0% found this document useful (0 votes)

888 views

Fluent Tutorial

Fluent tutorial

Uploaded by

luis900000

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

888 views

Fluent Tutorial

Fluent tutorial

Uploaded by

luis900000

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 126

FLUENT 14.

5 Beta Features Manual

ANSYS, Inc. Southpointe 275 Technology Drive Canonsburg, PA 15317 [email protected] http://www.ansys.com (T) 724-746-3304 (F) 724-514-9494

Release 14.5 October 2012

ANSYS, Inc. is certified to ISO 9001:2008.

Beta Features Manual Copyright and Trademark Information

2012 SAS IP, Inc. All rights reserved. Unauthorized use, distribution or duplication is prohibited. ANSYS, ANSYS Workbench, Ansoft, AUTODYN, EKM, Engineering Knowledge Manager, CFX, FLUENT, HFSS and any and all ANSYS, Inc. brand, product, service and feature names, logos and slogans are registered trademarks or trademarks of ANSYS, Inc. or its subsidiaries in the United States or other countries. ICEM CFD is a trademark used by ANSYS, Inc. under license. CFX is a trademark of Sony Corporation in Japan. All other brand, product, service and feature names or trademarks are the property of their respective owners.

Disclaimer Notice
THIS ANSYS SOFTWARE PRODUCT AND PROGRAM DOCUMENTATION INCLUDE TRADE SECRETS AND ARE CONFIDENTIAL AND PROPRIETARY PRODUCTS OF ANSYS, INC., ITS SUBSIDIARIES, OR LICENSORS. The software products and documentation are furnished by ANSYS, Inc., its subsidiaries, or affiliates under a software license agreement that contains provisions concerning non-disclosure, copying, length and nature of use, compliance with exporting laws, warranties, disclaimers, limitations of liability, and remedies, and other provisions. The software products and documentation may be used, disclosed, transferred, or copied only in accordance with the terms and conditions of that software license agreement. ANSYS, Inc. is certified to ISO 9001:2008.

U.S. Government Rights

For U.S. Government users, except as specifically granted by the ANSYS, Inc. software license agreement, the use, duplication, or disclosure by the United States Government is subject to restrictions stated in the ANSYS, Inc. software license agreement and FAR 12.212 (for non-DOD licenses).

Third-Party Software
See the legal information in the product help files for the complete Legal Notice for ANSYS proprietary software and third-party software. If you are unable to access the Legal Notice, please contact ANSYS, Inc. Published in the U.S.A.

Table of Contents
1. Introduction ........................................................................................................................................... 1 2. Files ......................................................................................................................................................... 5 2.1. Fluid-Structure Interaction (FSI) ......................................................................................................... 5 3. Meshes ................................................................................................................................................... 7 3.1. Adjacency ........................................................................................................................................ 7 3.2. Meshing Mode Access ...................................................................................................................... 8 4. Boundary Conditions ............................................................................................................................. 9 4.1. Reference Temperature from a Boundary .......................................................................................... 9 4.2. Non-Reflecting Boundary Conditions .............................................................................................. 10 5. Heat Exchangers ................................................................................................................................... 11 5.1. Improved Curve Fitting for Heat-Exchanger Model .......................................................................... 11 5.1.1. Limitations ............................................................................................................................ 11 5.1.2. Usage .................................................................................................................................... 12 5.2. Alternate Formulation for the Dual Cell Heat Exchanger .................................................................. 12 6. Turbulence ............................................................................................................................................ 13 6.1. Explicit Algebraic Reynolds Stress Model ......................................................................................... 13 6.1.1. Accessing the WJ-BSL-EARSM Model ....................................................................................... 16 6.1.2. References ............................................................................................................................. 16 6.2. Wall-Modeled LES (WMLES) S-Omega Formulation .......................................................................... 17 6.2.1. Accessing the WMLES S-Omega Model ................................................................................... 17 6.3. Near Wall Treatment for the Porous Media Interface ......................................................................... 18 6.3.1. Accessing the Turbulent Wall Treatment Option ...................................................................... 19 6.3.2. Example ................................................................................................................................. 19 6.4. Near Wall Treatment for Models .............................................................................................. 22 6.4.1. Theory ................................................................................................................................... 24 6.4.1.1. Momentum Equations ................................................................................................... 24 6.4.1.2. Turbulence Models ............................................................................................... 24 6.4.1.3. Iteration Improvements ................................................................................................. 25 6.4.2. User Interface ......................................................................................................................... 26 6.4.3. Example ................................................................................................................................. 27 6.4.4. References ............................................................................................................................. 31 6.5. Buoyancy Effects on Omega-Based Turbulence Models .................................................................... 31 6.5.1. Theory ................................................................................................................................... 32 6.5.2. User Interface ......................................................................................................................... 33 7. Combustion .......................................................................................................................................... 35 7.1. Char Burnout Kinetics (CBK) Model ................................................................................................. 35 7.1.1. References ............................................................................................................................ 36 7.2. Number of Species in Reacting Flows .............................................................................................. 36 8. Pollutants ............................................................................................................................................. 39 8.1. Coal Derived Soot .......................................................................................................................... 39 8.1.1. Using the Coal Derived Soot Model ........................................................................................ 41 8.1.1.1. References .................................................................................................................... 43 8.2. Atomic Balance for Sulfur ............................................................................................................... 43 8.3. Mercury Pollutant Formation ........................................................................................................... 43 8.3.1. Mercury Speciation in Coal Flames .......................................................................................... 43 8.3.1.1. Overview ....................................................................................................................... 44 8.3.1.2. Governing Equations for Mercury Transport ................................................................... 44 8.3.1.3. Mercury Speciation Model ............................................................................................. 46 8.3.1.3.1. One Step Mechanism ............................................................................................ 46 8.3.1.3.2. Two Step Mechanism ............................................................................................ 46
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates. iii

Beta Features Manual 8.3.1.3.3. Detailed (Wilcox) Mechanism ................................................................................ 47 8.3.1.4. Species Production Sources from Different Fuel Types .................................................... 48 8.3.1.4.1. Hg and HCl Production in a Gaseous Fuel .............................................................. 48 8.3.1.4.2. Hg and HCl Production in a Liquid Fuel .................................................................. 49 8.3.1.4.3. Hg and HCl Production from Coal .......................................................................... 49 8.3.1.4.4. Hg and HCl from Char ........................................................................................... 49 8.3.1.4.5. Hg and HCl from Volatiles ...................................................................................... 50 8.3.1.5. Species Production/Consumption due to Elementary Reactions ..................................... 50 8.3.1.6. Mercury Species Capture and Retention in Ash Residue .................................................. 50 8.3.1.7. Mercury Species Capture using Sorbent Injection ........................................................... 51 8.3.1.8. Mercury Formation in Turbulent Flows ........................................................................... 51 8.3.1.8.1. The Turbulence-Chemistry Interaction Model ........................................................ 51 8.3.1.8.2. The PDF Approach ................................................................................................ 51 8.3.1.8.3. The Mean Reaction Rate ........................................................................................ 51 8.3.1.8.4. The PDF Options ................................................................................................... 51 8.3.2. Using the Mercury Model ....................................................................................................... 52 8.3.2.1. Setting Up the One Step Model ...................................................................................... 52 8.3.2.2. Setting Up the Two Step Model ...................................................................................... 53 8.3.2.3. Setting Up the Detailed (Wilcox) Model .......................................................................... 54 8.3.2.4. Defining the Fuel Streams .............................................................................................. 55 8.3.2.5. Defining the Mercury Fuel Stream Settings ..................................................................... 57 8.3.2.6. Setting Mercury Parameters for Gaseous and Liquid Fuel Types ...................................... 58 8.3.2.7. Setting Mercury Parameters for a Solid Fuel .................................................................... 59 8.3.2.8. Setting Turbulence Parameters ...................................................................................... 60 8.3.2.9. Specifying a User-Defined Function for the Hg Rate ........................................................ 63 8.3.2.10. Defining Boundary Conditions for the Mercury Model .................................................. 63 8.3.3. Solution Strategies ................................................................................................................. 64 8.3.4. Postprocessing ....................................................................................................................... 65 8.3.5. DEFINE_HG_RATE UDF Macro ............................................................................................. 66 8.3.5.1. Description ................................................................................................................... 66 8.3.5.2. Usage ............................................................................................................................ 67 8.3.5.3. Example 1 ..................................................................................................................... 68 8.3.5.4. Hooking DEFINE_HG_RATE UDFs ................................................................................ 70 8.3.5.5. Hg Macros ..................................................................................................................... 71 8.3.6. Mercury Model Dialog Box A Quick Reference Guide .......................................................... 73 8.3.7. References ............................................................................................................................ 78 9. Acoustics ............................................................................................................................................... 79 9.1. Modal Analysis ............................................................................................................................... 79 9.1.1. Limitations ............................................................................................................................ 79 9.1.2. Modal Analysis Theory ........................................................................................................... 79 9.1.3. Using the Modal Analysis Model ............................................................................................ 80 9.1.4. Setting Model Parameters ...................................................................................................... 81 9.1.5. Postprocessing of the Modal Analysis Model .......................................................................... 81 9.1.5.1. References .................................................................................................................... 81 10. Discrete Phase .................................................................................................................................... 83 10.1. Extended Collision Stencil ............................................................................................................ 83 10.2. Tracking of Child Droplets Within the Same Time Step ................................................................... 83 10.3. Linearized Source Terms ................................................................................................................ 84 10.4. Temperature-Dependent Particle Density ...................................................................................... 85 11. Solver .................................................................................................................................................. 87 11.1. Recursive Projection Method ........................................................................................................ 87 11.2. Reduced Rank Extrapolation (RRE) Method .................................................................................... 88
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Beta Features Manual 11.2.1. References ........................................................................................................................... 91 11.3. Second Order in Time For Moving Deforming Meshes .................................................................... 91 11.4. Moving Averages for Monitors ....................................................................................................... 92 11.5. Executing Commands at a User-specified Iteration or Time Step ..................................................... 93 11.5.1. Executing a Command at a Particular Iteration ...................................................................... 93 11.5.2. Executing a Command at a Particular Time Step .................................................................... 94 12. Custom Field Functions ...................................................................................................................... 95 12.1. Postprocessing Unsteady Statistics ................................................................................................ 95 13. Turbomachinery .................................................................................................................................. 97 13.1. Pitch-Scale Model ......................................................................................................................... 97 13.2. Implicit Mixing-Plane Model .......................................................................................................... 98 14. Parallel Processing ............................................................................................................................ 101 14.1. Using Graphics Processing Units (GPUs) With the Algebraic Multigrid (AMG) Solver ...................... 101 14.2. Enhancing Parallel Performance and Convergence for the Algebraic Multigrid (AMG) Solver ......... 101 15. FLUENT in Workbench ...................................................................................................................... 103 15.1. Performing Transient Two-Way Simulations with FLUENT and ANSOFT ......................................... 103 15.2. Working with Custom Input Parameters ....................................................................................... 103 15.3. Using UDFs to Compute Output Parameters ................................................................................ 104 16. User-Defined Functions .................................................................................................................... 107 16.1. Six-DOF Motion Constraint Using UDFs ....................................................................................... 107 17. FLUENT as a Server ............................................................................................................................ 109 17.1. ANSYS Session Manager .............................................................................................................. 109 17.1.1. Using ANSYS Session Manager ............................................................................................ 109 17.1.2. Configuring ANSYS Session Manager .................................................................................. 110 17.2. FLUENT Remote Console ............................................................................................................. 110 17.2.1. Connecting to ANSYS Session Manager ............................................................................... 110 17.2.2. Concurrent Access .............................................................................................................. 111 17.3. FLUENT as a Server SDK ............................................................................................................... 112 17.3.1. IAnsysSessionManager CORBA Interface ............................................................................. 112 17.3.2. COM Connectors ................................................................................................................ 113 17.3.2.1. Interfaces .................................................................................................................. 113 17.3.3. Registering the COM Connectors ........................................................................................ 117 18. Population Balance .......................................................................................................................... 119 18.1. Coulaloglou and Tavlarides Breakage .......................................................................................... 119 18.1.1. References ......................................................................................................................... 119

Chapter 1: Introduction
This document contains information about ANSYS FLUENT 14.5 beta features, which provide options for modeling and reporting that are outside of the normal scope of ANSYS FLUENT. These features are not always accessible through the standard menus and dialog boxes, and will require the following text user interface (TUI) command to enable them: define beta-feature-access

Note
Please note that if you enable beta features in this case, use any beta features, and then disable beta features, the beta features you put into use may still be active, even though the text and graphical interfaces for these features may no longer be visible. It is therefore recommended that you save a separate copy of the case before any beta feature is activated. This will allow you to return to working on the case with only released features if you desire.

Important
Note that beta features have not been fully tested and validated. ANSYS, Inc. makes no commitment to resolve defects reported against these prototype features. However, your feedback will help us improve the overall quality of the product.

Note
Beta features are not subject to our Class 3 error reporting system. In addition, we will not guarantee that the input files using this beta feature will run successfully when the feature is finally released so you may, therefore, need to modify the input files. This Beta Features document presents information on the following topics: Meshes (p. 7) Boundary Conditions (p. 9) Heat Exchangers (p. 11) Turbulence (p. 13) Combustion (p. 35) Pollutants (p. 39) Acoustics (p. 79)

Introduction Discrete Phase (p. 83) Solver (p. 87) Custom Field Functions (p. 95) Turbomachinery (p. 97) Parallel Processing (p. 101) FLUENT in Workbench (p. 103) User-Defined Functions (p. 107) Population Balance (p. 119) The following is a list of the beta features discussed in this document, in alphabetical order: Adjacency (p. 7) Alternate Formulation for the Dual Cell Heat Exchanger (p. 12) Atomic Balance for Sulfur (p. 43) Char Burnout Kinetics (CBK) Model (p. 35) Coal Derived Soot (p. 39) Coulaloglou and Tavlarides Breakage (p. 119) Explicit Algebraic Reynolds Stress Model (p. 13) Extended Collision Stencil (p. 83) Implicit Mixing-Plane Model (p. 98) Improved Curve Fitting for Heat-Exchanger Model (p. 11) Mercury Pollutant Formation (p. 43) Meshing Mode Access (p. 8) Modal Analysis (p. 79) Near Wall Treatment for the Porous Media Interface (p. 18) Non-Reflecting Boundary Conditions (p. 10) Number of Species in Reacting Flows (p. 36) Pitch-Scale Model (p. 97) Postprocessing Unsteady Statistics (p. 95) Recursive Projection Method (p. 87) Reduced Rank Extrapolation (RRE) Method (p. 88)
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Reference Temperature from a Boundary (p. 9) Second Order in Time For Moving Deforming Meshes (p. 91) Six-DOF Motion Constraint Using UDFs (p. 107) Tracking of Child Droplets Within the Same Time Step (p. 83) Performing Transient Two-Way Simulations with FLUENT and ANSOFT (p. 103) Wall-Modeled LES (WMLES) S-Omega Formulation (p. 17) Included in the information about the beta features are references to related chapters and sections in the ANSYS FLUENT 14.5 Getting Started Guide, Users Guide, Theory Guide, UDF Manual, Fuel Cell Modules Manual, and Population Balance Module Manual.

Chapter 2: Files
2.1. Fluid-Structure Interaction (FSI)
When setting up a fluid-structure interaction problem, you can ensure that the forces mapped to the FEA mesh are conserved by performing the following steps: 1. Enable the beta feature access (as described in Introduction (p. 1)). 2. Read an FEA mesh, using either the Read button of the Volume FSI Mapping or Surface FSI Mapping dialog box, or the file/fsi/read-fsi-mesh text command. 3. Enable the conservation of the mapped forces by using the following text command: file fsi conserve-force?

Chapter 3: Meshes
3.1. Adjacency
You can determine the face zones which are adjacent to the cell zones, and display them using the Adjacency dialog box. After enabling beta feature access (Introduction (p. 1)), enter the following text command in the ANSYS FLUENT console: mesh adjacency Figure 3.1: The Adjacency Dialog Box

The Adjacency dialog box (Figure 3.1: The Adjacency Dialog Box (p. 7)) will be displayed where you can: 1. List the face zones that are adjacent to the cell zones in your case. Selecting a Cell Zone(s) will populate the Adjacent Face Zones list. 2. Display one or more adjacent face zones, using the Display Face Zones button. 3. Enable the Multiple Cell Zones option if you want to select more than one Cell Zone(s). By default, this option is enabled.

Meshes 4. Select Renaming Face Zones to allow you to rename or clean up some of the names that may cause confusion. You have several renaming options: Rename by Adjacency appends the name of the cell zone to the type of the adjacent face zone. For example, if fluid is selected in the Cell Zone(s) and interior is selected in the Adjacent Face Zones list, then renaming by adjacency produces interior-fluid. If the name is already in use, the original zone ID is appended in addition in order to create a unique name. Note that it is best if you avoid using long cell zone names.

Important
Only selected face zones will be renamed.

Rename to Default simply appends the original zone ID to the type. Enable Abbreviate Types to abbreviate the zone type. For example, vi is the velocity inlet, int is the interior, ifc is the interface. Enable Exclude Custom Names to allow for customized names that do not follow any pattern or any default names (such as default-interior) which are excluded from renaming. This is a protective measure so as not to accidentally destroy your desired naming. Disabling this option, will unconditionally rename all selected zones. 5. Enable Draw Default Mesh opens the Mesh Display dialog box, where you may choose to display mesh zones. These will be displayed permanently while others will be displayed as currently selected in the Adjacent Face Zones list. This is useful for finding your way through a new and complex mesh. 6. Specify a Face Zone Name Pattern and click Match to select face zones with names that match the specified pattern. For example, if you specify *inlet*, all face zones whose names contain inlet (e.g., velocity-inlet-5, velocity-inlet-6) will be selected automatically. If they are all selected already, they will be deselected. If you specify inlet?, all face zones whose names consist of inlet followed by a single character will be selected (or deselected, if they are all already selected). 7. The Cell Zone Types and Face Zone Types lists allow you to select the respective zones by type from the Cell Zone(s) list and the Adjacent Face Zones list, respectively.

3.2. Meshing Mode Access

For 3D serial processing, you have the ability to switch from the solution mode of FLUENT to the meshing mode at any point, even when a mesh or case file is in memory. By enabling beta feature access (Introduction (p. 1)), the following text command will always be available in the console, and can be used as described in Switching Between Meshing and Solution Modes in the Getting Started Guide : switch-to-meshing-mode

Important
This beta feature is not available for parallel ANSYS FLUENT.

Chapter 4: Boundary Conditions

4.1. Reference Temperature from a Boundary
When any fluid material inside the domain is an incompressible-ideal-gas or ideal-gas, the option of specifying the Density Method will appear as a drop-down list in the Operating Conditions dialog box. Select one of the inlet boundaries (velocity inlet, mass-flow-inlet, pressure-inlet ) for the calculation of the operating density. The temperature specified in the temperature tab of an inlet boundary dialog box will be used to calculate the operating density. If no boundary type is an Inlet, then ANSYS FLUENT will calculate the reference density using the default method.

Important
This option can be used only when you specify the temperature and/or species concentration on the boundary as constant. This option will not be available if the boundary has a profile or UDF for temperature. This option is only available for the pressure-based solver.

Specifying the inlet boundary for the calculation of reference density helps in predicting quiescent flows.

Boundary Conditions Figure 4.1: The Operating Conditions Dialog Box

After enabling beta feature access ( Introduction (p. 1) ), you can use the following text command: define operating-conditions use-inlet-temperature-for-operating-density Enter the Zone-id/name [()].

4.2. Non-Reflecting Boundary Conditions

The general non-reflecting boundary conditions (NRBC) are available for the pressure-based solver after enabling beta feature access ( Introduction (p. 1) ). The information in General Non-Reflecting Boundary Conditions in the User's Guide applies to the density-based and pressure-based solvers, with the exception that the general NRBC for the pressure-based solver is compatible with ideal gas, real gas, species transport, and mixture fraction transport (for premixed and partially premixed models).

Note
General non-reflecting boundary conditions (NRBC) are not available for the steady-state solver.

