the walls of the rotating cone's
internal and external surfaces. To investigate an influence of the rotating cone wall's
roughness, let us perform the calculation with the rotating cone's internal and external
surfaces' at 500 µm roughness under the same boundary conditions.
Create a new configuration by cloning the current project, and
name it 130000rpm - rough wall .
4 Click OK .
The calculated fields of flow velocity and Y-component of velocity in different section are
shown below and reveal practically no change in the vertical velocity of the flow. As a
result, the flying particles’ trajectories are nearly identical to those in the case of smooth
walls. It is seen that increase in the roughness from 0 to 500 µm increases the vortex flow's
tangential velocity.
Velocity in the XY section at Z = 0 (roughness = 500 µm)
9-12
Flow streamlines
Trajectories of 5 µm particles
Trajectories of 15 µm particles
9-14
10
Non-Newtonian Flow in a Channel with Cylinders
Problem Statement
Copy the Tutorial Advanced 3 - Non-Newtonian Flow folder into your working
directory and ensure that the files are not read-only since Flow Simulation will save input
data to these files. Open the Array of Cylinders.sldprt part.
3 Click New Item in the toolbar. The blank Item Properties tab appears.
Double-click the empty cell to set the corresponding property value.
4 Specify the material properties as shown in the table below:
Name XGum
Project Definition
10-2
Wall Conditions Adiabatic wall, default smooth walls,
default slip condition
Conditions
Specifying Goals
Specify surface goals for the Average
Total Pressure at the inlet and outlet.
Specify an equation goal for the total
pressure drop between the channel’s inlet
and outlet.
Run the calculation. When the calculation is finished, create the goal plot to obtain the
pressure drop between the channel’s inlet and outlet.
A r r a y o f C y li n d e r s .S L D P R T [X G S ]
G o a l Na m e U n it V a lu e A ve r a g e d V a lu e M in im u m V a lu e M a x im u m V a lu e P ro g r e ss [%]
S G A v To tal P re s s ure 1 [P a] 10562 2.4926 10562 2.4125 105620 .3901 1 05627 .4631 1 00
S G A v To tal P re s s ure 2 [P a] 10132 9.0109 10132 9.0091 101329 .0051 1 01329 .0109 1 00
P re s s ure D rop [P a] 4293. 481659 429 3.4034 4298.4 57377 4 291.3 80166 1 00
Let us now consider the flow of water in the same channel under the same conditions (at
the same volume flow rate).
Create a new configuration by cloning the current
project, and name it Water.
Run the calculation. After the calculation is finished, create the goal plot.
10-4
A r r a y o f C y li n d e r s .S L D P R T [w a te r ]
G o a l Na m e U n it V a lu e A ve r a g e d V a lu e M in im u m V a lu e M a x im u m V a lu e P ro g r e ss [%]
S G A v To tal P re s s ure 1 [P a] 1013 95.004 10139 5.0214 101394 .8731 1 01395 .1171 1 00
S G A v To tal P re s s ure 2 [P a] 10132 9.3912 10132 9.3378 101329 .3084 1 01329 .3912 1 00
P re s s ure D rop [P a] 65.6 128767 65.683 57061 65.765 66097 6 5.552 43288 1 00
As shown in the results table above, the channel's total pressure loss is about 60 Pa, i.e.
60...70 times lower than with the 3% aqueous solution of xanthan gum, this is due to the
water's much smaller viscosity under the problem's flow shear rates.
The XGS (above) and water velocity distribution in the range from 0 to 30 cm/s.
1 Georgiou G., Momani S., Crochet M.J., and Walters K. Newtonian and Non-Newtonian
Flow in a Channel Obstructed by an Antisymmetric Array of Cylinders. Journal of
Non-Newtonian Fluid Mechanics, v.40 (1991), p.p. 231-260.
10-6
11
Heated Ball with a Reflector and a Screen
Problem Statement
Copy the Tutorial Advanced 4 - Surface-to-surface Radiation folder into your working
directory and ensure that the files are not read-only since Flow Simulation will save input
data to these files. Open the Heated Ball Assembly.SLDASM assembly.
11-2
Case 1
Project Definition
Using the Wizard create a new project as follows:
Unit system SI
11-4
5 Click Flow Simulation, Insert, Radiative Surface.
6 Under Type, expand the list of Pre-Defined radiative
surfaces and select the Whitebody wall.
7 Select the Reflector’s inner surface.
Case 2
4 Click OK .
11-6
Goals Specification
Specify the additional surface goals for the maximum, average, and minimum
temperatures of the Reflector's inner and outer surfaces.