Chapter 5: Heat Exchangers

5.1. Improved Curve Fitting for Heat-Exchanger Model
In the dual cell heat exchanger model, you can specify the performance data table (either heat transfer or NTU) for calculating local heat transfer in the cell. If the operating mass flow rates fall within the range of the mass flow rates provided in the performance data table, then linear interpolation has been found to be the best method to calculate the NTU value. However, if the operating mass flow rates fall outside the range specified in the performance data table, then the NTU value corresponding to the maximum mass flow rate is taken if the operating mass flow rate is greater; otherwise the NTU value corresponding to the minimum mass flow rate is taken if the operating mass flow rate is lower. Due to this clipping of NTU values, unexpected heat transfer may occur. To avoid this, curve fitting allows you to use the exponential curve for extrapolation. You can use the following flavors of exponential decay curves for NTU versus mass flow rates. (5.1) = +

(5.2)

is the primary mass flow rate. where a,b,c,d,e,g are user-specified coefficients and "
To use Equation 5.1 (p. 11), you have to create a file named coefficient3.dat in your working directory, which contains the coefficients a,b, and c for each auxiliary mass flow rate row by row. For example, if the number of auxiliary mass flow rates is 3, then the file will read as
a1 b1 c1 a2 b2 c2 a3 b3 c3

ANSYS FLUENT will read the file coefficient3.dat and use the coefficients in Equation 5.1 (p. 11) to compute the NTU value if the primary mass flow rate is out of range. If the primary mass flow rate is within the range, the above coefficients will be ignored and linear interpolation will be used. Similarly, to use Equation 5.2 (p. 11), you have to create a file named coefficient5.dat in your working directory, which contains the coefficients a,b,c,d, and g for each of the auxiliary mass flow rates row by row. For example if the number of auxiliary mass flow rates is 3 then the file will read as
a1 b1 c1 d1 g1 a2 b2 c2 d2 g2 a3 b3 c3 d3 g3

5.1.1. Limitations
This feature can be used only for one heat exchanger since it can read only one file for coefficients. This feature is available only with the dual cell heat exchanger model (see Using the Dual Cell Heat Exchanger Model in the User's Guide). This feature cannot be used for interpolation. Linear interpolation is used for such cases. Curves of the type outlined above can only be used for extrapolation.

Heat Exchangers

5.1.2. Usage
Make sure you first enable beta feature access, as described in Introduction (p. 1). The feature can be activated by setting an rpvar as follows: For curve (1), enter (rpsetvar dc/extrapolation-method exponential3) in the console. A coefficient file named coefficient3.datwill be created in the working directory when you perform iterations. For curve (2), enter (rpsetvar dc/extrapolation-method exponential5) in the console. A coefficient file named coefficient5.dat will be created in the working directory when you perform iterations. To go back to the default extrapolation, use the following rpvar: (rpsetvar dc/extrapolation-method default)

5.2. Alternate Formulation for the Dual Cell Heat Exchanger

It is a well known fact that the dual cell model depends on the resolution of the core meshes. If the core mesh is very coarse, then accuracy is severely affected. Make sure you first enable beta feature access, as described in Introduction (p. 1), then activate the alternate formulation for heat transfer using the following text command: define models heat-exchanger dual-cell-model alternate-formulation? The results obtained using the alternate formulation is mesh independent and gives a reliable solution even on very coarse meshes. Please note that the default formulation and alternate formulation results are comparable on a sufficiently fine core mesh. Also the alternate formulation should not be used for non-matching core meshes. For background information about the dual cell heat exchanger, see Using the Dual Cell Heat Exchanger Model in the User's Guide.

Chapter 6: Turbulence
This chapter contains information relating to turbulence models implemented as beta features in ANSYS FLUENT 14.5. 6.1. Explicit Algebraic Reynolds Stress Model 6.2. Wall-Modeled LES (WMLES) S-Omega Formulation 6.3. Near Wall Treatment for the Porous Media Interface 6.4. Near Wall Treatment for Models 6.5. Buoyancy Effects on Omega-Based Turbulence Models

6.1. Explicit Algebraic Reynolds Stress Model

Explicit Algebraic Reynolds Stress Models (EARSM) represent an extension of the standard two-equation models. They are derived from the Reynolds stress transport equations and give a nonlinear relation between the Reynolds stresses and the mean strain-rate and vorticity tensors. Due to the higher order terms, many flow phenomena are included in the model without the need to solve transport equations for individual Reynolds stresses. The WJ-BSL-EARSM allows an extension of the BSL turbulence model to capture the following flow effects: Anisotropy of Reynolds stresses Secondary flows The BSL model is the basic model underlying the SST model. The BSL model is described in [4]. The implementation of the WJ-BSL-EARSM in ANSYS FLUENT is based on the explicit algebraic Reynolds stress model of Wallin and Johansson [1]. Differences from the original formulation by Wallin and Johansson are explained in the following text.

Note
The beta implementation of the WJ-BSL-EARSM mentioned in this section is still under development. This model may not yield accurate results and it is recommended that you do not use the model currently implemented in FLUENT 14.0 and 14.5. If you want to use the corrected WJ-BSL-EARSM, contact your ANSYS Customer Support representative. With EARSM, the Reynolds stresses are computed from the anisotropy tensor according to its definition:

= +
where the anisotropy tensor is searched as a solution of the following implicit algebraic matrix equation:

= +

= + !

(6.1)

Turbulence The coefficients in this matrix equation depend on the -coefficients of the pressure-strain term in the underlying Reynolds stress transport model. Their values are selected here as =1.245, =0, =1.8, =2.25. The values of , , and are the same as those used in the original model by Wallin and Johansson [1]. As for the value of , it is increased from 1.2 to 1.245 in the course of calibrating EARSM for its use together with the BSL

! model.

"#$ and % &' denote the non-dimensional strain-rate and vorticity tensors, respectively. They are defined
as:

02 0 3 + (23 = ) 1 3 12 68 6 9 4 89 = 5 7 9 78
where the time-scale

(6.2)

(6.3)

A =B C =

@ is given by: DFE DF =

(6.4)

In order to arrive at an explicit solution of the Equation 6.1 (p. 13), the anisotropy tensor is expressed as a polynomial based on the strain rate and the vorticity tensors as follows:

G TU = HYITU + H` P TV P VU QQWR TU + Ha ITV P VU P TV IVU + Hb ITV P VXP XU + P TV P VXI XU

The c-coefficients are evaluated to:

QSR TU QQWITU

(6.5)

dg = e f ht =
ip

q r q u iis

vx = w y =
where the denominator Q is:

=
The invariants, which appear in the formulation of the anisotropy tensor and the coefficients, are defined by:

=
14 Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Explicit Algebraic Reynolds Stress Model

The model representation of the anisotropy tensor Equation 6.5 (p. 14) and its coefficients polynomial contribution (the

follows

the original model by Wallin and Johansson [1] with two differences. First, the fourth order tensor

term) is neglected in Equation 6.5 (p. 14). Second,

the tensor basis is slightly changed for convenience according to Apsley and Leschziner [2]. Although the tensor basis is changed, the model remains algebraically equivalent to the original model of Wallin and Johansson. The latter change results in correspondingly changed expressions for the coefficients . In three-dimensional flows, the equation to be solved for the function is of sixth order and no explicit can be derived from solution can be derived, whereas in two-dimensional mean flows the function a cubic equation, an analytic solution of which is recommended by Wallin and Johansson [1] also for three-dimensional cases:

! =
where
)4 =

+ +

#% +

%' $

& #% #&

%' (

#% & #% #&

(6.6)
#& <

6 05

040 7

1 12

1 13 05

C 8C = 8D

C 9E

9D9 F

@ @A + @ @B

In the original model by Wallin and Johansson [1], the diffusion terms in the transport equations for and H were calculated using the effective eddy viscosity,
`aa WY = c f TU U = I S TUU PV PV

R, of

EARSM, where

Xb
e for

. The EARSM model, implemented in ANSYS FLUENT, uses the standard eddy viscosity the diffusion terms. This model change helps to avoid the problems with the asymptotic

behavior at the boundary layer edge, which were reported by Hellsten [3]. For the underlying BSL g
h model, the

standard coefficients are used.

Turbulence

6.1.1. Accessing the WJ-BSL-EARSM Model

The EARSM model is available after enabling beta feature access, as described in Introduction (p. 1).

Note
It is only available for 3D cases. The model is available in the interface when the k-omega model is selected. The WJ-BSL-EARSM option becomes available under k-omega Model, as shown in Figure 6.1: The Viscous Model Dialog Box (p. 16). Figure 6.1: The Viscous Model Dialog Box

6.1.2. References
1. Wallin, S. and Johansson A., A complete explicit algebraic Reynolds stress model for incompressible and compressible flows, Journal of Fluid Mechanics, 403, pp. 89-132, 2000. 2. Apsley, D.D. and Leschziner, M.A., A new low-reynolds-number nonlinear two-equation turbulence model for complex flows, International Journal of Heat and Fluid Flow, 19, pp. 209-222, 1998

Wall-Modeled LES (WMLES) S-Omega Formulation 3. Hellsten, A., New advanced turbulence model for high-lift aerodynamics, AIAA Paper 2004-1120, Reno, Nevada, 2004. 4. Menter, F. R., Two-Equation Eddy-Viscosity Turbulence Models for Engineering Applications. AIAA Journal. 32(8). 15981605. August 1994.

6.2. Wall-Modeled LES (WMLES) S-Omega Formulation

One of the deficiencies of the WMLES using a modified Smagorinsky model in the LES zone, is that the model does not provide zero eddy-viscosity for flows with constant shear. For this reason, the model does not allow the computation of transitional effects, and can produce overly large eddy-viscosities in separating shear layers (like from a backstep). An enhancement to the WMLES formulation given in Equation 4.261 in the Theory Guide is therefore to compute the LES-portion of the model with the use of the difference abs(S-) instead of S (S being the shear strain rate and the vorticity rate).

6.2.1. Accessing the WMLES S-Omega Model

The WMLES S-O model is available after enabling beta feature access, as described in Introduction (p. 1). The model is available in the interface when the Large Eddy Simulation (LES) model is selected. The WMLES S-O option becomes available under Subgrid-Scale Model, as shown in Figure 6.2: The Viscous Model Dialog Box (p. 17). Figure 6.2: The Viscous Model Dialog Box

You can also access the WMLES S-O subgrid-scale model via the embedded LES zone, as shown in Figure 6.3: The Fluid Dialog Box (p. 18).
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates. 17

Turbulence Figure 6.3: The Fluid Dialog Box

6.3. Near Wall Treatment for the Porous Media Interface

Two models are available in ANSYS FLUENT for the treatment of porous media: The (full) porous media model The porous jump condition The porous media model is applied in a cell zone. Several input parameters can be specified to determine the pressure loss in the flow. The porous jump condition is a 1D simplification of the porous media model and is applied to a face zone. In the porous medium the standard conservation equations for turbulence quantities are solved by default. The turbulence inside the porous medium is treated as though the solid medium has no effect on the turbulence production and dissipation rates. A detailed description of these models can be found in the ANSYS FLUENT Users Guide. An enhancement to the porous jump condition is now available as a beta feature, allowing you to enable a turbulent wall treatment at the fluid side of the interface. When you assume, for example, turbulent fluid flow over a porous medium, then the porous medium has an effect on the fluid similar to a wall at the interface depending on its porosity. This beta option has been introduced to include the effects of the porous material on the fluid side of such an interface. An Example (p. 19) shows the combination of the beta porous jump option with a porous media zone. The turbulent near wall treatment on the fluid side of a fluid/porous media interface has mainly two components: In the momentum equation, the wall shear stress is introduced at the interface.

Near Wall Treatment for the Porous Media Interface In the transport equation for the specific dissipation rate scribed.

or dissipation rate wall values are pre-

Furthermore, the diffusion term is set to zero at the interface, and the near wall distance is updated to obtain a proper wall treatment. The near wall treatment which is used at the interface behaves identical to that of a solid wall, also with respect to y+.

6.3.1. Accessing the Turbulent Wall Treatment Option

Near wall treatment for the porous media interface is available after enabling beta feature access, as described in Introduction (p. 1). The Enable turbulent wall treatment option in the Porous Jump dialog box (see Figure 6.4: The Porous Jump Dialog Box (p. 19)) is used to enable or disable a turbulent near wall treatment on the fluid side of a porous jump interface. In the Viscous Model dialog box, different near wall treatments can be selected depending on the turbulence model. The selected treatment is then applied both for solid walls and at the fluid side of the interface. The Enable turbulent wall treatment option can be used together with the k-epsilon and k-omega turbulence models, SST and Transition SST, Reynolds Stress models, Detached Eddy Simulation (DES) and Scale-Adaptive Simulation (SAS). Figure 6.4: The Porous Jump Dialog Box

6.3.2. Example
As outlined in the introduction, the purpose of the Enable turbulent wall treatment option is to include the effects of the porous material on the turbulent flow, for example at an interface between a fluid and a porous medium. Figure 6.5: Setup of Two Channel Flows Separated by Wall / Porous Jump Interface; Color Denotes Contours of the Streamwise Velocity Component. (p. 20) shows the setup of two channel flows, which are separated by a wall in the first part and by a porous jump interface in the larger second part. The Reynolds number based on the centerline velocity and channel half height in each + channel is .

The lower zone in the second part is defined as a cell zone where the porous media model has been in the activated. A viscous loss term has been specified with a small viscous resistance of

Turbulence streamwise direction. The interface between the lower porous media zone and the upper pure fluid zone has been defined as porous jump with a medium thickness of zero. Therefore, no pressure jump occurs across the interface. It is only used to enable the turbulent wall treatment at the fluid side of the interface. Two simulations have been performed with and without the turbulent wall treatment at the interface in order to investigate the influence. Inlet boundary condition profiles are prescribed for the velocity and the turbulence quantities which have been obtained from a 1D periodic channel flow. The outlet is specified as pressure outlet. The grid is refined both to the walls (y+ ~ 2) and to the interface. The SST turbulence model has been used for both simulations. Figure 6.5: Setup of Two Channel Flows Separated by Wall / Porous Jump Interface; Color Denotes Contours of the Streamwise Velocity Component.

Figure 6.6: Profiles of the Streamwise Velocity Component Near the Outlet at Position x=0.9m Without (top) and With (bottom) Near Wall Treatment at the Interface. Red Denotes the Pure Fluid Side, Black Represents the Side of the Porous Media (p. 21) shows profiles of the streamwise velocity component at position x=0.9m near the outlet without (top) and with (bottom) near wall treatment at the interface. The profiles in the pure fluid zone are colored in red. The profiles in the zone where the porous media model is activated are colored in black. The inclusion of the wall shear stress in the momentum equation has reduced the streamwise velocity at the interface on the fluid side. In the figures cell center, values have been plotted to allow a direct comparison with the effect caused by the walls. The effect of the new treatment at the interface is similar to the influence of the upper wall. The profiles of the turbulent viscosity ratio are shown in Figure 6.7: Profiles of the Turbulent Viscosity Ratio Near the Outlet at Position x=0.9m Without (top) and With (bottom) Near Wall Treatment at the
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Near Wall Treatment for the Porous Media Interface Interface. Red Denotes the Pure Fluid Side, Black Represents the Side of the Porous Media (p. 22). In this simulation, a fine grid has been used which allows you to resolve the viscous sublayer. The eddy viscosity ratio is nearly zero close to the walls. The interface shows the same effect due to the additional source term in the -equation when the near wall treatment has been enabled and reduces the turbulent viscosity ratio.

Turbulence Figure 6.7: Profiles of the Turbulent Viscosity Ratio Near the Outlet at Position x=0.9m Without (top) and With (bottom) Near Wall Treatment at the Interface. Red Denotes the Pure Fluid Side, Black Represents the Side of the Porous Media

6.4. Near Wall Treatment for Models

The near wall treatment is an essential element of any RANS turbulence model for the simulation of wall boundary layers. Previously, the near wall treatment would be either a wall function approach or

Near Wall Treatment for # $ Models a low-Reynolds number model. The wall function approach requires the cell centre of the first grid point to lie in the logarithmic layer. The low-Reynolds formulation requires an integration to the wall using + a resolution of < . Both approaches produce large errors if used outside their range of validity. + The wall function method deteriorates under mesh refinement. For meshes coarser than = , the low-Reynolds formulation results in inaccurate wall values for the wall shear stress and heat transfer. -insens+ itive. The computed wall values (shear stress and heat transfer) are largely independent of the + value provided by the grid. Any -insensitive wall treatment reverts back to its underlying lowReynolds formulation when the grid is sufficiently fine, and reverts to the wall function formulation when the grid is coarse. The lack of a suitable low-Reynolds formulation was the main obstacle in the formulation of such wall treatments for the -equation. While many low-Reynolds formulations have been developed and published, they all suffered from one of more of the following problems. The formulations: are complicated, as the formulation involved numerous highly non-linear damping terms are numerically not robust for complex applications may produce multiple (non-unique) solutions for the same application may produce pseudo-transitional results, or unphysical laminar zones. Due to these problems, the model formulation of choice in todays industrial codes for the -equation is a two-layer formulation. The two-layer formulation avoids the solution of the -equation in the viscous sublayer and overwrites it with an algebraic formulation based on a simple mixing-length model. In FLUENT the two-layer model is the basis of the Enhanced Wall Treatment (EWT), which is the + -insensitive formulation for all -equation based models. In the two-layer approach, the fluid domain is subdivided into a viscosity-affected region and a fully-turbulent region. The blending of the two regions is determined by a turbulent Reynolds number , which is defined as: In order to obtain a less sensible formulation, wall models have been developed which are
+

(6.7)

where is the density, is the wall-normal distance, is the turbulence kinetic energy, and is the molecular viscosity. The blending of the two regions is centred around:

>
!<

: fully turbulent region

! : viscosity-affected near wall region

(default value). In the fully turbulent region the transport equations for the turbulence "= kinetic energy and the dissipation rate are evaluated. In the viscosity-affected near-wall region the one-equation model of Wolfstein is employed. where

Turbulence While the EWT based on the two-layer formulation works well in most cases, using the turbulent Reynolds number defined in for the demarcation of the flow regime can have the following drawbacks: Regions far away from the wall with very low values of turbulence kinetic energy might have a turbulent Reynolds number smaller than 200, and these regions will therefore be treated with a near-wall formulation (ex. regions with a very low turbulence level). The Wolfstein model is not consistent with the -equation for non-equilibrium (pressure gradient) flows, and so the solution of the combination depends on the switching location.
+ In case the mesh is coarse with a of the first cell near the switching location, the model has a tendency to oscillate as it switches back and forth between time steps. The oscillation prevents convergence.

An alternative formulation has been developed which is not based on the two-layer approach but uses a new low-Reynolds formulation. This alternative formulation is designed to avoid the above listed deficiencies of existing low-Reynolds formulations, as well as those of the two-layer formulation.

6.4.1. Theory
+ + The goal of a insensitive near wall treatment is the -independent prediction of the wall shear stress and wall heat flux (assuming a sufficient resolution of the boundary layer). The formulation should switch gradually from wall functions to a low-Reynolds formulation when the grid is refined. This switch requires a blending of various quantities between the viscous sublayer and the logarithmic region.

6.4.1.1. Momentum Equations

The wall shear stress is needed as a boundary condition and is calculated as:
=

(6.8) and are friction velocities. Both the friction

where is the shear stress, is the density, and velocities blending for
!

and are blended between the viscous sublayer and the logarithmic region. The is:

The blending for the friction velocity, 5 6, is:

"# )01 = + % 234 ( &' $

+ 8 7@A B

234

(6.9)

89 =

+ 8 CDE

FGB

(6.10)

6.4.1.2. H I Turbulence Models

The main idea of the new near wall treatment is to add a source term to the transport equation of the dissipation rate , which accounts for near wall effects. The modified standard P Q model reads:

ST V a

W T W + c = Y d S` Xd U b

(6.11)

Near Wall Treatment for Y ` Models

+
' 1 (

# = " # +

(6.12)

= %&0 )

(6.13) , 576 = , 89 = , @A = , and BC = .

where 243 =

The additional source term D EFGHIGPP is active only in the viscous sublayer and accounts for low-Reynolds number effects. The value for this source term is zero by default in the logarithmic region. The exact formulation of this source term is at this point proprietary and therefore not given here. Adding this source term to the standard Q R model is more or less straightforward since this model has no low-Reynolds number terms in its original formulation. However, adding the source term becomes more complex for the realizable and RNG S T models since both models already have low-Reynolds terms in their formulations. In order to avoid an interaction of these low-Reynolds terms with the new near wall source term, the original low-Reynolds terms are turned off in the viscous sublayer, which allows the new source term to become active.

Note
Please keep in mind that the two-layer approach in FLUENT also does not use the lowReynolds terms of these models, since the transport equation for the dissipation rate is replaced by an algebraic relation in the near wall region.

6.4.1.3. Iteration Improvements

In combination with the new near wall treatment, the iterative treatment and linearization of the U V two-equation model has been improved. This modification is activated when the new wall treatment is used. As an example, results of an incompressible flat plate boundary layer calculation are given below in Figure 6.8: Residual history using new iterative procedure (p. 26) and Figure 6.9: Residual history using standard procedure (p. 26). The standard W X model has been used together with the new near wall treatment. The solution converges well with the improved treatment. With the standard treatment, significant oscillations in the residuals can be observed.

Turbulence Figure 6.8: Residual history using new iterative procedure

Figure 6.9: Residual history using standard procedure

6.4.2. User Interface

The new near wall treatment is available in the Viscous Model panel once the beta features have been enabled. The additional option Near wall treatment Menter-Lechner in the section Near-Wall Treat26 Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Near Wall Treatment for Models ment is shown in Figure 6.10: New option "Near wall treatment Menter-Lechner" in Viscous Model panel (p. 27). Figure 6.10: New option "Near wall treatment Menter-Lechner" in Viscous Model panel

You can enable the new near wall treatment using the following text command:
/define/models/viscous/near-wall-treatment/ke-nwt-menter-lechner? yes

The new near wall treatment can be used together with the standard, realizable, and RNG turbulence models.

6.4.3. Example
A zero pressure gradient flat plate boundary layer, demonstrated in Figure 6.11: Zero pressure gradient flat plate boundary conditions (p. 28), was used as one of the basic validation cases. The experiment was carried out by Wieghardt (1951) and was included in the AFOSR-IFP Stanford conference on turbulent flows (Coles and Hirst (1969)). Experimental data is available for velocity profiles at different locations and for the skin friction coefficient. Since no measurements of heat transfer are available for this case, the Reynolds analogy has been used to estimate the distribution of the Stanton number for the experiment.

Turbulence The length of plate is 5 [m] and the Reynolds number, based on the plate length and freestream velocity, is equal to 107. At the inlet, constant profiles for velocity and turbulence have been specified. The inlet turbulence intensity is 1% and eddy viscosity ratio is 0.2. The wall is maintained at a constant temperature with a temperature difference of

between wall and inlet.

Figure 6.11: Zero pressure gradient flat plate boundary conditions

The calculations have been performed with the standard model on six different grids. The corres+ ponding values are given in Table 1.
+ Table 6.1: y values for the different grids at position x 4.7 [m]

grid 1 2 3 4 5 6

+
0.09 0.4 2.2 5.8 10.6 21.4

The distribution of the skin friction coefficient and the Stanton number along the wall is shown in Figure 6.12: Skin friction coefficient (p. 29) and Figure 6.13: Stanton number (p. 30). The results are in agreement with the experimental values and show only a small variation for the different grids. Velocity profiles in wall units for different grids at position x 4.7 [m] are present in Figure 6.14: Velocity profiles in wall units for different grids at position x 4.7 [m] (p. 31). The velocity profiles show a correct behavior in the logarithmic region.

Near Wall Treatment for Models Figure 6.12: Skin friction coefficient

Turbulence Figure 6.13: Stanton number

Buoyancy Effects on Omega-Based Turbulence Models Figure 6.14: Velocity profiles in wall units for different grids at position x 4.7 [m]

6.4.4. References
Bibliography
[1] D.E. Coles and E.A. Hirst. "Computation of turbulent boundary layers". 1968 AFOSR-IFP-Stanford Conference. Stanford University, CA. 1968. [2] K Wieghardt and W. Tillmann. On the turbulent friction layer for rising pressure". Technical Memorandum 1314. National Advisory Committee for Aeronautics. 1951.

6.5. Buoyancy Effects on Omega-Based Turbulence Models

Until now, the effects of buoyancy on turbulence have only been included in turbulence models based on the transport equation of the dissipation rate . The details of the formulation are described in Effects of Buoyancy on Turbulence in the k- Models in the FLUENT Theory Guide. Using a beta feature in ANSYS FLUENT 14.5 you can account for buoyancy effects for turbulence models based on the transport equation of specific dissipation rate . The formulation of buoyancy terms in the -equation is derived from the corresponding terms in the transport equations of the turbulence kinetic energy and the dissipation rate .

Turbulence

6.5.1. Theory
The standard --model that includes the buoyancy term is written as:

+ = + =
where !#" = , $&% =

+ + + + + +
, '( = , )0 = , and 253 =

(6.14)

1 . 4

For details on modeling of turbulence generation due to buoyancy (6 7), see Effects of Buoyancy on Turbulence in the k- Models in the FLUENT Theory Guide. The buoyancy term in the 8-equation is derived from the 9 and @ equations using the following relations:

A = B CD E FG EH

(6.15)

=I P

E FQ EH

+I Q

E FP EH

(6.16)

This derivation leads to the following transformation of the buoyancy source terms: S S RVW = UYX U`X R W R W T T

(6.17)

The first part of the buoyancy term abc in the d equation comes from the transport equation of the dissipation rate. The second part of the term comes from the turbulence kinetic energy equation. The model coefficient, egf , is replaced with + h , where i is the corresponding coefficient of the production term in the p-equation. In the SST model, this coefficient is a linear combination of the corresponding coefficients of the q-r and the transformed s -t models. For the u -v model, the value of w is 0.44. The value for xy in the standard - model is recovered from this value of . The SST model with buoyancy terms is written as:

+ = + =

+ + d + e d g + + d + fe f + h f ig
(6.18)

The coefficients, j, for the SST model are functions of kl: no mp m = nomo +

(6.19)

Buoyancy Effects on Omega-Based Turbulence Models Where, and stand for coefficients of the - and - models respectively. Without low Reynolds number corrections, the coefficients have the following values:

= = = =
=

! " #$ = %&'( = )0 = 12 =
+ 7 8BA4 @ 4 @
(6.20)

5 49@ = 6

The final formulation of the buoyancy source terms for the 3 transport equations are:

The C-equation buoyancy formulation term, DEF, can be applied to the following turbulence models: Standard and SST G-H models Transition SST

I-based differential Reynolds stress models

Explicit Algebraic Reynolds Stress models (Beta) SST-DES and SAS The value of the coefficient P depends on the turbulence model being implemented. The buoyancy term in the transport equation for Reynolds stresses is identical to the term which appears in the Q-based differential Reynolds stress models.

6.5.2. User Interface

The new buoyancy option for turbulence models based on transport equations of the specific dissipation rate (R) can be accessed in the viscous model panel once the beta features have been enabled. When you select the k-omega (2 eqn) option under the Viscous Model panel, you can select one of the following options: No Buoyancy Effects on Turbulence

Turbulence This option is the default setting which will not enable the buoyancy effects on turbulence. Buoyancy Effects on Turbulence Production Enabling this option will include the buoyancy term in the transport equation for the turbulence kinetic energy or Reynolds stresses. Enabling this option also will also include the second term in Equation 6.20 (p. 33) for the transport equation.

Full Buoyancy Effects on Turbulence This option activates the full buoyancy model. This option also includes the first term in Equation 6.20 (p. 33) which represents the contribution from the -equation.

You can also enable the new buoyancy option using the following text command:
define/models/viscous/kw-buoyancy-effects?

Chapter 7: Combustion
7.1. Char Burnout Kinetics (CBK) Model
The Char Burnout Kinetics (CBK) model describes char oxidation under conditions relevant to pulverized coal combustion processes. It includes effects of thermal annealing and ash inhibition on the char combustion. The form of this model that is available in ANSYS FLUENT only applies to atmospheric conditions and has no statistical kinetics. The CBK model can be defined as a material property in the Create/Edit Materials dialog box ( Figure 7.1: The Create/Edit Materials Dialog Box with the CBK Model Selected (p. 35) ) for problems in which you have defined discrete-phase injections. Materials Edit... Figure 7.1: The Create/Edit Materials Dialog Box with the CBK Model Selected

Combustion Once you have selected combusting-particle from the Material Type drop-down list, you can select cbk from the Combustion Model drop-down list in the Properties group box. The New CBK-8 Combustion Model dialog will open, allowing you to set the char reactivity parameters (see Figure 7.2: The New CBK-8 Combustion Model Dialog Box (p. 36) ). The default values are acceptable. Figure 7.2: The New CBK-8 Combustion Model Dialog Box

7.1.1. References
1. FORTRAN program CBK8. Brown University, Providence, 1998. 2. R. Hurt, J. K. Sun, and M. Lunden. A Kinetic Model of Carbon Burnout in Pulverized Coal Combustion. Combustion and Flame, 113:181-197, 1998. 3. R. E. Mitchell, R. Hurt, L. L. Baxter, and D. R. Hardesty. Compilation of Sandia Coal Char Combustion Data and Kinetic Analyses: Milestone Report. Technical Report SAND92-8208, Sandia National Laboratory, Livermore, CA, 1992. 4. J. K. Sun and R. Hurt. Mechanics of Extinction and Near-Extinction in Pulverized Solid Fuel Combustion. Proceedings of the Combustion Institutes, 28:2205-2213, 2000.

7.2. Number of Species in Reacting Flows

After enabling beta feature access (Introduction (p. 1)), the maximum number of species for the Species Transport model is increased from 50 to 100. DPM may be used with more than 50 species with the following restrictions: No evaporating, devolatizing, oxidizing or product species may have an index greater than 49 If used in conjunction with the NOx model, the SNCR reagent specie from a liquid injection may not have an index above 49 The following models and features may not be used with more than 50 species: Density-based solver Eulerian PDF Transport model Melting/solidification model Crevice model
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Number of Species in Reacting Flows Thermal diffusivity model Surface species Site species

Chapter 8: Pollutants
This chapter contains information relating to the pollution models implemented as beta features in ANSYS FLUENT 14.5. 8.1. Coal Derived Soot 8.2. Atomic Balance for Sulfur 8.3. Mercury Pollutant Formation

8.1. Coal Derived Soot

The coal derived soot beta feature is an extension to the Moss-Brookes soot model, which is discussed in The Moss-Brookes Model in the Theory Guide and Setting Up the Moss-Brookes Model and the Hall Extension in the User's Guide. The present implementation accounts for coal-derived soot, based on the work of Brown [1]. This extension includes an additional transport equation for the tar evolved during coal devolatilization. The Moss-Brookes model assumes that the physical and thermodynamic properties of tar are similar to those of volatiles, such that the combined effect of volatile and tar on the gas phase flame simulation may be replaced by a single volatile stream (consisting of volatile and tar). In reality, however, this may not be the case. Therefore, it is important that you realize that the shortcomings of the above assumption may have an impact on the coal-derived soot model. If the tar in flame is modeled using a secondary devolatilization model, then it is advised that you use the tar stratum thus solved instead of using the simplified assumption. The following set of paths was assumed for the coal-derived soot formation [2] (see Figure 8.1: Presumed Path for Coal-Derived Soot (p. 39)). Figure 8.1: Presumed Path for Coal-Derived Soot

Nucleation is assumed to be the first step in formation of soot in most light gas flames, and acetylene is understood to be the major species involved. In heavier gas flames, benzene and other polycyclic aromatic hydrocarbons (PAHs) may contribute to soot formation as well. Soot formation in coal flames is thought to occur when tars or the higher molecular weight hydrocarbons given off during devolatilization combine and condense to form soot particles. This is a different mechanism to that of soot formation from gaseous fuels. The related source term for each path is given as follows:

=

(8.1)

Pollutants

= "E ! 6789@A = $%&'()0%1GBBH 2334%'5&()0%1 6789@A " 6BCD #F I GBBH UV

(8.2) (8.3)

xy = d efghihjfkhlm tuv = notuvp w xyz {|}~| =

where P QRST is equal to particles, WX is Avogadros Number, and Y` d abbc is the molecular weight of the soot particle. The remaining terms in the previous overall source expressions are defined as follows: efghipqfr uvw = stuvw (8.4)

qw xyz rs

(8.5) (8.6) (8.7)

(8.8)

The soot oxidation term () is similar to that shown in Equation 14.139 in the Theory Guide.

and the collision constant is set to 3.0. (= 12 kg/kgmol) is the molecular weight of carbon and (= 1.3806503e-23 J/K) is the Boltzmann constant. An incipient soot particle is assumed to consist of 9e+04 carbon atoms, thus making = 108e+04 kg/kgmol. is the normalized radical nuclei concentration (i.e., the number of particles /kg). Since the
is assumed to be 1950 kg/ coal-derived soot particles are large, the turbulent Schmidt number used in the transport equations for soot mass fraction and the normalized number density must be modified to account for the particle size. A value of 700 for the turbulent Schmidt number is suggested for soot mass fraction and nuclei transport. The term is the tar release rate from coal (kg/ -s) and comes from the coal particle source computations of the discrete phase model. It is assumed that the mass fraction of tar in coal volatiles is in the range 0.30.5, and therefore the term is related to the volatile source term via the tar mass fraction in volatiles. One of the main assumptions of this implementation is that tar may be decoupled from the flow field computations, since tar is assumed to be a known fraction of volatiles and volatile transport is fully coupled with the flow field. In addition, it is assumed that the amount of soot formed is too small to affect the major species concentration. However, there are some suggestions that soot formed in a coal flame may be significant enough to couple the major species concentration with the soot formation and oxidation. This, however, is not included in the present implementation. The values used for the pre-exponential constant and the activation energy in Equation 8.5 (p. 40) Equation 8.7 (p. 40) are listed in Table 8.1: Rate Constants for Coal-Derived Soot (p. 40) Table 8.1: Rate Constants for Coal-Derived Soot Term

in the Moss-Brookes soot model theory (Fenimore-Jones or Lee oxidation model). Soot density ( )

(kJ/kgmol)
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Coal Derived Soot

!"#

6.77e+05 ( /kg-s) 9.77e+10 (1/s) 5.02e+08 (1/s)

52,300 286,900 198,900

$%&'()0%1 2334

8.1.1. Using the Coal Derived Soot Model

The steps that follow describe how to use the coal-derived soot extension of the Moss-Brookes model. For details about the theory and equations related to this extension, see Soot Model Theory in the Theory Guide

Important
Make sure you first enable beta feature access, as described in Introduction (p. 1) . 1. (optional) Set up and solve a combustion simulation that solves for the tar species. 2. Enable and set up the discrete phase model, using the Discrete Phase Model dialog box. Make sure that the injections have coal or heavy oil specified as the injection material. See Modeling Discrete Phase in the User's Guide. 3. Enable the soot model and set up the definitions and parameters using the following text commands: define models soot? When asked to Enable the soot model?, respond by entering yes 4. Make sure you have enabled beta feature access, as described in Introduction (p. 1). 5. Enable the coal-derived soot extension using the following text command: define models soot-parameters soot-model-parameters You can use the settings for most of the prompts that follow this text command, as they reflect the settings you made in the Soot Model dialog box and the default Moss-Brookes model settings. The prompts that relate to the coal-derived soot extension are the following: a. Coal-derived soot? Enter yes to enable the coal-derived soot extension. b. Solve tar equation? Enter yes if you need to solve for the tar species. Enter no if you have already run a combustion simulation that calculated the tar species. Your answer will affect the inputs to the define/models/soot-parameters/soot-process-parameterstext command (as described below). c. Collision Constant Enter a value for the collision constant. The recommended value is 3.

Pollutants 6. Set the process parameters for the coal-derived soot extension using the following text command: define models soot-parameters soot-process-parameters Enter 1080000 kg/kgmol for the Mass of incipient soot particle, as this represents 9 atoms. A value of 1950 kg/

carbon

is recommended for the Mean density of soot particle.

You can also perform the above steps in the Soot Model dialog box. Select Moss-Brookes in the Model list and set up the definitions and parameters. See Using the Soot Models in the User's Guide. The prompts that relate to the coal-derived soot extension will vary, depending on whether you have already run a combustion simulation that calculated the tar species. If you requested that the tar equations be solved using the define/models/soot-parameters/soot-model-parameters text command, then the prompts that relate to the coal-derived soot extension are the following: a. Number of tar streams Enter the number of tar streams for the model. This value will depend on the number of different coal or heavy oil injections defined using the Injections dialog box. b. Mass fraction of tar in coal volatiles For each tar stream, enter the mass fraction of tar in the coal volatiles. It is recommended that this value be between 0.3 and 0.5. c. Species name For each tar stream, enter the name of the fuel species of the associated volatile stream. The tar evolution will then be calculated as a fraction of the volatiles that evolve from this fuel species. d. Remove fuel species from list? If you made a mistake when entering the number of tar streams or the species name, you can remove the erroneous species from the calculation by entering yes. If you requested that the tar equations not be solved using the define/models/soot-parameters/soot-model-parameters text command, then the prompt that relates to the coal-derived soot extension is the following: Tar species name Enter the name of the tar species from your previous combustion model solution. 7. Specify the desired turbulent Schmidt number for the soot mass fraction and nuclei transport, using the following text command: define models soot-parameters modify-schmidt-number? The suggested value is 700.

Mercury Pollutant Formation

8.1.1.1. References
1. A. L. Brown, Modeling Soot in Pulverized Coal Flames, MSc thesis, Brigham Young University, Utah, USA, 1997. 2. J. Ma, Soot Formation and Soot Secondary Reactions During Coal Pyrolysis, PhD thesis, Brigham Young University, Utah, USA, 1996.

8.2. Atomic Balance for Sulfur

In order to verify the pollutant sulfur species balance, i.e. inflow, outflow and source from coal combustion etc, you can use the sulfur atomic balance calculation. The feature may be enabled using the following text command (after you have enabled beta feature access, as described in Introduction (p. 1)): define models sox-parameters s-atom-balance? Once enabled, during each of the postprocessing iterations of the SOx pollutant computation, the S species atomic balance will be computed and displayed in the console.

Important
You are advised to enable this option only for the final few iterations of the SOx model computations to save cpu time. Note that you may not see a proper S balance during the initial stages of the SOx model solution.

8.3. Mercury Pollutant Formation

This chapter discusses the theory, usability, and the UDFs used in ANSYS FLUENT for modeling Mercury formation. Information is presented in the following sections: Mercury Speciation in Coal Flames (p. 43) Using the Mercury Model (p. 52) DEFINE_HG_RATE UDF Macro (p. 66) Mercury Model Dialog Box A Quick Reference Guide (p. 73)

8.3.1. Mercury Speciation in Coal Flames

The following sections present the theoretical background of mercury speciation in coal flames. For information about using the mercury model in ANSYS FLUENT, see Using the Mercury Model (p. 52) . Overview (p. 44) Governing Equations for Mercury Transport (p. 44) Mercury Speciation Model (p. 46) Species Production Sources from Different Fuel Types (p. 48)

Pollutants Species Production/Consumption due to Elementary Reactions (p. 50) Mercury Species Capture and Retention in Ash Residue (p. 50) Mercury Species Capture using Sorbent Injection (p. 51) Mercury Formation in Turbulent Flows (p. 51)

8.3.1.1. Overview
The principal anthropogenic source of mercury in the atmosphere is generated from the coal-fired utility boilers. Due to its potential toxicity, environmental persistence and potential for bioaccumulation, mercury is a particularly insidious and difficult pollutant to manage. Due to the above mentioned reasons, there is a great concern over the emission of heavy metal mercury into the atmosphere and as a result there is an impetus to investigate the reduction of mercury emission to the environment from power plant utility boilers. Mercury released due to combustion of coal or waste co-utilization, may be captured using sorbent injections as a post combustion reduction technique (in water, is highly soluble while HgO has is insoluble). It is also important to note that part of the mercury released only a low solubility and may be retained in the ash residue. Therefore, a general mathematical tool must be able to predict both the mercury retention in the ash residue as well as gas-phase mercury speciation accurately (e.g. mercury chlorination between Hg and HCl released from fuel). Due to the very low ppm levels of mercury present in the fuels, usual post processing technique is applicable to the mathematical modeling of mercury retention and speciation. Dajnak and Lockwood [1] reported a model for mercury speciation and retention in sub-micron particles applicable to coal-fired power plants. However, their mercury speciation model was limited to Hg + HCl single step reaction only. More emphasis was given to the mercury retention modeling in ash residues. Madsen et al. [2] have reported on a detailed study of computational fluid dynamic (CFD) modeling of mercury capture using sorbent injection method using ANSYS FLUENT CFD code. However, they have only considered the capture process of mercury in the post-combustion region and did not model the mercury speciation. The mercury concentration of various coal types is available through the U.S. Geological Surveys (USGS) COALQUAL database [3].