Case 3
Results
In Case 1, due to the radiation returned by the reflector, the ball’s surface facing the
reflector is hotter than the ball’s surface facing the screen (see pictures below). Therefore,
screen temperature in Case 1 is higher than in the other cases.
In Case 2, radiation coming from the ball to the reflector heats up the reflector and heat is
radiated from the reflector’s outer surface to ambient, therefore being lost from the
system. Since the heat returned to the ball by the reflector’s radiation is smaller, the ball’s
temperature is lower, although distributed over the ball in the same manner as in Case 1.
The heat coming from the reflector to the screen is also smaller. As a result, the screen’s
temperature is lower than in Case 1.
Since the reflector is removed in Case 3, there is no noticeable heat radiated back to the
ball. The ball’s temperature is lower than in Case 2 and mostly uniform (the
non-uniformity is lower than 1 K). Since the screen acquires radiation from the ball only,
the screen’s temperature is the lowest among all the cases.
The ball temperature distribution (front plane cross-section) in CASE 1 (left), CASE 2 (center) and
CASE 3 (right) in the range from 1200 to 1220 K (the reflector is arranged at the left).
The screen temperature distribution (surface plot of solid temperature) in CASE 1 (left), CASE 2
(center) and CASE 3 (right) in the range from 295 to 340 K.
11-8
P a rameter C ase 1 C a se 2 C ase 3
M a ximum 1 2 3 3 .0 4 1 2 0 6 .3 1 1 1 9 5 .9 7
T he b all’s temp erature, K A vera ge 1 2 2 2 .1 6 1 2 0 3 .3 0 1 1 9 5 .4 7
M inimum 1 2 1 1 .9 8 1 2 0 0 .3 8 1 1 9 5 .3 9
M a ximum 3 4 5 .9 3 3 2 4 .8 9 3 1 3 .6 0
T he scree n’s temp era ture , K A vera ge 3 1 7 .7 8 3 0 8 .9 0 3 0 3 .5 5
M inimum 3 0 6 .7 4 3 0 2 .5 2 2 9 9 .5 1
11-10
12
Rotating Impeller
Problem Statement
Let us consider the air flow through a centrifugal pump having a rotating impeller (see
below).1 This pump has a stationary axial inlet (an eye), a pipe section of 92 mm radius
with a central body of circular arc contour, which turns the flow by 90o from the axial
direction. At the inlet's exit the radial air flow is sucked by a rotating impeller, which has
seven untwisted constant-thickness backswept blades with wedge-shape leading and
trailing edges. Each blade is cambered from 65o at the impeller inlet of 120 mm radius to
70o at the impeller exit of 210 mm radius, both with respect to the radial direction. These
blades are confined between the impeller shrouding disks rotating with the same (as the
blades) angular speed of 2000 rpm. Downstream of the impeller the air enters a stationary
(non-rotating) radial diffuser.
To complete the problem statement, let us specify the following inlet and outlet boundary
conditions: inlet air of 0.3 m3/s volume flow rate having uniform velocity profile with
vectors parallel to the pump's axis; at the radial-directed outlet a static pressure of 1 atm is
specified.
Ω = 2000 rpm
Outlet Static Pressure
Copy the Tutorial Advanced5 - Rotating Impeller folder into your working directory.
Open the Pump.SLDASM assembly.
Project Definition
Unit system SI
12-2
Conditions
Relative to rotating frame. When the Relative to rotating frame option button is
selected, the specified velocity (Mach number) is assumed to be relative to the rotating
reference frame (Vr):
V specified = V r = V abs − ω × r
Here, r is the distance from the rotation axis and ω is the angular velocity of the
rotating frame. The mass or volume flow rate specified in the rotating reference frame
(the Relative to rotating frame option is selected) will be the same in the absolute
(non-rotating) frame of reference if the tangential velocity component is perpendicular
to the opening’s normal, thus not influencing the mass (volume) flow rate value, e.g.
when the opening's normal coincides with the rotation axis.
Outlet Outlet Environment
Boundary Pressure:
Condition Default value (101325 Pa)
for the Environment pressure
(in the absolute frame of
reference - the Pressure
potential is disabled) at the
radial outlet face.
Pressure potential. If you enable a rotating reference frame, you can select the
Pressure potential check box. When the Pressure potential check box is selected, the
specified static pressure is assumed to be equal to the rotating frame pressure (Pr) and
may be calculated using following parameters: absolute pressure, density, angular
velocity and radius:
1
Pspecified = Pr = Pabs − ρω 2 ⋅ r 2
2
When the Pressure potential check box is unchecked, the specified static pressure is
assumed to be a pressure in terms of the absolute frame of reference (Pabs).
When you specify a rotating reference frame, it is assumed that all model walls are rotated
with the reference frame's angular velocity unless you set a specific wall to be stationary.
To specify a non-rotating wall, the Stator moving wall boundary condition can be applied
to this wall. Specifying the stator boundary condition is the same as specifying the zero
velocity of this wall in the absolute (non-rotating) frame of reference. Note that stator face
must be axisymmetric with respect to the rotation axis.
4 Select Stator.
5 Click OK and rename the new Real Wall 1 condition to Stator Walls.
12-4
Impeller’s Efficiency
Engineers dealing with pump equipment are interested in the pump efficiency. For the
pump under consideration the efficiency (η) can be calculated in the following way
(F.M.White "Fluid Mechanics", 3rd edition, 1994):
η=
( Poutlet − Pinlet ) ⋅ Q
Ω ⋅M
where Pinlet is the static pressure at the pump’s inlet, Poutlet is the bulk-average static
pressures at the impeller’s outlet (Pa), Q is the volume flow rate (m3/s), Ω is the impeller
rotation angular velocity (rad/s), and M is the impeller torque (N·m). To obtain Poutlet, an
auxiliary measure component was placed where the flow exits the impeller.
The measure component is only used for the pressure
measurement (the corresponding goal will be specified at the
inner face of the measure thin ring), thus it should be
disabled in the Component Control dialog box.
1 Click Flow Simulation, Component Control.
2 Select the Measure item and click Disable.
3 Click OK to close the dialog.
First, since the pressure and volume flow rate boundary condition are specified, it makes
sense to set the mass flow rate surface goal at the pump’s inlet and outlet to inspect the
mass balance as an additional criterion for converging the calculation.
Next, specify the goals that are necessary for calculating the impeller’s efficiency:
Surface Goal Z - Component of Torque All impeller faces in contact with air
(see details below).
To avoid manual selecting of all impeller’s faces in contact with air (more than 150) we
will use the Filter Faces feature.
1 Select the Impeller component by clicking on it in the graphic area or in the
FeatureManager Design tree.
2 Right-click on Goals item in the Analysis tree and select
Insert Surface Goals. All impeller faces (including those
we do not actually need) appear in the Faces to Apply the
Surface Goal list.
12-6
Finally, specify the following Equation goals:
To add inlet volume flow value to the equation goal’s expression, click the
Inlet Volume Flow 1 item in the Analysis tree and then click Volume flow rate normal
to face in the Parameter list.
Results
The velocity vectors and static pressure distribution are shown below. To display vectors
in the rotating reference frame select the Velocity RRF parameter under the Vectors tab of
the View Settings dialog box.
The flow velocity vectors in the frame rotating with the impeller (left) and in the
stationary frame (right) at the impeller flow passage midsection (Z = - 0.02 m, Front
plane, vector spacing = 0.02m, arrow size = 0.03m).
12-8
The flow pressure distribution
12-10
13
CPU Cooler
Problem Statement
Let us consider a CPU cooler consisting of a copper core and an aluminum heat sink with
62 fins. An eight-blade propeller generates a constant flow of air through the heat sink.
The CPU is mounted on a socket installed on a PCB. Heat produced by the CPU is
transferred through the core to the heat sink and then released into the air flow.
To calculate the problem using
Flow Simulation, it is convenient Fan
to use the concept of local rotating
regions. In order to simplify the
problem statement, we do not
consider the thermal interface
layer between the processor and
the cooler. Also, we neglect the
Heat sink
thermal conduction through the
Copper core
processor socket and PCB.
CPU
A quantitative measure of the
cooler efficiency is the thermal
characterization parameter
Ψ CA = ( T C – T A ) ⁄ PD , where Tc is
the temperature of the CPU cover,
TA is the surrounding air
temperature, and PD is the thermal
design power (TDP) of the CPU.
An exploded view of the CPU cooler assembly.
SolidWorks Model
Configuration
Copy the Tutorial Advanced 6 - CPU Cooler folder into your working directory. Open
the CPU Cooler.SLDASM assembly.
Project Definition
Unit system SI
Result and Geometry Resolution Set the Result resolution level to 5; Minimum gap
size = 0.001 m, other options are default
Computational Domain
13-2
Rotating Region
The Rotating region is used to calculate flow through rotating components of model
(fans, impellers, mixers, etc.) surrounded by non-rotating bodies and components, when a
global rotating reference frame cannot be employed. For example, local rotating regions
can be used in analysis of the fluid flow in the model including several components
rotating over different axes and/or at different speeds or if the computational domain has a
non-axisymmetrical (with respect to a rotating component) outer solid/fluid interface.