8.3.1.2. Governing Equations for Mercury Transport

ANSYS FLUENT solves the mass transport equations for the Hg species, taking into account convection, diffusion, production and consumption of Hg and related species. This approach is completely general, being derived from the fundamental principle of mass conservation. The effect of residence time in mercury mechanisms, a Lagrangian reference frame concept, is included through the convection terms in the governing equations written in the Eulerian reference frame. If all mercury released from fuel is assumed to convert directly to Hg and the other product and intermediate species are assumed negligible, then only the Hg species transport equation is needed:

(8.9)

As will be discussed in Mercury Speciation Model (p. 46), mercury formation mechanisms are more involved. Hence, the tracking of mercury-containing intermediate species is also important. ANSYS FLUENT solves transport equations for , and species in addition to the Hg species:

#$%&

Mercury Pollutant Formation

+ + +
8

(8.10) (8.11) (8.12)

= =

+ "&'() + %&'()

!"&'()

! # "&'()

where 012, 3456 7 , 9@ABC and DEFG are mass fractions of HI, PQR ST , UVW and XY`a in the gas phase. The source term
bcd, efgh i , qrstu p

and vwxy are to be determined depending on the form of fuel

mercury release (Hg is released from coal during devolatilization) and inclusion of , and in the mercury mechanism.

It is important to note that in the gas-phase, mercury undergoes a set of chlorination reactions (mercury oxidation). Hence, in addition to the set of transport equations for the mercury containing species, further set of chlorine containing species needs to be modeled. Following set of transport equations for chlorine containing species are required to complete the mercury chlorination mechanism.

e m v

+ + +

=
fi

d + d

(8.13) (8.14) (8.15) (8.16)

fg kl

f hg kl

= =

gkl + jkl os tu + rs tu
wz

no s tu

n po s tu

wx|}~

w y x|}~

x|}~ + {|}~

where , , and are mass fractions of , , and in the gas phase. The source term , , and are to be determined depending on the form of fuel chlorine

release ( is assumed to release from coal during devolatilization, but further flexibility is given to the user to partition as well) and inclusion of , and in the mercury chlorination mechanism. The source term in each transport equation may be expressed in terms of the source from fuel/char combustion (or volatile release), source/sink from production (or consumption) rate of the given species due to each gas phase reaction the species participates and source due to retention or capture of mercury species into the particulate or ash residues, as follows: = + + (8.17)

= source of i ( Cl,...

), where i = Hg, HgCl2,..., HCl,

= source of i (

) from fuel, where i = Hg, HCl

or Cl

= source of i (

) from gas phase reactions,

where i = Hg, HgCl2,..., HCl, Cl,...

Pollutants

= source of i ( ) due to capture into solid

particles or retention in ash residues, where i = Hg, HgCl2,..., HCl, Cl,... Each respective source term will be discussed further in detail in Species Production Sources from Different Fuel Types (p. 48) and Species Production/Consumption due to Elementary Reactions (p. 50).

8.3.1.3. Mercury Speciation Model

The rate of elemental mercury (Hg) release during coal particle devolatilization may be assumed proportional to the rate of devolatilization. Chlorine (in the form of HCl) may also be released in a similar manner (i.e. at a rate proportional to the rate of devolatilization). It is generally accepted that chlorine is released to the gas-phase mainly in the form of HCl [1]. Cl radical formation then occurs via the decomposition of HCl and , where is formed from those original radicals of HCl. Dajnak and Lockwood [1] reported a single step mercury chlorination model proposed by Hall [4] while Wilcox [5] proposed a complete set of elementary reaction mechanism for mercury speciation, involving 12 species and 13 reactions. In this version of ANSYS FLUENT, you have access to both models. All reactions considered here are assumed to have Arrhenius type rate constants of the form:

$ ! " #

(8.18)

8.3.1.3.1. One Step Mechanism

The single step reaction for mercury chlorination estimated by Hall et al. [4] and reported by Dajnak and Lockwood [1], considered only %&, '()01 and 234.

56 + 578 56789 + 59

(8.19)

Rate constant for the above reaction were reported as pre-exponential factor, @= A + BCD GBHE G F = and activation energy E = 18 kJ/mol. Gasper et al. [9] reported a slightly different set of rate constants for the same reaction, in which I= P + QRS VQWT V U = and E = 28.77 kJ/mol, respectively. For the present implementation, Hall [4] constants were used.

8.3.1.3.2. Two Step Mechanism

In addition to the oxidized product, XY` ab , mercury may be oxidized into the product HgO as well. In order to accommodate the second oxidized species, a two step model is proposed by combining the Hall [4] global reaction (Equation 8.19 (p. 46)) with the following oxidation reaction. cd + e + f cde + f (8.20) Here, the rate constants for the above reaction (Equation 8.20 (p. 46)) were as proposed in the Wilcox model (see Table 8.2: Reaction Rate Constant for Mercury Speciation [1] (p. 47), reaction R7). While the Wilcox model may need considerably larger number of iterations to converge a solution, two-step model was found to give comparable accuracy for predicted ghi pq and rst concentrations with much less CPU penalty.

Mercury Pollutant Formation

8.3.1.3.3. Detailed (Wilcox) Mechanism

The following set of reactions were proposed by Wilcox [5] to model the gas-phase mercury speciation. In addition to the mercury related reactions, chlorine also undergoes a set of reactions on its own and must be considered together with the mercury chlorination reactions. Figure 8.2: Mercury Chlorination Pathway (p. 47) and Figure 8.3: Chlorine Only Reaction Subset (p. 47) show the pathways of mercury chlorination and HCl/Chlorine reaction set. Figure 8.2: Mercury Chlorination Pathway

Figure 8.3: Chlorine Only Reaction Subset

, !, "#$%, &', (, )0 and 1). All reactions considered here are assumed to be reversible. Rate
Table 8.2: Reaction Rate Constant for Mercury Speciation [1] Reaction Number R1 R2 R3 R4 R5 R6 R7 R8 Reaction Forward A (cm3molsec) Forward b Forward E (kcal/mol) 716.13 30.27 93.3 43.3 76.08 0 8.8 36.6 Reverse A (cm3molsec) 2.55e12 7.23e12 6.87e11

Wilcox [5] mercury chlorination/oxidation model consists of 12 species (

, , , , ,

constants for the complete set of mercury reaction mechanism are given in Table 8.2: Reaction Rate Constant for Mercury Speciation [1] (p. 47) and Table 8.3: Chlorine Specific Reactions [2] (p. 48).

ReReverse verse b E (kcal/mol) 0 0 0 13.8 11.8 6.2

2345 + 6 = 23 + 45 + 6 4.25e13 789@ + 79@ = 789@A + 7 4.50e13 1.93e13 BC + BDE = BCDE + B 6.15e13 FG + HIP = FGHI + HI QRSTV + U = QRST + ST + U 1.35e8 WXY` + Y`a = WXY`a + Y` 1.8115e10 3.09e9 bcd + e = bc + d + e fg + fhip = fgip + hf 3.06e13

0 0 0 0 0 0 0 0

Pollutants Reaction Number R9 R10 Reaction Forward A (cm3molsec) Forward b Forward E (kcal/mol) 0.5 87.0 Reverse A (cm3molsec) ReReverse verse b E (kcal/mol) -

+ = + 1.827e10 + = + + 3.19e11

0 0

[1] Rate constants for some reactions are different to those reported in Wilcox [5]. New constants for R1 are from Wilcox et. al [6], R2 from Wilcox et. al [7], R3, R4 and R8 from Wilcox [8], R6 updated with new values (A = 1.43e9 3.48e10, average value used), R7 updated using RRKM (to be published), R9 updated with new values (A = 1.74e9 3.48e10, average value used). Note: Details about RRKM may be obtained at http://en.wikipedia.org/wiki/RRKM_theory. Table 8.3: Chlorine Specific Reactions [2] Reaction Number R11 R12 R13 R14 Reaction Forward A (cm3molsec) Forward b Forward E (kcal/mol) 55.84 3.16 1.17 0.0 Reverse A (cm3molsec) 2.23e14 1.45e12 1.0e17 4.4e13 Reverse b Reverse E (kcal/mol) -1.8 2.46 47.49 81.67

+ = + 8.51e15 + = + 1.47e12 " + ! = ! + 8.59e13 #$ + % + & = %#$ + & 7.19e21

0 1.63 0 -2

0.0 1.44 0.0 0.0

[2] Rate constants were taken from NIST data base (except forward rate of R14, which was obtained using RRKM).

8.3.1.4. Species Production Sources from Different Fuel Types

Here, it is assumed that only Hg and HCl are released into gas phase due to combustion of fuel or char in coal or due to coal volatile release. Hence, for other species, the respective source term, , is set to zero.

'( 0 )

8.3.1.4.1. Hg and HCl Production in a Gaseous Fuel

2 3 4 15 D 6 = 78 9 D 8@AB CD 6 4CD 9 QR T S a bc
48

The rate of Hg or HCl production is proportional to the rate of combustion of the gaseous fuel: (8.21)

source of i ( or Cl

EFG IH P), where i = Hg, HCl

= mean limiting reaction rate of fuel ( )

UVW YX `

Mercury Pollutant Formation

= mass fraction of j" in fuel, where j = Hg or Cl

8.3.1.4.2. Hg and HCl Production in a Liquid Fuel

The rate of Hg or HCl production is proportional to the rate of fuel release into the gas phase through droplet evaporation:

(8.22)

( ) ! # " = source of i ($%& ' ), where i = Hg, HCl or Cl 01234 = rate of fuel release from the liquid droplets to the gas 8 (567 )

9@ E ABCD = mass fraction of j" in fuel, where j = Hg or Cl F = cell volume (GH)

8.3.1.4.3. Hg and HCl Production from Coal
For coal, it is assumed that both mercury and chlorine are distributed between volatiles and char. Since there is no reason to assume that mercury (or chlorine) is equally distributed between the volatiles and char, the fraction of mercury (or chlorine) in the volatiles and char should be specified separately.

IP Y Q = IRSTU Y Q + IVWX Y Q

Combined source of Hg and HCl may be expressed as (8.23)

`a c b = source of i (def hg i), where i = Hg, HCl or Cl pqrst v u = source of i (wxy ) from char, where i = Hg, HCl or
Cl

= source of i ( ) from volatiles, where i = Hg, HCl

or Cl

8.3.1.4.4. Hg and HCl from Char

The source of Hg or HCl from the char is related to the rate of char combustion:

def j g = h j def i j g i j h

(8.24)

kl = char burnout rate (mno p) qr w stuv = mass fraction of j" in char, where j = Hg
or Cl

= cell volume (y )

Pollutants

8.3.1.4.5. Hg and HCl from Volatiles

The source of Hg or HCl from the volatiles is related to the rate of volatile release:

(8.25)

= volatile release (devolatilization) rate from coal

particles into the gas phase ( Cl

)

$ !"# = mass fraction of j" in volatiles, where j = Hg or % = cell volume (&' )

8.3.1.5. Species Production/Consumption due to Elementary Reactions

In addition to the production sources from combusting fuel to the Hg and HCl, each species considered in the mercury chlorination and oxidation reaction mechanism will have further source/sink term from each participating reaction, as described in Mercury Speciation Model (p. 46), depending on the respective speciation mechanism used. The source/sink term of the ( variable (1 2) of the 3

45 reaction may be expressed as follows:

)0 species in terms of the rate-of-progress

6@ D A = 7 BC = E8 A D B9 B

(8.26)

where F G I H = F G I H FG I H and J is the total number of reactions involved. The stoichiometric coefficients P Q S R are related by the general form of J elementary reactions (reversible or irreversible) involving I

T bc = eUb f dV b = T bc = eU b f dV b WXY` =

chemical species.

a hi

(8.27)

The rate-of-progress variable for the g reaction is given by the difference of the forward and reverse rates as q y tr vw = s v x (8.28) p t = qu tr vw = s v x Where

is the molar concentration of the species and and are the forward and reverse d rate constants of the reaction.

8.3.1.6. Mercury Species Capture and Retention in Ash Residue

Mercury containing species such as ef, ghijk , lmn and opqr may be retained by the ash residue in coal combustion applications or be captured by sorbent injection. Instead of proposing a global model for adsorption due to sorbent injection, which is usually dependent upon the type of sorbent stream used (e.g. activated carbon), a user defined function hook will be provided to incorporate the sink term due to mercury capture. On the other hand, retention of mercury species in the ash residue

Mercury Pollutant Formation (for coal applications) may be modeled following the work of Dajnak and Lockwood [1]. However, they have not provided the model constants in their publication.

8.3.1.7. Mercury Species Capture using Sorbent Injection

You may implement your own sorbent injection and capture model using the ANSYS FLUENT UDF utility. Please refer to the UDF Manual for further details.

8.3.1.8. Mercury Formation in Turbulent Flows

The kinetic mechanisms of Mercury formation and destruction are obtained from laboratory experiments in a similar fashion to the NOx model. In any practical combustion system, however, the flow is highly turbulent. The turbulent mixing process results in temporal fluctuations in temperature and species concentration that will influence the characteristics of the flame. The relationships among Mercury formation rate, temperature, and species concentration are highly nonlinear. Hence, if time-averaged composition and temperature are employed in any model to predict the mean Mercury formation rate, significant errors will result. Temperature and composition fluctuations must be taken into account by considering the probability density functions which describe the time variation.

8.3.1.8.1. The Turbulence-Chemistry Interaction Model

In turbulent combustion calculations, ANSYS FLUENT solves the density-weighted time-averaged NavierStokes equations for temperature, velocity, and species concentrations or mean mixture fraction and variance. To calculate Mercury species concentration, a time-averaged Mercury species formation rate must be computed at each point in the domain using the averaged flow-field information.

8.3.1.8.2. The PDF Approach

The PDF method has proven very useful in the theoretical description of turbulent flow [10]. In the ANSYS FLUENT Mercury model, a single- or joint-variable PDF in terms of a normalized temperature, species mass fraction, or the combination of both is used to predict the Mercury emission. If the nonpremixed combustion model is used to model combustion, then a one- or two-variable PDF in terms of mixture fraction(s) is also available. The mean values of the independent variables needed for the PDF construction are obtained from the solution of the transport equations.

8.3.1.8.3. The Mean Reaction Rate

The mean turbulent reaction rate described in NOx Formation in Turbulent Flows for the NOx model also applies to the Mercury model. The PDF is used for weighting against the instantaneous rates of production of Mercury species and subsequent integration over suitable ranges to obtain the mean turbulent reaction rate as described in Equation 14.105 and Equation 14.106 for NOx.

8.3.1.8.4. The PDF Options

As is the case with the NOx model, P can be calculated as either a two-moment beta function or as a clipped Gaussian function, as appropriate for combustion calculations [12], [11]. Equation 14.108 to Equation 14.112 apply to the Mercury model as well, with the variance computed by solving a transport equation during the combustion calculation stage, using Equation 14.113 or Equation 14.114.

Pollutants

8.3.2. Using the Mercury Model

Mercury species concentrations, generally in parts per million (ppm) levels, generated in combustion have minimal influence on the predicted flow field, temperature, and major combustion product concentrations. The most efficient way to use the Mercury model is as a postprocessor to the main combustion calculation. The procedure for activating and setting up the model for a decoupled solution is as follows: 1. Calculate your combustion problem using ANSYS FLUENT.

Important
The premixed combustion model is not compatible with the Mercury model.

2. Enable the Mercury model, define the fuel streams, and set the appropriate parameters, as described in this section. Models Mercury Edit...

3. Define the boundary conditions for all Mercury and Chlorine species at flow inlets. Boundary Conditions 4. In the Equations dialog box, turn off the solution of all variables except Mercury and Chlorine species. Solution Controls Equations... 5. Perform calculations until convergence (i.e., until the Mercury and Chlorine species residuals are below ) to ensure that the Mercury and Chlorine concentration fields are no longer evolving. Run Calculation 6. Review the mass fractions of Mercury and Chlorine species with alphanumerics and/or graphics tools in the usual way. 7. Save a new set of case and data files, if desired. File Write Case & Data...

8.3.2.1. Setting Up the One Step Model

You can enable and set up the One Step Mercury formation model by using the Mercury Model dialog box (Figure 8.4: The Mercury Model Dialog Box for the One Step Model (p. 53)). Models Mercury Edit...

Mercury Pollutant Formation Figure 8.4: The Mercury Model Dialog Box for the One Step Model

Under Model, select One Step. The dialog box will expand to show the appropriate inputs. Next, you must define the Fuel Streams data and Turbulence Interaction Mode data as described in detail later in this document.

8.3.2.2. Setting Up the Two Step Model

You can enable and set up the Two Step Mercury formation model by using the Mercury Model dialog box (Figure 8.5: The Mercury Model Dialog Box for the Two Step Model (p. 54)). Models Mercury Edit...

Pollutants Figure 8.5: The Mercury Model Dialog Box for the Two Step Model

Under Model, select Two Step. The dialog box will expand to show the appropriate inputs. Next, you must define the Fuel Streams data and Turbulence Interaction Mode data and Model Parameters (i.e. [O] model) as described in detail later in this document.

8.3.2.3. Setting Up the Detailed (Wilcox) Model

You can enable and set up the Detailed (Wilcox) Mercury formation model by using the Mercury Model dialog box (Figure 8.6: The Mercury Model Dialog Box for the Detailed (Wilcox) Model (p. 55)). Models Mercury Edit...

Mercury Pollutant Formation Figure 8.6: The Mercury Model Dialog Box for the Detailed (Wilcox) Model

Under Model, select Detailed. The dialog box will expand to show the appropriate inputs. Next, you must define the Fuel Streams data, Turbulence Interaction Mode data, and Model Parameters (i.e. [O] and [OH] models) as described in detail later in this document.

8.3.2.4. Defining the Fuel Streams

ANSYS FLUENT allows you to define multiple fuel streams when you are modeling Mercury formation, as shown in the following steps: 1. Specify the Number of Fuel Streams in the Fuel Streams group box. You are allowed up to three separate fuel streams. 2. Define the first fuel stream.

a. Select the fuel stream to be defined by using the arrow keys of the Fuel Stream ID text box. b. When the non-premixed combustion model is not enabled, select the fuel species from the Fuel Species list. You cannot select more than 5 fuel species for each fuel stream, and the total number of fuel species selected for all the fuel streams combined cannot exceed 10. c. When the non-premixed combustion model is enabled (Figure 8.7: The Mercury Model Dialog Box with Non-Premixed Combustion (p. 57)), make a selection from the PDF Stream drop-down list to
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates. 55

Pollutants define the species for this stream. You can select either the primary or secondary fuel stream species, as defined in the PDF table. d. Specify the parameters for this particular fuel stream in the Fuel Stream Settings group box. See Defining the Fuel Streams 3. Repeat steps 2.(a)-2.(c) for each additional fuel stream. 4. Set the model parameters that apply to all of the fuel streams in the Model Parameters group box (only required by the Two Step and Detailed (Wilcox) models):

Specify the method by which O and OH (for the Detailed model only) will be calculated. The Mercury model routines employ three methods for reduction calculations of Mercury: You can select equilibrium, partial-equilibrium, or instantaneous in the [O] Model drop-down list. You can select none, partial-equilibrium, or instantaneous in the [OH] Model drop-down list.

Important
To use the predicted O and/or OH concentration, select instantaneous in the [O] Model or [OH] Model drop-down list.