Each rotating solid component is surrounded by an axisymmetrical rotating region which
has its own coordinate system rotating together with the component.
A rotaing region is defined by an additional component of the model. This additional
component must meet the following requirements:
• the rotating component must be fully enclosed by it,
• it must be axisymmetric (with respect to the rotating component's rotation axis),
• its boundaries with other fluid and solid regions must be axisymmetrical too, since
the boundaries are sliced into rings of equal width and the flow parameters' values
transferred as boundary conditions from the adjacent fluid regions are
circumferentially averaged over each of these rings,
• the components defining different rotating regions must not intersect.
Specify the rotating region as follows:
1 Click Flow Simulation, Insert, Rotating Region.
2 In the flyout FeatureManager design tree select Rotation Region component. Note that
the Disable solid components check box is automatically selected to treat the Rotating
Region as a fluid region.
By placing the
rotating region
boundary within a
solid instead of
putting it into a
narrow channel
between the fan and
the attach clip we
avoid the additional
mesh refinement
and the negative
effects of the
non-axisymmetrical
flow in this narrow
channel
13-4
3 Under Parameter, in the Angular Velocity box, specify the angular rotation
velocity of -4400 RPM.
4 Click OK .
When you specify a rotating region, it
is assumed that all model walls within
this region rotate with the region's
angular velocity unless you set a
specific wall to be stationary. To specify a non-rotating wall, the Stator real wall boundary
condition should be applied to the wall. Specifying the stator boundary condition is the
same as specifying the zero velocity of this wall in the absolute (non-rotating) frame of
reference. Note that the stator face (or a part of the face that is located inside the rotating
region in the case when the given face intersects with the rotating region boundary) must
be axisymmetric with respect to the rotation axis.
We will specify the stator condition at the appropriate walls of the fan attach and the
attachment clip. To easily select the necessary faces, hide the Fan and Rotation Region
components.
1 Click Flow Simulation, Insert, Boundary Condition.
Since we have specified the Exclude internal space option in the Wizard, the faces in
contact with the cavity between the Fan Attach and the Copper Core are considered
outer faces. Therefore we need to select the Remove outer faces option in Filter
Faces in order to exclude them.
6 Click Filter.
7 Click OK .
Solid Materials
Heat Source
Define the volume source with the heat generation rate of 75 W in the CPU component.
To resolve the complex geometry of the fan and heat sink better, let us define six
additional control planes and specify the proper Ratios for the intervals between them to
make the mesh denser in the central region containing the complex geometry and coarser
near the computational domain’s boundaries.
13-6
1 Click Flow Simulation, Initial Mesh.
2 Clear the Automatic settings check box.
3 On the Basic Mesh tab, under
Control Intervals select the 0 m
value (either as a Max of X1 interval
or as a Min of X2 interval) and click
Delete plane.
Ratio is the ratio of cell sizes on the given interval. The cell sizes are changed
gradually along the selected direction so that the proportion between the first and the
last cells of this interval is close (but not necessarily equal) to the entered value of the
Ratio. Negative values of the ratio correspond to the reverse order of cell size increase.
13-8
Specifying Project Goals
Specify surface goals for maximum temperature on the CPU cover and mass flow rate for
the flows entering the rotating region and exiting from it. To select the necessary faces,
you will probably need to hide temporarily some components of the assembly.
Equation goal ({SG Mass Flow Rate The disbalance of the inlet and outlet mass
1}+{SG Mass Flow flow rates. We are using the "+" operand
Rate 2})/{SG Mass since the inlet and outlet mass flow rate
Flow Rate 1} values have opposite signs.
Select No units for Dimensionality.
To calculate the thermal characterization parameter we will need the temperature of the
center of the CPU cover. To get more accurate value of the parameter we will specify a
separate point goal.
1 Click Flow Simulation, Insert, Point Goals.
6 Click OK .
13-10
Results
Use the goal plot tool to obtain the value of the temperature of the center of the CPU
cover. Now we can calculate the thermal characterization parameter of the heat sink:
Ψ CA = ( T C – TA ) ⁄ PD = (329.9-311.15)/75 = 0.25 °C/W. The second most important
characteristic of the CPU Cooler is the velocity of the flow above PCB. We can assess the
value of this parameter as well as the distribution of the temperature by looking at the cut
plots made in the Front and Right planes (see below).
Temperature field and velocity vectors distribution (Front plane, no offset, vector
spacing = 0.003 m, uniform plot, projected vectors, arrow size = 0.015 m).
Temperature field and velocity vectors distribution (Right plane, no offset, vector
spacing = 0.003 m, uniform plot, projected vectors, arrow size = 0.015 m).
13-12