Mercury Pollutant Formation Figure 8.7: The Mercury Model Dialog Box with Non-Premixed Combustion

Note that the following limitations apply when you are modeling Mercury formation with multiple fuel streams, if more than one fuel stream has the same fuel type (as defined in the Fuel Type list in the Fuel Stream Settings group box): For multiple liquid fuel streams or multiple solid (coal) fuel streams, the injectors associated with the fuel streams should have different destination species, as defined in the Devolatilizing Species drop-down list in the Set Injection Properties dialog box (see Defining Injection Properties for details). The Mercury calculations will be erroneous if the destination species are the same. For multiple solid (coal) fuel streams, the fuel streams should have the same char-related parameter values in the Fuel Stream Settings group box, i.e., the Char HG and CL Mass Fraction and the Partition Fractions (for char HCL) values. Note that even if different values are set for these char-related parameters, ANSYS FLUENT will only recognize those specified for the solid fuel stream with the lowest ID number, and then apply them to all of the other solid fuel streams. For more information about the limitations associated with multiple fuel streams with the same fuel type, contact your ANSYS FLUENT support engineer.

8.3.2.5. Defining the Mercury Fuel Stream Settings

When using the Mercury model, you must set the parameters in the Fuel Stream Settings group box for each fuel stream specified in the Fuel Stream ID text box.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates. 57

Pollutants To begin, specify the fuel type in the following manner:

To calculate Mercury formation from a solid fuel, select Solid under Fuel Type. To calculate Mercury formation from a liquid fuel, select Liquid under Fuel Type. To calculate Mercury formation from a gaseous fuel, select Gas under Fuel Type. Figure 8.8: The Mercury Model Dialog Box Displaying Liquid Fuel Parameters

Note that you can use only one of the fuel types for a given fuel stream. The Gas option is available only when the Species Transport model is enabled (see Enabling Species Transport and Reactions and Choosing the Mixture Material).

8.3.2.6. Setting Mercury Parameters for Gaseous and Liquid Fuel Types
If you have selected Gas or Liquid as the Fuel Type, you will also need to specify the following: Set the correct mass fraction of Hg in the fuel (kg Hg per kg fuel) in the Fuel HG Mass Fraction field. Select the Chlorine intermediate species (hcl, cl, or hcl/cl) in the CL intermediate drop-down list. Set the correct mass fraction of Cl in the fuel (kg Cl per kg fuel) in the Fuel CL Mass Fraction field.

Mercury Pollutant Formation If you selected hcl/cl as the Chlorine intermediate, you will need to set the fraction of the converted fuel chlorine, by mass, that will become HCl under HCL Partition Fraction. The fraction of fuel chlorine that will become Cl will be calculated by the remainder. Note that setting a partition fraction of 0 for HCl is equivalent to assuming that all fuel chlorine is converted to the final product Cl.

Important
Note that there is a limitation that must be considered when defining more than one liquid fuel stream. See Using the NOx Model for details.

8.3.2.7. Setting Mercury Parameters for a Solid Fuel

For solid fuel, several inputs are required for the Mercury model. Figure 8.9: The Mercury Model Dialog Box Displaying Solid Fuel Parameters

Specify the mass fraction of Hg in the volatiles in the Volatile HG Mass Fraction field. Specify the mass fraction of Hg in the char in the Char HG Mass Fraction field. Select the Chlorine intermediate species (hcl, cl, or hcl/cl) in the CL Intermediate drop-down list.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Pollutants Set the correct mass fraction of Cl in the volatiles (kg Cl per kg volatiles) in the Volatile CL Mass Fraction field. If you selected hcl/cl as the Chlorine intermediate, you will need to set the fraction of the converted fuel chlorine, by mass, that will become HCl under HCL Partition Fraction. The fraction of fuel chlorine that will become Cl will be calculated by the remainder. Select the char chlorine conversion path from the Char CL Conversion drop-down list as hcl, cl, or hcl/c. Specify the mass fraction of chlorine in the char in the Char CL Mass Fraction field. If you selected hcl/cl from the Char CL Conversion drop-down list, you will need to specify the fraction of the converted char chlorine, by mass, that will become HCl under Char HCL Partition Fraction. The fraction of char chlorine that will become Cl will be calculated by the remainder.

Important
Note that there are limitations that must be considered when defining more than one solid fuel stream. See Using the NOx Model for details.

The equations, described in Using the SOx Model (Equation 22.8 to Equation 22.13), are used in a similar manner to determine the mass fraction of mercury or chlorine in the volatiles and char.

8.3.2.8. Setting Turbulence Parameters

If you want to take into account turbulent fluctuations when you compute the specified Mercury formation, define the turbulence parameters in the Turbulence Interaction Mode group box.

Mercury Pollutant Formation Figure 8.10: The Mercury Model Dialog Box for a Gas Fuel Type with Turbulence

Select one of the options in the PDF Mode drop-down list: Select temperature to take into account fluctuations of temperature. Select temperature/species to take into account fluctuations of temperature and mass fraction of the species selected in the Species drop-down list (which appears when you select this option). (non-premixed and partially premixed combustion calculations only) Select mixture fraction to take into account fluctuation in the mixture fraction(s).

Important
When modeling the formation of other pollutants along with Mercury, you should compare the selections made in the PDF Mode drop-down lists in the Turbulence Interaction Mode tab of the NOx Model dialog box and the Turbulence Interaction Mode group boxes of

Pollutants the SOx Model and Soot Model dialog boxes. If mixture fraction is selected in any of these dialog boxes, then it must be selected in all of the others as well. The mixture fraction option is available only if you are using either the non-premixed or partially premixed combustion model to model the reacting system. If you use the mixture fraction option, the instantaneous temperatures and species concentrations are taken from the PDF look-up table as a function of mixture fraction and enthalpy and the instantaneous Mercury and Chlorine rates are calculated at each cell. The PDF used for convoluting the instantaneous Mercury or Chlorine rates is the same as the one used to compute the mean flow-field properties. For example, for single-mixture fraction models the beta PDF is used, and for two-mixture fraction models, the beta or the double delta PDF can be used. The PDF for mixture fraction is calculated from the values of mean mixture fraction and variance at each cell, and the instantaneous Mercury and Chlorine rates are convoluted with the mixture fraction PDF to yield the mean rates in turbulent flow. If you selected temperature or temperature/species for the PDF Mode, you should define the following parameters in the Turbulence Interaction Mode group box: PDF Type allows you to specify the shape of the PDF, which is then integrated to obtain mean rates for the temperature and (if you selected temperature/species for the PDF Mode) the species. If you select beta, the PDF will be modeled using Equation 14.108 in the FLUENT Theory Guide of the Theory Guide. If you select gaussian, the PDF will be modeled using Equation 14.111 in the FLUENT Theory Guide of the Theory Guide. PDF Points allows you to specify the number of points used to integrate the beta or Gaussian function in Equation 14.105 in the FLUENT Theory Guide or Equation 14.106 in the FLUENT Theory Guide of the Theory Guide on a histogram basis. The default value of 10 will yield an accurate solution with reasonable computation time. Increasing this value may improve accuracy, but will also increase the computation time. Temperature Variance allows you to specify the form of transport equation that is solved to calculate the temperature variance. The default selection is algebraic, which is an approximate form of the transport equation (see Equation 14.114 in the FLUENT Theory Guide of the Theory Guide). You have the option of selecting transported to instead solve Equation 14.113 in the FLUENT Theory Guide of the Theory Guide. Though the transported form is more exact, it is also more expensive computationally. Tmax Option provides various options for determining the maximum limit(s) for the integration of the PDF used to calculate the temperature: The default selection is global-tmax, which sets the limit as the maximum temperature in the flow field. You can select local-tmax if you would rather obtain cell-based maximum temperature limits by multiplying the local cell mean temperature by the value entered in Tmax Factor. You can select specified-tmax to set the limit for each cell to be the value entered in Tmax. If you have selected a user-defined function from the Hg Rate drop-down menu in the User-Defined Functions group box, then you can select user-defined so that the limit is specified by a UDF. See the separate UDF Manual for details about user-defined functions.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Mercury Pollutant Formation Species only appears if you have selected temperature/species for the PDF Mode. Your selection in this dropdown menu determines which species mass fraction is included in the Mercury formation calculations.

Important
Note that the species variance will always be calculated using the algebraic form of the transport equation (Equation 14.114 in the FLUENT Theory Guide in the separate Theory Guide).

8.3.2.9. Specifying a User-Defined Function for the Hg Rate

You can choose to specify a user-defined function for the rate of Mercury production. By default, the rate returned from the UDF is added to the rate returned from the standard Mercury production options. You also have the option of replacing ANSYS FLUENTs Mercury rate calculations with your own userdefined Mercury rate. In addition to or instead of using the UDF to specify the Mercury rate, you can use it to specify custom values for the maximum limit () that is used for the integration of the temperature PDF (when temperature is accounted for in the turbulence interaction modeling). To use a UDF to add a rate to ANSYS FLUENTs Mercury rate calculations, you must compile and load the desired function, and then select it from the Hg Rate drop-down list in the User-Defined Functions group box. After you have selected the UDF, you have the following options: You can specify that your custom rate is added to the ANSYS FLUENT Mercury rate calculations, by retaining the default selection of Add to the ANSYS FLUENT Rate in the UDF Rate group box. You can replace the ANSYS FLUENT Mercury rate calculations with your custom rate, by selecting Replace ANSYS FLUENT Rate in the UDF Rate group box. You can specify custom values for , by selecting user-defined from the Tmax Option drop-down list in the Turbulence Interaction Mode group box. See the separate UDF Manual for details about user-defined functions.

8.3.2.10. Defining Boundary Conditions for the Mercury Model

At flow inlet boundaries, you will need to specify the Pollutant HG Mass Fraction, and if necessary, the Pollutant HGCL2 Mass Fraction, Pollutant HCL Mass Fraction, Pollutant HGO Mass Fraction, Pollutant CL Mass Fraction, Pollutant CL2 Mass Fraction, Pollutant HGCL Mass Fraction and Pollutant HOCL Mass Fraction in the Species tab, as demonstrated in Figure 8.11: The Mass-Flow Inlet Dialog Box and the Species Tab (Detailed Mercury Model) (p. 64). Boundary Conditions You can retain the default inlet values of zero for these quantities or you can input nonzero numbers as appropriate for your combustion system.

Pollutants Figure 8.11: The Mass-Flow Inlet Dialog Box and the Species Tab (Detailed Mercury Model)

8.3.3. Solution Strategies

To solve for Mercury products: (optional) If the discrete phase model (DPM) is activated (by turning on the Interaction with Continuous Phase) to run with the Mercury model, then set the Number of Continuous Phase Iterations per DPM Iteration to 0 such that no DPM iterations are performed as the Mercury case is being solved. In the Equations dialog box, turn off the solution of all variables except Pollutant hg, Pollutant hgcl2 and Pollutant hcl (Pollutant hgo, Pollutant cl, Pollutant cl2, Pollutant hgcl, and Pollutant hocl, if applicable). Solution Controls Equations... Also in the Solution Controls task page, set suitable values for the pollutant HG, HGCL2, HCL (and HGO, CL, CL2, HGCL and HOCL, if applicable) under-relaxation factors. A value of 0.95 is suggested, although
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Mercury Pollutant Formation lower values may be required for certain problems. That is, if convergence cannot be obtained, try a lower under-relaxation value. Solution Controls Under Spatial Discretization, in the Solution Methods task page select the desired scheme for each of the pollutants, HG, HGCL2, HCL (and HGO, CL, CL2, HGCL, and HOCL, if applicable) Solution Methods In the Residual Monitors dialog box, decrease the convergence criterion for the pollutants, HG, HGCL2, . HCL (and HGO, CL, CL2, HGCL, and HOCL, if applicable) to Monitors Residuals Edit...

Perform calculations until convergence (i.e., until the HG, HGCL2, HCL (and HGO, CL, CL2, HGCL, and HOCL, ) to ensure that the HG, HGCL2, HCL (and CL, CL2, HGCL, if applicable) pollutant residuals are below HGO, and HOCL, if applicable) concentration fields are no longer evolving. Review the mass fractions of pollutants, HG, HGCL2, HCL (and HGO, CL, CL2, HGCL, and HOCL, if applicable) alphanumerics and/or graphics tools.

8.3.4. Postprocessing
When you compute Mercury formation, the following additional variables will be available for postprocessing. They are contained in the Hg... category of the variable selection drop-down list that appears in postprocessing dialog boxes. Mass fraction of pollutant hg Mass fraction of pollutant hgcl2 Mass fraction of pollutant hcl Mass fraction of pollutant hgo Mass fraction of pollutant cl Mass fraction of pollutant cl2 Mass fraction of pollutant hgcl Mass fraction of pollutant hocl Mole fraction of pollutant hg Mole fraction of pollutant hgcl2 Mole fraction of pollutant hcl Mole fraction of pollutant hgo Mole fraction of pollutant cl
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Pollutants Mole fraction of pollutant cl2 Mole fraction of pollutant hgcl Mole fraction of pollutant hocl hg Density hgcl2 Density hcl Density hgo Density cl Density cl2 Density hgcl Density hocl Density Rate of hg Rate of hgcl2 Rate of hcl Rate of hgo Rate of cl Rate of cl2 Rate of hgcl Rate of hocl

8.3.5. DEFINE_HG_RATE UDF Macro

8.3.5.1. Description
You can use DEFINE_HG_RATE to specify a custom Mercury rate that can either replace the internallycalculated Mercury rate in the source term equation, or be added to the ANSYS FLUENT rate. Example 1 demonstrates this use of DEFINE_HG_RATE. The default functionality is to add user-defined rates to the ANSYS FLUENT-calculated rates. If the Replace with UDF Rate option is enabled in the Mercury Model dialog box, then the ANSYS FLUENT-calculated rate will not be used and it will instead be replaced by the Mercury rate you have defined in your UDF. When you hook a Mercury rate UDF to the graphical interface without checking the Replace with UDF Rate box, then the user-defined Mercury rate will be added to the internally-calculated rate for the source term calculation. DEFINE_HG_RATE may also be used to calculate the upper limit for the integration of the temperature PDF (when temperature is accounted for in the turbulence interaction modeling). You can calculate a custom maximum limit () for each cell and then assign it to the POLLUT_CTMAX(Pollut_Par) macro (see Hg Mac-

Mercury Pollutant Formation ros (p. 71) of the UDF Manual for further details about data access macros). Example 1 demonstrates this use of DEFINE_HG_RATE.

Important
If you want to use DEFINE_HG_RATE only for the purpose of specifying , be sure that the user-defined Mercury rate does not alter the internally-calculated rate for the source term calculation.

8.3.5.2. Usage
DEFINE_HG_RATE (name,c,t,Pollut,Pollut_Par,Hg) Argument Type symbol name cell_t c Thread *t Pollut_Cell*Pollut Pollut_Parameter *Pollut_Par Hg_Parameter *Hg Function returns void There are six arguments to DEFINE_HG_RATE: name, c, t, Pollut, Pollut_Par, and Hg. You will supply name, the name of the UDF. c, t, Pollut, Pollut_Par, and Hg are variables that are passed by the ANSYS FLUENT solver to your function. A DEFINE_HG_RATE function does not output a value. The calculated HG rates (or other mercury pollutant species rates) are returned through the Pollut structure as the forward rate POLLUT_FRATE(Pollut) and reverse rate POLLUT_RRATE(Pollut), respectively. Description UDF name. Cell index. Pointer to cell thread on which the Mercury rate is to be applied. Pointer to the data structure that contains the common data at each cell. Pointer to the data structure that contains auxiliary data. Pointer to the data structure that contains data specific to the Mercury model.

Important
The data contained within the Hg structure is specific only to the Mercury model. Alternatively, the Pollut structure contains data at each cell that is useful for all pollutant species (e.g., forward and reverse rates, gas phase temperature, density). The Pollut_Par structure contains auxiliary data common for all pollutant species (e.g. equation solved, universal gas constant, species molecular weights). Note that molecular weights extracted from the Pollut_Par structure (i.e., Pollut_Par sp[IDX(i)].mw for pollutant speciesHG, HCL, etc.and Pollut_Par sp[i].mw for other species, such as O2) has units of kg/kg-mol.

Pollutants

8.3.5.3. Example 1
The following compiled UDF, named user_hg, computes the rate for formation according to the reaction given in Equation 8.29 (p. 68). Note that this UDF will replace the ANSYS FLUENT rate only if you select the Replace with UDF Rate option in the Mercury Model dialog box. Otherwise, the rate computed in the UDF will be added to ANSYS FLUENTs default rate. See Hooking DEFINE_HG_RATE UDFs (p. 70) for details. It is assumed that the release of Hg and Cl from fuel is proportional to the rate of release of volatiles and all chlorine is in the form of HCl when released to the gas phase. The irreversible reaction for is given below:

+ + =

(8.29)

with forward rate of reaction

(Gasper et al., 1997) in the Arrhenius form:

"#$$%& '! ()0 8 12345637 and 9@A H BCADEFAG respectively) are

Here, all constants are given in SI units. The molar rates of release of Hg and Cl from volatiles ( given by:

dwx pqrsturv PQR P S ITU g VWXY`aXb = hiiicpqrsturv ey wxf mh p fghijkhl o = deeefghijkhl np mh qrstuvwtx is the rate of release of volatiles in kg/sec, yz{ |}~~ is the mass fraction of mercury species in volatiles, is the mass fraction of chlorine species in volatiles, and are the molecular weights of mercury and chlorine in kg/kg-mol, and is the cell volume in .
where See Hg Macros (p. 71) of the UDF Manual for details about the Hg macros (e.g., POLLUT_EQN, MOLECON, ARRH) that are used in pollutant rate calculations in this UDF.
/********************************************************************* UDF example of User-Defined Hg Rate: compute the rates of mercury formation according to the rate constants provided by Gasper et al. (1997). Hg + 2 HCl --> HgCl2 + H2 A = 22.0e+03, b = 0.0, E = 28770.0 (SI units) * * Arguments: * char hg_func_name - UDF name * cell_t c - Cell index * Thread *t - Pointer to cell thread on * which the Hg rate is to be * applied * Pollut_Cell *Pollut - pointer to Pollut structure * Pollut_Parameter *Pollut_Par - pointer to * Pollut_Par structure * Hg_Parameter *Hg - pointer to Hg structure * Description of Pollut_Par->pollut_io_pdf: 1. Pollut_Par->pollut_io_pdf == IN_PDF

Mercury Pollutant Formation

All rate terms which are subjected to turbulent fluctuations must be included within this flag. 2. Pollut_Par->pollut_io_pdf == OUT_PDF All rate terms which must be outside of the pdf integration (are not affected by turbulence) must be included within this flag. e.g. Char contributions to pollutant formation. 3. Pollut_Par->pollut_io_pdf == SET_VAR To modify a parameter value, user may use this flag. e.g. Top temperature setting for pdf integration. 4. Pollut_Par->pollut_io_pdf == GET_VAR This flag may be used to obtain a mean rate of reaction or any other cell based value at the end of the source term computation. Pollut_Par->nfstreams : Number of fuel streams Pollut->r_fuel_gls[i] : rate of volatile release for stream "i" per unit volume in kg/m3-sec Hg->Yhg_fuelvolat[i] : mass fraction of Hg in fuel/vol stream "i" Hg->Yhcl_fuelvolat[i] : mass fraction of HCl in fuel/vol stream "i" Hg->Ycl_fuelvolat[i] : mass fraction of Cl in fuel/vol stream "i" Hg->Yhg_char[i] : mass fraction of Hg in char stream "i" Hg->Yhcl_char[i] : mass fraction of HCl in char stream "i" Hg->Ycl_char[i] : mass fraction of Cl in char stream "i" *********************************************************************/ #include "udf.h" DEFINE_HG_RATE(user_hg, c, t, Pollut, Pollut_Par, Hg) { int ifstream; real rf=0., rr = 0., kf1=0.; /*Rate_Const KF_HG = {1.204409e4, 0.0, 18000.0};*/ /* Hall (1991)*/ Rate_Const KF_HG = {2.2e4, 0.0, 28770.0}; /* Gasper et al. (1997)*/ POLLUT_FRATE(Pollut) = 0.0; POLLUT_RRATE(Pollut) = 0.0; kf1 = ARRH(Pollut, KF_HG); switch (Pollut_Par->pollut_io_pdf) { case IN_PDF: /* Include source terms other than those from char */ switch (POLLUT_EQN(Pollut_Par)) { case EQ_HG: /* Hg production */ for(ifstream=0; ifstream<Pollut_Par->nfstreams; ifstream++) { rf += Pollut->r_fuel_gls[ifstream]*Hg->Yhg_fuelvolat[ifstream] *1000./Pollut_Par->sp[IDX(HG)].mw; } rr = -kf1*MOLECON(Pollut, IDX(HG))*MOLECON(Pollut, IDX(HCL)); break; case EQ_HGCL2: rf = kf1*MOLECON(Pollut, IDX(HG))*MOLECON(Pollut, IDX(HCL)); break; case EQ_HCL: for(ifstream=0; ifstream<Pollut_Par->nfstreams; ifstream++) { rf +=Pollut->r_fuel_gls[ifstream]*Hg->Yhcl_fuelvolat[ifstream] *1000./Pollut_Par->sp[IDX(CL)].mw; } rr = -2.*kf1*MOLECON(Pollut, IDX(HG))*MOLECON(Pollut, IDX(HCL)); break; default: break; } break; case OUT_PDF: /* Char Contributions are not included in PDF integral */ switch (POLLUT_EQN(Pollut_Par)) { case EQ_HG: for(ifstream=0; ifstream<Pollut_Par->nfstreams; ifstream++) { if (Pollut_Par->pollut_type[ifstream] == FUEL_S) { rf += Pollut->r_char[ifstream]*Hg->Yhg_char[ifstream]*

Pollutants
1000./Pollut_Par->sp[IDX(HG)].mw; break; /*char stream cannot be split at present*/ } } break; case EQ_HCL: for(ifstream=0; ifstream<Pollut_Par->nfstreams; ifstream++) { if (Pollut_Par->pollut_type[ifstream] == FUEL_S) { if (Hg->char_cl_conv[ifstream] == 0 || Hg->char_cl_conv[ifstream] == 2) { rf += Pollut->r_char[ifstream]*Hg->Yhcl_char[ifstream]* 1000./Pollut_Par->sp[IDX(CL)].mw; break; /*char stream cannot be split at present*/ } } } break; default: break; } break; case SET_VAR: /* Set a value at each cell before Hg rate computations */ break; case GET_VAR: /* Get values at the end of Hg computations */ break; default: /* Not used */ break; } POLLUT_FRATE(Pollut) = rf; POLLUT_RRATE(Pollut) = rr; }

8.3.5.4. Hooking DEFINE_HG_RATE UDFs

After you have interpreted or compiled your DEFINE_HG_Rate UDF, the name of the function you supplied as a DEFINE macro argument will become visible and selectable in the Mercury Model dialog box (Figure 8.12: The Mercury Model Dialog Box (p. 71)) in ANSYS FLUENT. To hook the UDF to ANSYS FLUENT, open the Mercury Model dialog box (Figure 8.12: The Mercury Model Dialog Box (p. 71)). Models Mercury Edit...

Mercury Pollutant Formation Figure 8.12: The Mercury Model Dialog Box

Select the function name (e.g., user_hg::libudf) from the Hg Rate drop-down list in the Mercury Model dialog box, and click OK. See DEFINE_HG_RATE UDF Macro (p. 66) for details about DEFINE_HG_RATE functions.

8.3.5.5. Hg Macros
The following macros can be used in Mercury model UDFs in the calculation of pollutant rates. These macros are defined in the header file sg_pollut.h, which is included in udf.h. They can be used to return real Hg variables in SI units and are available in both the pressure-based and the densitybased solver. See DEFINE_HG_RATE UDF Macro (p. 66) for examples of DEFINE_HG_RATE UDFs that utilize these macros. Table 8.4: Macros for Mercury UDFs Defined in sg_pollut.h Macro POLLUT_EQN(Pollut_Par) MOLECON(Pollut,SPE) NULLIDX(Pollut_Par,SPE) Returns index of pollutant equation being solved (see below) molar concentration of species specified by SPE (see below) TRUE if the species specified by SPE does not exist in a ANSYS FLUENT case (i.e., in the Species dialog box)
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Pollutants Macro ARRH(Pollut,K) ARRH_TR(Pollut,K, Tref) POLLUT_FRATE(Pollut) POLLUT_RRATE(Pollut) POLLUT_FLUCTDEN(Pollut) POLLUT_FLUCTTEM(Pollut) POLLUT_FLUCTYI(Pollut,SPE) POLLUT_CTMAX(Pollut_Par) Returns Arrhenius rate calculated from the constants specified by K (see below) Arrhenius rate calculated from the constants specified by K and Tref (see below) production rate of the pollutant species being solved reduction rate of the pollutant species being solved fluctuating density value (or, if no PDF model is used, mean density at a given cell) fluctuating temperature value (or, if no PDF model is used, mean temperature at a given cell) fluctuating mass fraction value (or, if no PDF model is used, mean mass fraction at a given cell) of the species given by index SPE upper limit for the temperature PDF integration (see below)

Important
Pollut_Par is a pointer to the Pollut_Parameter data structure that contains auxiliary data common to all pollutant species and Hg is a pointer to the Hg_Parameter data structure that contains data specific to the Mercury model. POLLUT_EQN(Pollut_Par) returns the index of the pollutant equation currently being solved. The , etc. indices are EQ_HG for Hg and EQ_HGCL2 for

MOLECON(Pollut, SPE) returns the molar concentration of a species specified by SPE. SPE is either the name of the species or IDX(i) when the species is a pollutant (like Hg). For example, for molar concentration you should call MOLECON(Pollut, O2), whereas for Hg molar concentration the call should be MOLECON(Pollut, IDX(HG)).

ARRH(Pollut,K) returns the Arrhenius rate calculated from the constants specified by K. K is defined using the Rate_Const data type and has three elements - , , and . The Arrhenius rate is given in the form of

= where is the temperature. Note that the units of must be in m-gmol-J-s.
ARRH_TR(Pollut,K,Tref) returns the Arrhenius rate calculated from the constants specified by K and Tref. K is defined using the Rate_Const data type and has three elements - , , and . Tref is of data type real and is the reference temperature in Kelvin (setting Tref = 1 degenerates the ARRH_TR to ARRH). The Arrhenius rate is given in the form of

# = $ % %&'( 2')0
72

Mercury Pollutant Formation POLLUT_CTMAX(Pollut_Par) can be used to modify the value used as the upper limit for the integration of the temperature PDF (when temperature is accounted for in the turbulence interaction modeling). You must put this macro under the SET_VAR condition within the UDF and must make sure not to put this macro under any other conditions within the UDF (e.g., IN_PDF, OUT_PDF or GET_VAR). DEFINE_POLLUT_RATE UDF conditions (where POLLUT can be either NOX, SOX or HG): IN_PDF is used to specify rates that are included within the PDF integration loop for turbulence interaction. OUT_PDF is used to specify rates that are not allowed to fluctuate (e.g. sources due to char combustion). SET_VAR is used to set or modify a common pollutant model parameter or cell dependant value (e.g. upper limit of the temperature integral). This condition will be checked at each computational cell before fluctuating rate computations. GET_VAR is used to obtain computed values (e.g. species rates due to each individual reaction) at each computational cell at the end of pollutant rate computations.

8.3.6. Mercury Model Dialog Box A Quick Reference Guide

The Mercury Model dialog box allows you to set parameters related to the mercury model. See Using the Mercury Model (p. 52) for details about the items below.

Controls

Pollutants Model specifies which model should be used for mercury speciation.

Off disables the calculation of mercury formation. One-Step enables the one-step mercury model described in One Step Mechanism (p. 46). Two-Step enables the two-step mercury model described in Two Step Mechanism (p. 46). Detailed enables the detailed (Wilcox) model described in Detailed (Wilcox) Mechanism (p. 47). Fuel Streams allows you to define multiple fuel streams for mercury formation. Number of Fuel Streams sets the number of fuel streams. You are allowed up to three fuel streams. Fuel Stream ID specifies the fuel stream you are defining in the PDF Stream drop-down list or the Fuel Species selection list. PDF Stream specifies the PDF stream species associated with a particular Fuel Stream ID, when calculating mercury formation in conjunction with the non-premixed combustion model. You can select either the primary or secondary fuel streams, as defined in the PDF table. Fuel Species is a list containing all of the defined species. Specify the species that is associated with a particular Fuel Stream ID. You cannot select more than 5 fuel species for each fuel stream, and the total number of fuel species selected for all the fuel streams combined cannot exceed 10. Turbulence Interaction Mode contains inputs that specify how turbulent fluctuations are accounted for in the mercury formation calculations. For further details on these inputs, see Setting Turbulence Parameters (p. 60).

PDF Mode is a drop-down list that contains the options for addressing turbulent fluctuations in the mercury rate calculations. Note that mixture fraction is the most accurate option, and should be used if it is available. none specifies the use of laminar mercury rate calculations, so that the effects of turbulence are ignored. temperature specifies that the mercury rate calculations include the effect of temperature fluctuations.

Mercury Pollutant Formation temperature/species specifies that the mercury rate calculations include the effect of fluctuations of temperature, as well as fluctuations of the mass fraction of the species selected in the Species drop-down list (which appears when you select this option). mixture fraction (non-premixed and partially premixed combustion calculations only) is the most accurate option, specifying that the mercury rate calculations include the effect of fluctuations of mixture fraction(s). Note that this option is not available if you are using the eddy-dissipation model. PDF Type allows you to specify the shape of the PDF. beta models the PDF using Equation 14.108 in the FLUENT Theory Guide in the Theory Guide. gaussian models the PDF using Equation 14.111 in the FLUENT Theory Guide in the Theory Guide. PDF Points controls the number of points at which the beta function will be integrated on a histogram basis. Increasing this number may improve accuracy, but will also increase compute time. This field is only available when temperature or temperature/species is selected from the PDF Mode drop-down list. Temperature Variance allows you to specify the form of transport equation that is solved to calculate the temperature variance. algebraic is an approximate form of the transport equation (see Equation 14.114 in the FLUENT Theory Guide in the Theory Guide). transported solves Equation 14.113 in the FLUENT Theory Guide in the Theory Guide. Tmax Option provides various options for determining the maximum limit(s) for the integration of the PDF used to calculate the temperature. global-tmax sets the limit as the maximum temperature in the flow field. local-tmax-factor yields cell-based maximum temperature limits by multiplying the local cell mean temperature by the value entered in Tmax Factor. specified-tmax sets the limit for each cell to be the value entered in Tmax. user-defined allows you to hook a user-defined function that specifies custom values for the maximum limit (), which is used for the integration of the temperature PDF. This option is only available if you have already compiled a UDF.
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Pollutants Species is a drop-down list which appears when temperature/species is selected from the PDF Mode dropdown list. Here you will select the species whose mass fraction fluctuations will be factored into the mercury rate calculations. User-Defined Functions allows you to use a user-defined function (UDF) to contribute to the rate of mercury production. See the separate UDF Manual for details. Note that you may also use a UDF to specify custom values for the maximum limit () that is used for the integration of the temperature PDF (when temperature is accounted for in the turbulence interaction modeling).

Hg Rate allows you to select a compiled UDF. UDF Rate provides options for the treatment of the mercury production specified by the UDF. Replace FLUENT Rate replaces ANSYS FLUENTs mercury rate calculations with the custom mercury rate produced by your UDF. Add to FLUENT Rate adds the custom mercury rate produced by your UDF to ANSYS FLUENTs mercury rate calculations. Fuel Stream Settings contains the parameters associated with a particular fuel stream of the mercury model, as specified in the Fuel Stream ID text box in the Fuel Streams group box.

Fuel Type enables selection of the fuel. Solid enables the calculation of mercury from a solid fuel. Liquid enables the calculation of mercury from a liquid fuel. Gas enables the calculation of mercury from a gaseous fuel. Fuel HG Mass Fraction field sets the correct mass fraction of Hg in the fuel (kg Hg per kg fuel). Volatile HG Mass Fraction specifies the mass fraction of mercury in the volatiles. This parameter appears only for Solid fuel streams. Char HG Mass Fraction specifies the mass fraction of mercury in the char. This parameter appears only for Solid fuel streams.

Mercury Pollutant Formation CL Intermediate drop-down list enables you to select Chlorine intermediate species (hcl, cl, or hcl/cl). Volatile CL Mass Fraction specifies the mass fraction of chlorine in the volatiles (kg Cl per kg volatiles). This parameter appears only for Solid fuel streams. Fuel CL Mass Fraction field sets the value for correct Chlorine intermediate species (hcl, cl, or hcl/cl) mass fraction in the fuel (kg chlorine per kg fuel). This parameter appears only for Liquid and Gas fuel streams. HCL Partition Fraction specifies the fraction of fuel chlorine that will become Cl. A partition fraction of 0 for HCl is equivalent to assuming that all fuel chlorine is converted to the final product Cl. Char CL Conversion drop-down list enable you to select the char chlorine path as hcl, cl, or hcl/cl. This parameter appears only for Solid fuel streams. Char CL Mass Fraction specifies the mass fraction of chlorine in the char. This parameter appears only for Solid fuel streams. Model Parameters contains parameters that control the formation of mercury.

[OH] Model is a drop-down list that allows you to specify the method by which the OH radical concentration is calculated, i.e., none, instantaneous, or partial-equilibrium. This list appears only for the Detailed mercury model. [O] Model is a drop-down list that specifies the method by which the O radical concentration is calculated, i.e., equilibrium, partial-equilibrium, or instantaneous.

Important
To use the concentration of OH or O predicted by the combustion model, select instantaneous for [OH] Modelor [O] Model. hg? Enable/disable the mercury model. hg-parameters/ Enter the mercury parameters menu. hg-chemistry Select mercury chemistry model. hg-turbulence-interaction Set mercury-turbulence interaction model.

Pollutants inlet-diffusion? Enable/disable inclusion of diffusion at inlets.

8.3.7. References
1. D. Dajnak and F. C. Lockwood., Modelling of Toxic Heavy Metal Mercury Partitioning from Pulverized Fuel Combustion , IFRF Combustion Journal, March 2001 2. J. I. Madsen, W. A. Rogers, and T. J. OBrien, Unstructured Multigrid Through Agglomeration In Computational Modelling of Mercury Control by Sorbent Injection , Baltimore, Maryland, 2004 3. B. Toole-ONeil, S. J. Tewalt, R. B. Finkleman, and D. J. Akers, "Mercury Concentration in Coal Unravelling the Puzzle", Fuel, 78(1), 4774, 1999 4. B. Hall, An Experimental Study of Mercury Reactions in Combustion Flue Gases , PhD thesis, Chalmers University of Technology, Sweden, 1991 5. J. Wilcox, "On the Path to Elucidating the Speciation of Mercury in the Flue Gases of Coal Combustion , PhD thesis, Dept. of Chemical and Environmental Engineering, he University of Arizona, Arizona, U.S.A., 2004 6. J. Wilcox, D. C. J. Marsden, and P. Blowers, "Evaluation of Basis Sets and Theoretical Methods for Estimating Rate Constants of Mercury Oxidation Reactions Involving Chlorine", Fuel Processing Technology, v85(5), 391400, 2004 7. J. Wilcox, J. Robles, D. C. J. Marsden, and P. Blowers, Theoretically Predicted Rate Constants for Mercury Oxidation by Hydrogen Chloride in Coal Combustion Flue Gases , Environmental Science and Technology, v37(18), 41994204, 2003 8. J. Wilcox, , "A Kinetic Investigation of High-Temperature Mercury Oxidation by Chlorine", Journal of Physical Chemistry A, v113(24), 66336639, 2009 9. J. A. Gasper, N. C. Widmer, J. A. Cole, and W. R. Seeker, Study of Mercury Speciation in a Simulated Municipal Waste Incinerator Flue", In Proceedings of 1997 International Conference on Incineration and Thermal Treatment Technologies, Oakland, California, 1997 10. J. Janicka and W. Kollmann, A Numerical Study of Oscillating Flow Around a Circular Cylinder , Combustion and Flame, v44, 319336, 1982 11. M. Missaghi, Mathematical Modelling of Chemical Sources in Turbulent Combustion , PhD thesis, The University of Leeds, England, 1987 12. G. Hand, M. Missaghi, M. Pourkashanian, and A. Williams, "Experimental Studies and Computer Modeling of Nitrogen Oxides in a Cylindrical Furnace", In Proceedings of the Ninth Members Conference, volume 2. IFRF Doc No K21/g/30, 1989

Chapter 9: Acoustics
9.1. Modal Analysis
ANSYS FLUENT implements a finite-volume method for computing the resonance frequencies and natural acoustic modes for any enclosure. Sound wave propagation, reflection, diffraction, and convection are taken into account. The formulation requires an input of the mean flow obtained from a steady state solution, together with prescribed boundary conditions. The method involves solving the Linearized Navier-Stokes Equations with the iterative Implicitly Restarted Arnoldi method to find the eigenvalues (frequencies) and eigenvectors (mode shapes of the pressure and velocity fluctuations).

9.1.1. Limitations
The following limitations apply to the modal analysis model currently implemented in ANSYS FLUENT: The modal analysis model is applicable to a steady-state, compressible ideal-gas solution. The modal analysis model is available for the double precision, serial ANSYS FLUENT version. The modal analysis model treats all domain boundaries as sound reflecting boundaries. Fluid particles can enter and leave the domain but acoustic waves are reflected.

9.1.2. Modal Analysis Theory

The system of 3D Linearized Navier-Stokes equations, in a Cartesian coordinate system are:

+ + + + =

(9.1)

+ + + + + + + ! " ! "

=
(9.2)

# %1 % 14 # # + % 14 + % 1 6 + ' #6 + ' # $ & 1 & 1 & 1 & 1

=
(9.3)

% ( 24 21 & 1

4 %2( 21 + & 1

4 ( 21 ( 21 0 + ) 5 % 23 %2 & 1 & 1 & 1

where 7 89 is the molecular stress tensor

Acoustics

(9.4)

and the subscript represents the mean values, and superscript represents the acoustic fluctuation about this mean (e.g. = + ). Note that a thermally perfect ideal gas is assumed. is the Kronecker delta function, is the molecular viscosity, is the conductive heat diffusivity coefficient, and are the specific heats at constant volume and constant pressure, respectively. The mean flow conditions are first solved from a steady-state solution of the Reynolds-Averaged NavierStokes equations. The fluctuating quantities are assumed harmonic functions in time, so that ! = "!# $% ! , where & = ' ( ) 0 1 . Substituting this into the linearized Navier-Stokes gives
23

4 5 5

789 5@ 5

(9.5)

The Jacobians

CDE FG A and B F

are the direct linearizations of the discretized flow equations.

With H = IP, this system of equations can be written as:

Q R

= SR
T

(9.6)
VWX Yb
T

` and c = d e , f is the solution of Y Y the eigenvalues problem. g is the eigenvector for the complex eigenvalue h = ip. The proper acoustic response of boundaries is included in the modal analysis model [1].

where

Equation 9.6 (p. 80) is an eigen-system and the iterative Implicitly Restarted Arnoldi method is use to compute a small set of eigenvalues and eigenvectors in a limited range of interest. The Implicitly Restarted Arnoldi method uses the ARPACK package (http://www.caam.rice.edu/software/ARPACK/), which is a collection of Fortran 77 subroutines designed to solve large-scale eigenvalue problems.

9.1.3. Using the Modal Analysis Model

Make sure you first enable beta feature access, as described in Introduction (p. 1). The procedure for computing the resonance frequencies and acoustic modes using the modal analysis model in ANSYS FLUENT is as follows: 1. Calculate a converged steady state compressible RANS solution. 2. Enable the Modal Analysis acoustics model. Models Acoustics Edit...

3. Set the associated Model Constants in the Acoustics Model dialog box. 4. Compute the resonance frequencies and acoustic modes by clicking the Solve button. 5. Postprocess the acoustic modes.

Modal Analysis Graphics and Animations Contours Set Up...

9.1.4. Setting Model Parameters

Under Model Constants in the Acoustics Model dialog box (Figure 9.1: The Acoustics Model Dialog Box (p. 81)), specify the relevant acoustic parameters used by the model. Figure 9.1: The Acoustics Model Dialog Box

Number of Frequencies is the requested maximum number of eigenvalues (natural acoustic frequencies). The default of 30. Frequency Shift is the frequency around which the eigenvalues will be solved. When the frequency shift is zero, the Arnoldi algorithm computes the eigenvalues around the smallest magnitude (SM). The default is 200Hz. Maximum Number of Arnoldi Iterations : the Implicitly Restarted Arnoldi Method is terminated after this many iterations if not converged. The default is 500. Residual Tolerance is the convergence criterion for the Arnoldi algorithm. The default is 0.001.

9.1.5. Postprocessing of the Modal Analysis Model

Postprocessing of the acoustic modes is accomplished by selecting the Acoustics category in the postprocessing dialog boxes: Acoustic Pressure Mode n where n ranges from 1 to 10.

9.1.5.1. References
1. Caraeni M., Devaki R. K., Aroni M., Oswald M., Srikanth KVSS, Efficient Acoustic Modal Analysis for Industrial CFD, AIAA-2009-1332, 2009
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates. 81

Chapter 10: Discrete Phase

10.1. Extended Collision Stencil
The extended collision stencil uses particles in neighboring cells to calculate the collision probability. The current model only uses particles in the current cell to calculate the collision frequency, where in this case particles in the closest N cells to the current particle location (where N is set by the user) are also allowed to collide. The model should be used when the mesh is very fine and the spray is represented by relatively few particles per cell. Using a larger sample volume improves the statistics of the collision calculation. Note that the cost for calculating collision depends on the square of n, where n is the number of particles in the sampled volume, so increasing the number of cells beyond 3 or 4 can increase the computation time significantly. To use this model (after enabling beta feature access, described in Introduction (p. 1) ), enter the following text command: define models dpm spray-model droplet-collision? and enter the following responses to the commands:
Spray collision model [yes] Spray collision model type [0-ORourke, 1-Stencil, 2-NTC] [0] 1 Spray collision stencil size [1] Spray collision event type [0-All, 1-Eff.Diam, 2-Impinging] [0]

10.2. Tracking of Child Droplets Within the Same Time Step

The DPM model is extended to consider the tracking of child droplets within the same time step when transient breakup models are enabled. The new child droplets are collected and tracked when the droplets are advanced within the time step. After enabling beta feature access, described in Introduction (p. 1), you can enable Consider Child Particles in the Same Tracking Step in the Physical Models tab of the Discrete Phase Model dialog box, as shown in Figure 10.1: The Discrete Phase Model Dialog Box (p. 84).

Discrete Phase Figure 10.1: The Discrete Phase Model Dialog Box

You can also enable this feature using the following text command: /define/models/dpm/spraymodel/consider-childs-in-the-same-tracking-step? Yes

10.3. Linearized Source Terms

Source terms for discrete phase momentum, energy, and species can be linearized with respect to the cell variable, :

= +

(10.1)

Temperature-Dependent Particle Density This linearization strongly increases numerical stability for steady flows. For transient flows, it typically allows the use of larger time steps and larger under-relaxation factors for the DPM model. The source term linearization can be combined with the Average DPM Source Terms (Node Based Averaging of Particle Data in the FLUENT User's Guide) and Update DPM Sources Every Flow Iteration (Options for Interaction with the Continuous Phase in the FLUENT User's Guide) options. However, you should disable Average DPM Source Terms when using linearized DPM source terms for vaporizing particles. Combination of these two features in vaporization cases may lead to numerical instabilities and unphysical results for the gas temperature. To enable source term linearization, use the TUI command: /define/models/dpm/interaction/linearized-dpm-source-terms? yes

10.4. Temperature-Dependent Particle Density

The density of discrete phase materials can be defined as a function of temperature by selecting one of the function types from the Density drop-down list in the Create/Edit Materials dialog box. See Defining Properties Using Temperature-Dependent Functions in the FLUENT User's Guide for details about how to select and use the temperature-dependence functions.

Chapter 11: Solver

This chapter contains information about the beta features related to the following topics: 11.1. Recursive Projection Method 11.2. Reduced Rank Extrapolation (RRE) Method 11.3. Second Order in Time For Moving Deforming Meshes 11.4. Moving Averages for Monitors 11.5. Executing Commands at a User-specified Iteration or Time Step

11.1. Recursive Projection Method

When using the density-based solver, in the Multigrid tab (Figure 11.1: The Multigrid Tab in the Advanced Solution Controls Dialog Box (p. 88)), you can choose the stabilization method to be the recursive projection method (RPM) in order to improve the convergence of the linear solver.

Note
Make sure you enable the access to the beta features (Introduction (p. 1)).

Solver Figure 11.1: The Multigrid Tab in the Advanced Solution Controls Dialog Box

For information about the Recursive Projection method, see Setting Algebraic Multigrid Parameters in the User's Guide.

11.2. Reduced Rank Extrapolation (RRE) Method

The Reduced Rank Extrapolation method (RRE) is a technique used to accelerate the convergence of numerical methods involving non-linear iterative solution algorithms. A Finite Volume simulation of the Navier-Stokes flow equations, as performed in ANSYS FLUENT, is such a method. Benefits of the application of RRE include better convergence rates, removal of residual stalling, and improved coupling between equations among different numerical models. The algorithm is independent of the type of

Reduced Rank Extrapolation (RRE) Method flow solver and equally applicable to explicit, implicit, and pressure and density based algorithms. The drawback is an increased memory consumption. The RRE method is representative of the set of so called Krylov type methods. Such methods explore the idea originally proposed by Krylov [1] that the solution of the linear system = lies within the space defined as (11.1)

which is also known as the Krylov subspace.

as a linear combination of orthogonal basis vectors of , which minimizes the Euclidian norm of the residual vector = . This is usually a better approximation
The RRE method obtains a vector method operates on that subset once the user specified size # (i.e. number of non-linear flow solver iterations) is reached. Then a new subspace is populated with the solution vectors from the following $ iterations and subsequently the RRE procedure is restarted. It was found that in certain cases it might be beneficial to not store the solution vector at each iteration, but to skip a certain number of iterations. This is because of the fact that two subsequent solution vectors are almost linearly dependent and do not provide much new information for the Krylov subspace. The number of skipped iterations is also a user specifiable value. ANSYS FLUENTs RRE method operates simultaneously on a predefined set of main flow variables, which are stored into a shared solution vector % & and uses the regular flow solver as a preconditioner to generate the products of the solution at the given iteration. A linear least squares procedure is applied in order to solve the minimization problem. Due to the memory requirements imposed by the need to keep solution vectors from previous iterations in memory, only a limited subset of the entire Krylov space ! " is stored. The

') (.

It can be shown that when applied to a linear problem, the RRE method is equivalent to the GMRES method derived by Y. Saad and M. H. Schultz [4]. Details of the method as well as investigations of its behavior when applied to the Navier-Stokes problems can be found in [2] and [3].

Note
The Krylov subspace data is not written into the data file for reasons of keeping the file size small and the i/o times limited. A certain jump of residual values if a precomputed data is read into a new ANSYS FLUENT session is expected. To use the Reduced Rank Extrapolation option, follow the steps outlined below: 1. Make sure you enable access to the beta features (Introduction (p. 1)). 2. In the Solution Methods task page, enable the Reduced Rank Extrapolation option.

Solver Figure 11.2: The Solution Methods Task Page

3. Click the Options... button to specify the RRE settings in the RRE Options dialog box. Figure 11.3: The RRE Options Dialog Box

a. Specify the size of the subspace in the RRE Options dialog box.

Second Order in Time For Moving Deforming Meshes b. Specify the number of iterations to skip while building it.

Note
In certain cases, it may be beneficial to increase the size of the Krylov subspace to 25 and store each 10th vector.

4. Run the solution.

11.2.1. References
1. A.N. Krylov, On the numerical solution of the equation by which in technical questions frequencies of small oscillations of material systems are determined, Izvestija AN SSSR (News of Academy of Sciences of the USSR), Otdel. mat. i estest. nauk, 1931, VII, Nr.4, 491-539 (in Russian). 2. A. Sidi, Efficient Implementation of Minimal Polynomial and Reduced Rank Extrapolation Methods. NASA Technical Memorandum 103240, ICOMP-90-20 3. A. Jemcov and J. P. Maruszewski, Nonlinear Flow Solver Acceleration by Reduced Rank Extrapolation. AIAA journal 2008-609 4. Y. Saad and M.H. Schultz, "GMRES: A generalized minimal residual algorithm for solving nonsymmetric linear systems", SIAM J. Sci. Stat. Comput., 7:856-869, 1986. doi:10.1137/0907058

11.3. Second Order in Time For Moving Deforming Meshes

For cases involving moving deforming meshes with layering, you can use the second order implicit formulation only after enabling beta feature access. To use the Second Order Implicit formulation, follow the steps outlined below: 1. Make sure you have a case file with dynamic mesh with layering defined. 2. Enable access to the beta features (Introduction (p. 1)). 3. In the Solution Methods task page, select Second Order Implicit from the Transient Formulation drop-down list.

Solver Figure 11.4: The Solution Methods Task Page

Note
The following limitations exist when using the second order formulation with moving deforming meshes: Manual mesh manipulations, swapping and remeshing are not supported. If your case has remeshing enabled, you will have the choice to either switch to First Order Implicit or disable Remeshing (in the Dynamic Mesh task page).

11.4. Moving Averages for Monitors

When using the Convergence Manager to monitor convergence, you may choose to employ the moving averages method if monitor convergence is proving difficult or the solutions contains many oscillations. The purpose of the moving averages option is to attain monitor convergence faster by replacing the values at each iteration/timestep with the calculated average values. Moving averages is only applicable to convergence monitors and works in conjunction with the Convergence Manager. All other calculations to decide monitor convergence remain same.

Executing Commands at a User-specified Iteration or Time Step Moving averages is available through the text user interface (TUI) console. /solve/monitors/convergence/ average-over-last-n-iterations-timesteps/ Option to average over previous values for checking convergence. To activate this feature, enable beta feature access (Introduction (p. 1)). You can enable moving averages by entering the following command into the TUI console:
/solve/monitors/convergence> average-over-last-n-iterations-timesteps Average every iteration [1]>

Important
average-over-last-n-iterations-timestep is only available when the Convergence Manager is invoked.

11.5. Executing Commands at a User-specified Iteration or Time Step

You can make modifications at a particular point during a simulation with the Execute command only once? command. This command gives you a choice to execute commands: at the beginning of a particular iteration, or at the end of a particular time step. The purpose of this option is to give you greater control over when modifications are applied; changes are made at a specified iteration or time step even if you halt the run and write the files. The Execute command only once? command is available through the text user interface (TUI) console. To activate this feature, enable beta feature access (Introduction (p. 1)) and set the session variable execute-command-at? to true.

11.5.1. Executing a Command at a Particular Iteration

The following example demonstrates how to execute a command at the beginning of a given iteration:
/solve/execute-commands> add-edit Name of the command [command-1] command-1 Adding command-1 Execute command only once? [no] yes

Options: "iteration" When ["iteration"] "iteration" iteration no. [1] 5 Command [""] "display close-window 1"

Note
This is available for both transient and steady-state cases.

Solver

11.5.2. Executing a Command at a Particular Time Step

The following example demonstrates how to execute a command at the end of a particular time step:
/solve/execute-commands> add-edit Name of the command [command-1] command-2 Adding command-2 Execute command only once? [no] yes

Options: "iteration" "time-step" When ["iteration"] "time-step" time-step no. [1] 5 Command [""] "display close-window 1"

Note
This is available only for transient cases.

Chapter 12: Custom Field Functions

12.1. Postprocessing Unsteady Statistics
You can postprocess unsteady statistics of any variable in your ANSYS FLUENT simulation, by performing the following steps: 1. Set up your transient problem. 2. Create a custom field function for the each of the variables for which you want to postprocess unsteady statistics, using the Custom Field Function Calculator (Figure 12.1: The Custom Field Function Calculator Dialog Box (p. 95)). For detailed instructions, see Creating a Custom Field Function. Define Custom Field Functions... Figure 12.1: The Custom Field Function Calculator Dialog Box

Important
The maximum number of custom field functions that can be calculated and postprocessed for unsteady statistics is 50.

3. Enable the beta feature access (as described in Introduction (p. 1)). 4. Enable data sampling for the unsteady calculation. Run Calculation Data Sampling for Time Statistics

5. Enable unsteady statistics for custom field functions by using the following text command:
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Custom Field Functions solve set unsteady-statistics-cff You will be prompted to enter the frequency at which the unsteady statistics will be sampled, as well as to specify the custom field functions you want to postprocess. 6. Run the calculation. Run Calculation Calculate 7. When the calculation is complete, the unsteady statistics for your custom field functions will be available for postprocessing as field variables. The root-mean-square of the function will be named RMS-function_name, and the mean value of the function will be named Mean-function_name, where function_name is the name of the custom field function you defined in step 2. For example, in the Contours dialog box, you could select Unsteady Statistics... and RMS-uns-custom-functon-0 for the Contours of drop-down lists (see Figure 12.2: The Contours Dialog Box (p. 96)). Figure 12.2: The Contours Dialog Box

Chapter 13: Turbomachinery

13.1. Pitch-Scale Model
In most turbomachines, the number of blades in the rotor row and the stator row are not equal. Therefore, the pitch angles of the rotor and stator passages are different. The most accurate simulations necessitate full domain modeling or whenever full pitchwise periodicity can be achieved. However, full domain modeling is demanding in terms of CPU and memory requirements. The pitch-scaling method will alleviate these requirements by modeling the flow between the rotor and stator passages by using a single or a few blades per row. It is important to note that this method is an approximation. The accuracy of the solution will improve as the pitch ratio between the rotor and stator section approaches unity. Therefore, if the pitch difference is large, it can be made small by adding an appropriate number of blade passages until the ensemble pitch-ratio is within acceptable limits. To activate this feature, enable beta feature access, as described in Introduction (p. 1). Now in the situation when a "Periodic Repeat" interface is created between two blade passages of different pitch angles, the pitch-scaling mechanism will be automatically deployed on the non-conformal interface. If the pitch between the blade passage is equal to unity, then the standard interface will be deployed. The pitch-scale method will maintain connectivity and full flow conservation between blade passages by stretching or compressing the solution profile on both sides of the interfaces. In the density-based implicit, and pressure-based segregated and coupled solvers, the pitch-scale implementation is fully coupled. Pitch-scaling is valid for 3D rotational turbomachines, such as radial or axial turbines and compressors. In steady-state modeling, it is used with the Multiple Reference Frame method (MRF) for modeling flow in an unequal-pitch rotor-stator problem. In order to maintain an acceptable solution accuracy, it is recommended that the pitch ratio between the two passages will not be more than 10%. In situations where the pitch difference is larger than 10%, it is recommended that a sufficient number of blade passages are added into the computational domain on each side of the interface to reduce the pitch difference. Compatibilities and limitations of the pitch-scale method, implemented as a beta feature, include the following: Applies to a 3D rotational turbomachine rotor-stator configuration, such as axial or radial compressors or turbines, or a torque converter Not available in 2D configuration or 3D translational problems (such as linear cascade) Not available currently with FMG initialization Available for steady-state simulations only Not available with the following models: Multiphase
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

Turbomachinery Radiation DPM Species Acoustics

13.2. Implicit Mixing-Plane Model

This beta feature is an alternate implementation of the mixing-plane model described in The Mixing Plane Model in the Theory Guide. However, in this implementation the non-conformal interfaces are utilized to provide full conservation as well as the mixing process instead of using boundary profile exchange as in the standard mixing plane model. Two mixing methods are provided in this model: Area Averaging Mass Averaging Area averaging should be sufficient for most flow cases. Mass averaging should be used only if there is no reverse flow in the pitch-wise direction. The implicit nature of this implementation should provide a much more robust mixing-process as well as fast convergence when used with implicit solution methods, such as the density-based implicit solver or both pressure-based solvers. This model is not available with the density-based explicit solver. To activate this feature, enable beta feature access, as described in Introduction (p. 1).

Implicit Mixing-Plane Model Figure 13.1: The Create/Edit Mesh Interfaces Dialog Box

This implementation works in conjunction with Periodic Repeats. Therefore, the upstream and downstream boundaries should be of interface types. Once Periodic Repeats is selected or created, the Mixing Plane model will be available for use. If you have already created Periodic Repeats and you want to apply a mixing process to it, then you should enable the Mixing Plane option and then click Update. Otherwise, if you have not yet created the Periodic Repeats, you can engage the Mixing Plane model at the same time you are creating the Periodic Repeats. The Interpolation Points are the number of radial or axial pitch-wise (depending on the machine type) locations used in reconstructing the circumferential averaging.

Note
If you want to convert the old standard mixing plane model and use this mixing plane model you must do the following:

Turbomachinery 1. Delete the old mixing plane. 2. Change the boundary type from inflow and outflow to interface type. 3. Follow the steps mentioned earlier to enable beta feature access and engage the mixing plane model with Periodic Repeats. Restrictions applied to Periodic Repeats in terms of model availability also apply to this model.

100

Chapter 14: Parallel Processing

This chapter contains information relating to parallel processing as beta features in ANSYS FLUENT 14.5. 14.1. Using Graphics Processing Units (GPUs) With the Algebraic Multigrid (AMG) Solver 14.2. Enhancing Parallel Performance and Convergence for the Algebraic Multigrid (AMG) Solver

14.1. Using Graphics Processing Units (GPUs) With the Algebraic Multigrid (AMG) Solver
Graphics Processing Units (GPUs) can be used to accelerate the Algebraic Multigrid (AMG) solver inside ANSYS FLUENT. GPU acceleration is currently only available for the pressure coupled equation system arising from the 3D pressure-based coupled solver. You can enable GPU acceleration using the solve/set/multi-grid-amg text user interface (TUI) command. An NVIDIA Fermi or later GPU is required along with CUDA 4.1. GPU acceleration of the AMG solver is limited to serial and shared memory parallel FLUENT sessions using a single GPU on lnamd64 (Red Hat Enterprise Linux 5/6, and SUSE Linux Enterprise Server 11) and win64 (Windows 7) platforms. As with any CPU system, the problem size that can be solved using a single GPU will depend on the amount of memory available on the GPU. The AMG cycle-type, coarse group size, and smoother weights all can impact convergence and performance. For instance, the V-cycle type converges more quickly than the default F-cycle type on the GPU. The group sizes for GPU execution are limited to 2, 4, and 8. Using a larger group size is faster, however, using a smaller group size leads to improved convergence. In addition, a smoother relaxation/weight can improve convergence for complicated problems (e.g., you can use the Scheme command (rpsetvar 'amg/ilu-cpld-relax 0.5) in order to explicitly set the relaxation to 0.5).

14.2. Enhancing Parallel Performance and Convergence for the Algebraic Multigrid (AMG) Solver
To improve parallel performance and convergence of a class of difficult problems, an AMG solver coarsening option, Conservative Coarsening? is provided as a beta feature in the Coupled Parameters dialog box. You can also enable the AMG solver coarsening using the solve/set/conservativeamg-coarsening? text user interface (TUI) command. With this option enabled, the solver considers parallel partitions as part of the multigrid coarsening process, as well as using different group sizes at different levels, generally improving convergence.

101

102

Chapter 15: FLUENT in Workbench

This chapter contains information relating to using beta features in ANSYS FLUENT 14.5 in Workbench. 15.1. Performing Transient Two-Way Simulations with FLUENT and ANSOFT 15.2. Working with Custom Input Parameters 15.3. Using UDFs to Compute Output Parameters

15.1. Performing Transient Two-Way Simulations with FLUENT and ANSOFT

If Beta Options are enabled in Workbench, you can perform transient two-way coupling simulations between FLUENT and ANSOFT. If Beta Options are disabled in Workbench, then FLUENT will not generate the feedback temperature file. Also note that once Beta Options are enabled in Workbench, you may also have to restart the FLUENT application and/or Workbench to proceed with the simulation. The following table illustrates the current support scenarios for FLUENT and ANSOFT coupling. Table 15.1: Current Support Scenarios for FLUENT and ANSOFT Coupling ANSOFT (Maxwell) 1 2 3 4 Steady Steady Transient Transient FLUENT Steady Transient Steady Transient One-way Support? Yes Yes Yes BETA Two-way Support? Yes Yes BETA BETA

Note
Since automatic system updates are not available for coupled Ansoft-FLUENT systems, you must perform cyclic updates of individual system components until the solution stops changing within a desired level of tolerance using Python scripts. For more information, see the FLUENT in Workbench Users Guide.

15.2. Working with Custom Input Parameters

Various ANSYS FLUENT setup related input quantities can be assigned to an input parameter. You can define a series of simulations based on a set of parametric values that are managed both in ANSYS FLUENT and in Workbench. These parameters may be defined for numeric cell zone and boundary condition settings using the New Input Parameter option in the corresponding drop-down list or by a small p icon adjacent to a specific input setting. However, various ANSYS FLUENT settings are not supported by these methods. If the beta-feature-access option is enabled in ANSYS FLUENT (as described in Introduction (p. 1)), you can mitigate this limitation using custom input parameters, and define input parameters for various ANSYS FLUENT simulation related settings. The define/parameters/custom-inputparameters/create command is used to define custom input parameters that will use other text
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates. 103

FLUENT in Workbench user interface (TUI) commands in FLUENT to change the desired simulation settings in a parametric manner. Each numerical component of the TUI command string can be marked and treated as a parameter. Setting up custom input parameters requires using Scheme functions that convert TUI commands into Scheme variables. Once defined, the Scheme variables are set as custom input parameters and are displayed in the Parameters dialog box alongside other input parameters. The input parameter passes a constant numeric value to the registered scheme function. Therefore, the associated Scheme function (and corresponding FLUENT text command) uses the constant parameter values using the units that were already defined for the designated text command quantity. /custom-input-parameters/ Enter the custom input parameters menu. create/ Create a custom input parameter. The following example demonstrates the create command. where you create a custom input parameter using a Scheme file called my-funct:
/define/parameters/custom-input-parameters> create Name of parameter ["parameter-1"] parameter-1 value [0] 0.3 Enter the name of custom-input-var variable as symbol [custom-input-var1] Enter the name of apply-function [()] my-funct /solve/set/under-relaxation/pressure 0.3

where the my-funct Scheme file contents are:

(define my-funct (lambda (value ) (ti-menu-load-string (format #f "/solve/set/under-relaxation/pressure ~g" value))))

delete Delete a selected custom input variable, but not the associated input parameter (the input parameter has to be deleted separately). Using the wildcard * allows you to delete all custom input variables at once. For example:
/define/parameters/custom-input-parameters> delete (custom-input-var3) custom-input-var name(1) [custom-input-var3] * custom-input-var name(2) [()] Are you sure you want to delete input parameter ("custom-input-var1" "custom-input-var2" "custom-input-var3")?

list Shows a list of defined custom input parameters along with their associated variables and apply functions. For example:
parameter-name -------------------parameter-3 parameter-2 parameter-1 -------------------custom-input-var -----------------------------custom-input-var3 custom-input-var2 custom-input-var1 -----------------------------apply-function -------------------my-funct my-funct my-funct --------------------

15.3. Using UDFs to Compute Output Parameters

ANSYS FLUENT allows you to create output parameters that let you compare reporting values for different cases. Output parameters are typically defined and computed through the graphical user interface and not accessible through user defined functions (UDFs). If the beta-feature-access option is enabled in ANSYS FLUENT (as described in Introduction (p. 1)), you can compute and publish real output parameter values computed by UDFs to Workbench (or ANSYS FLUENT). When the beta feature is enabled, a new output parameter type called udf is
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

104

Using UDFs to Compute Output Parameters available once you invoke the create command in the define/parameters/output-parameters menu. The create command has been extended to allow you to create a UDF-based output parameter. Using a registered UDF, you can compute your own output quantity. Currently, this command is called by FLUENT in Workbench at the end of a calculation. When you select the udf type, you are prompted for information regarding the name of the output parameter, the user-defined function, the name of any input parameters, etc. When you are finished, you are left with a specific output parameter that uses a UDF that in turn may consume one or more input parameters. For example:
define/parameters/output-parameters> create Output Parameter Type> drag-coefficient lift-coefficient flux moment-coefficient force surface-integral

udf volume-integral

Name of Output Parameter ["parameter-1"] Available udf of type output-parameter: ("mylibudf::libudf") output-parameter UDF function name ["mylibudf::libudf"] Do you want to use Input Parameters in the UDF Output Parameter? [no] yes Enter the no. of Input Parameters to be used in UDF [0] 2 Name of Input Parameter ["parameter-3"] parameter-3 value [0] 2 Name of Input Parameter ["parameter-4"] parameter-4 value [0] 3

To see the value of a particular output parameter, run the calculation for a few iterations (or initialize the solution), then type the define/parameters/outputparameters/printtoconsole text command.

105

106

Chapter 16: User-Defined Functions

This chapter contains information about the beta features related to the following topics: 16.1. Six-DOF Motion Constraint Using UDFs

16.1. Six-DOF Motion Constraint Using UDFs

For each moving object you have an associated 6DOF properties UDF in which you set e.g. the mass and the moments of inertia:
prop[SDOF_MASS] prop[SDOF_IXX] = prop[SDOF_IYY] = prop[SDOF_IZZ] = = 907.185; 27.116; 488.094; 488.094;

and other 6DOF properties. The entries of the 6DOF properties array are explained in DEFINE_SDOF_PROPERTIES in the UDF Manual. To constrain the motion of a 6DOF object, i.e. to set the motion to 0, you simply set in the UDF the corresponding 6DOF property array entries to TRUE, e.g.
prop[SDOF_ZERO_TRANS_X] = TRUE; prop[SDOF_ZERO_ROT_Y] = TRUE;

where SDOF_ZERO_TRANS_X, SDOF_ZERO_TRANS_Y, and SDOF_ZERO_TRANS_Z are the components of the translation and SDOF_ZERO_ROT_Y, SDOF_ZERO_ROT_Y, and SDOF_ZERO_ROT_Z are the components of the rotation. By default these entries are FALSE, i.e. the motion is not constrained.

107

108

Chapter 17: FLUENT as a Server

This chapter describes certain capabilities of the FLUENT as a Server feature that are released as beta features.

17.1. ANSYS Session Manager

ANSYS Session Manager provides a flexible approach to starting and managing multiple FLUENT as a Server sessions. It provides an abstraction layer through which the FLUENT Remote Console as well as client applications using the FLUENT as a Server SDK can start, stop, and access FLUENT solver sessions. ANSYS Session Manager is typically run on a Server machine on which the FLUENT as a Server sessions will be run. Once running, ANSYS Session Manager listens for connections from client applications and performs the requested session management actions. When ANSYS Session Manager receives a request for a new solver session to be started, it creates a folder for the new solver session in the TEMP folder belonging to the user running ANSYS Session Manager. It then starts the solver application in server mode and makes the connection information available to the client.

17.1.1. Using ANSYS Session Manager

The ANSYS Session Manager utility is provided as an executable and is located in the following directory within the ANSYS FLUENT install tree: v145\fluent\fluent14.5.0\addons\corba\<arch>

Note
If you are using the FLUENT Client Package instead of a full FLUENT installation, the ANSYS Session Manager files are located in a Beta subdirectory of the path above. Usage of the ANSYS Session Manager utility is as follows: > ANSYSSessionManager f<configfile> -p<portnumber> The following arguments and environment variables are recognized: <configfile> the name of a configuration file containing available installed solvers. For more information on the configuration file, see Configuring ANSYS Session Manager (p. 110). <portnumber> the port on which ANSYS Session Manager should listen for client connections. AAS_HOST an environment variable that is used to specify the IP address on which to listen for connections. Default value is localhost which will allow connections only from the local machine. To enable connections
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

109

FLUENT as a Server from remote clients, set this to the IP address of a network adapter on the local machine that is accessible remotely.

Important
When started, ANSYS Session Manager will display a message indicating two consecutive port numbers. The first of these is the port number used for connection from a client application.

17.1.2. Configuring ANSYS Session Manager

Before using ANSYS Session Manager you must create a configuration file. This file specifies which ANSYS solvers are installed on the Server machine and how to instantiate them. The configuration file is a text file of the following form where <userinput> indicates an item to be specified according to your specific installation:
__AnsysSessionManagerIni__ _NrAvailableApplications <integer> _Application_001 <solver 1 name> <solver 1 executable> <arguments> -wrapper=COWrapper.dll -wrapper-options=-multithreaded CORBA ICoFluentUnit _Application_002 <solver 2 name> <solver 2 executable> <arguments> -wrapper=COWrapper.dll -wrapper-options=-multithreaded CORBA ICoFluentUnit <etc...>

An example configuration file is included with the ANSYS FLUENT installation in the following location: %AWP_ROOT145%\fluent\fluent14.5.0\addons\corba\<arch>\AnsysSessionManager.ini You may need to edit or amend this file to reflect the local installation details on your Server Machine.

17.2. FLUENT Remote Console

17.2.1. Connecting to ANSYS Session Manager
To expose additional commands to connect to the ANSYS Session Manager from FLUENT Remote Console issue the following FLUENT Remote Console command:
>beta.enable asm

The ANSYS Session Manager commands are used to start, connect to, and disconnect from FLUENT solvers through an ANSYS Session Manager instance running on a local or remote machine. ANSYS Session Manager commands begin with the prefix asm. asm.connect_to_server <hostname> <portnumber> connect to the ANSYS Session Manager listening on <hostname>:<portnumber>. The portnumber is that specified with the -p command line option to ANSYS Session Manager. asm.list_applications list the available applications that can be started by the connected ANSYS Session Manager.

110

FLUENT Remote Console asm.start_application <appname> start a session of the application <appname> and connect to it. asm.list_sessions list the runnings sessions managed by the conntected ANSYS Session Manager that are available for connection. asm.connect_to_session <session_name> connect to the session <session_name> that is being managed by the connected ANSYS Session Manager. ANSYS Session Manager takes care of brokering the connection from the FLUENT Remote Console to the FLUENT as a Server session. However, once the client is connected to a solver session, it provides commands directly to the Component Session as described in FLUENT as a Server User's Guide without using the Session Manager as an intermediary.

17.2.2. Concurrent Access

Concurrent access allows you to control a FLUENT as a Server session that may already be processing a batch file or initiate batch processing from FLUENT Remote Console through an asynchronous call.

Important
No checking is currently performed to prevent conflicting commands when using concurrent access. Therefore you must use caution to prevent command conflicts that might result in corrupted files or unstable simulations. For example, reading a new case file into a paused simulation and then attempting to continue. You can enable concurrent access to a FLUENT as a Server session from the FLUENT Remote Console by issuing the following command in the FLUENT Remote Console:
>beta.enable concurrent

This feature exposes several additional commands in the FLUENT Remote Console: fluent.interrupt interrupt the batch processing. This is the equivalent of pressing Ctrl+C in an interactive FLUENT session. fluent.pause pause the batch processing. fluent.continue continue processing a batch job that has been paused with fluent.pause. fluent.read_journal <filename> instruct FLUENT to load and execute a remote journal file. This is an asynchronous call meaning FLUENT will return to FLUENT Remote Console immediately and then load and execute the journal file. Note the difference from file read-journal <filename> which is a TUI command and results in a syn-

111

FLUENT as a Server chronous call FLUENT will load and execute the journal file and return to FLUENT Remote Console after execution is complete.

Note
It is highly recommended that you set the listening interval using fluent.set_aaslistening_step_at prior to loading a batch file to ensure consistent behavior. fluent.set_aaslistening_step_at <number> iterations|timesteps sets the listening frequency on the FLUENT server to <number>\ iterations (or timesteps). For example, to specify that FLUENT should listen for FLUENT as a Server commands every 5 iterations use:
fluent.set_aaslistening_step_at 5 iterations

Once defined, the listening interval will be saved in the case file on case write and will be restored when the case file is read.

17.3. FLUENT as a Server SDK

Note
If you are using the FLUENT Client Package instead of a full FLUENT installation, the SDK files described in the following sections are located in a Beta subdirectory of the paths below.

17.3.1. IAnsysSessionManager CORBA Interface

The IAnsysSessionManager interface is a CORBA interface that includes methods to communicate with a running ANSYS Session Manager instance to query, start, and connect to solver sessions. In order to use the CORBA interface you must compile the AnsysSessionManager.idl file with a CORBA compiler suitable for your client development environment. This file is located in the following directory within the FLUENT install tree: v145\fluent\fluent14.5.0\addons\corba\<ARCH>\AnsysSessionManager.idl IAnsysSessionManager long getNrAvailableApplications(); returns the number of applications that ANSYS Session Manager has registered. string getApplicationNameByIndex(in long p_iApplicationIndex); returns the name of the application with index p_iApplicationIndex. string getApplicationInterfaceInfo(in string p_stringApplicationName); returns information about the interface of the named application. CoArrayString getApplicationAttributeNames(in string p_stringApplicationName); returns the attribute names of the named application. CoArrayString getApplicationAttributeDefaultValues(in string p_stringApplicationName); returns the default values of the attributes of the named application.

112

FLUENT as a Server SDK string startSession(in string p_stringApplicationName); starts a session of the application p_bstrApplicationName. string startSessionWithAttributes(in string p_stringApplicationName, in CoArrayString p_astringAttributeNames, in CoArrayString p_astringAttributeValues); starts a session of the application p_bstrApplicationName, with the specified attribute values. long cleanUp(); attempt to connect to all sessions started by the current session manager. If a session fails to connect it will be removed from the list of running sessions.

Note
p_stringLocation should be either the absolute path of an existing folder on the remote machine or a path relative to the folder in which ANSYS Session Manager is running.

17.3.2. COM Connectors

The CORBA interfaces described in FLUENT as a Server Software Development Kit (SDK) are also available as pre-compiled COM connectors in DLL libraries included with the FLUENT as a Server SDK.

17.3.2.1. Interfaces
You can make use of COM implementations of the ICoFluentUnit and ICoFluentSchemeController interfaces by including the following library in your application: ...\v145\fluent\fluent14.5.0\addons\corba\%ARCH%\COMCoFluentUnit.dll
Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

113

FLUENT as a Server This will make the following classes available: class CCoFluentUnit Calculate(void); iterates the solution for the number of iterations specified with put_NrIterations. This is mainly for use with steady simulations as it does not perform dual-time iteration. For transient cases you can issue the solve/dual-time-iterate TUI command using CCoSchemeController::DoMenuCommand get_ComponentName(void); returns the name of the connected component get_ComponentDescription(void); returns the description of the connected component get_NrInputParameters(void); returns the number of input parameters defined in the current case get_NrIterations(void); returns the number of iterations currently set for a Calculate command to perform get_NrOutputParameters(void); returns the number of output parameters defined in the current case getInputParameterNameByIndex(VARIANT &p_variantInputParameterIndex); returns the name of the input parameter with index p_variantInputParameterIndex getOutputParameterNameByIndex(VARIANT &p_variantOutputParameterIndex); returns the name of the output parameter with index p_variantOutputParameterIndex getOutputParameterValueByIndex(VARIANT &p_variantOutputParameterIndex); returns the value of the output parameter with index IOutputParameterIndex getOutputParameterValueByName(LPCTSTR p_bstrOutputParameterName); returns a string containing the name of the output parameter with name p_bstrOutputParameterName getSchemeControllerInstance(void); returns an object that can be used to send TUI or scheme commands to the FLUENT session and perform more advanced functions using the ICoFluentSchemeController Interface LoadCase(LPCTSTR p_bstrCaseFileName); load the case file p_bstrCaseFileName from the FLUENT working directory into FLUENT LoadData(LPCTSTR p_bstrDataFileName); load the data file p_bstrDataFileName from the FLUENT working directory into FLUENT put_ComponentDescription(LPCTSTR newValue); sets the connected component description to newValue put_ComponentName(LPCTSTR newValue); sets the connected component name to newValue put_NrIterations(VARIANT &newValue); sets the number of iterations that Calculate will perform to newValue.
114 Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

FLUENT as a Server SDK SaveCase(LPCTSTR p_bstrCaseFileName); save the current FLUENT case to p_bstrCaseFileName in the FLUENT working directory SaveData(LPCTSTR p_bstrDataFileName); save the current FLUENT data to p_bstrDataFileName in the FLUENT working directory setInputParameterValueByIndex(VARIANT &p_variantInputParameterIndex, VARIANT &p_variantInputParameterValue); sets the value of the input parameter with index p_variantInputParameterIndex to p_variantInputParameterValue setInputParameterValueByName(LPCTSTR p_bstrInputParameterName, VARIANT &p_variantInputParameterValue); sets the value of the input parameter with index p_lInputParameterIndex to p_lfInputParameterValue Terminate(void); terminate the connected FLUENT as a Server session setMaximumVerbosity(VARIANT &p_variantMaximumVerbosity); Reserved for future implementation setLoggingObject(LPDISPATCH p_pIDispatchCoLogger); Reserved for future implementation class CCoFluentSchemeController DoMenuCommand(LPCTSTR p_bstrMenuCommand); issues a TUI command to the connected FLUENT session. Output from the command is not returned DoMenuCommandToString(LPCTSTR p_bstrMenuCommand); issues a TUI command to the connected FLUENT session and returns the output DownloadFileToBuffer(LPCTSTR p_bstrRemoteFileName); returns the contents of the file named p_bstrRemoteFileName in the remote FLUENT session working directory. DownloadFileToFile(LPCTSTR p_bstrRemoteFileName, LPCTSTR p_bstrLocalFileName); writes the contents of the file named p_bstrRemoteFileName in the remote FLUENT session working directory to the local file p_bstrLocalFileName. ExecScheme(LPCTSTR p_bstrSchemeCommand); issues a scheme command to the connected FLUENT session. Output from the command is not returned ExecSchemeToString(LPCTSTR p_bstrSchemeCommand); issues a scheme command to the connected FLUENT session and returns the output SetRpVar(LPCTSTR p_bstrRpVar, LPCTSTR p_szRpVarValue); sets the value of the rpvar p_szRpVar to p_szRpVarValue. GetRpVar(LPCTSTR p_bstrRpVar); returns a string with the value of the rpvar p_bstrRpVar.

115

FLUENT as a Server UploadFileFromBuffer(LPCTSTR p_bstrRemoteFileName, VARIANT &p_variantLocalBufferContent); writes a file named p_bstrRemoteFileName in the remote FLUENT session working directory with the contents of p_variantLocalBufferContent. If p_bstrRemoteFileName exists, it is overwritten. UploadFileFromFile(LPCTSTR p_bstrRemoteFileName, VARIANT p_bstrLocalFileName); writes a file named p_bstrRemoteFileName in the remote FLUENT session working directory with the contents of p_bstrLocalFileName. If p_bstrRemoteFileName exists, it is overwritten. There is also a DLL library with an interface, IAnsysSessionManager, for connecting to ANSYS Session Manager. You can use this interface by including the following library in your application: ...\v145\fluent\fluent14.5.0\addons\corba\%ARCH%\COMAnsysSessionManager.dll class CAnsysSessionManager CleanUp(void); attempt to connect to all sessions started by the current session manager. If a session fails to connect it will be removed from the list of running sessions.

Important
If a FLUENT session is active but slow to respond (due to a long process of reading a case, for example) CleanUp will wait until the session responds. Thus, depending on the status of the running sessions this call may take a long time to complete. Use with caution. ConnectToRunningSession(LPCTSTR p_bstrRunningSessionName); connect to the managed session named p_bstrRunningSessionName. ConnectToSessionManager(LPCTSTR p_bstrHost, VARIANT &p_variantPort) establish a connection to the ANSYS Session Manager listening on the specified host and port. get_NrAvailableApplications(void); returns the number of applications that ANSYS Session Manager has registered. get_NrRunningSessions(void); returns the number of running solver sessions being managed by ANSYS Session Manager. getApplicationAttributeDefaultValues(LPCTSTR p_bstrApplicationName); returns the default attribute values of the named application. getApplicationAttributeNames(LPCTSTR p_bstrApplicationName); returns the attribute names of the named application. getApplicationInterfaceInfo(LPCTSTR p_bstrApplicationName); returns information about the interface of the named application. getApplicationNameByIndex(VARIANT &p_variantApplicationIndex); returns the name of the application with index &p_variantApplicationIndex. getRunningSessionInterfaceInfo(LPCTSTR p_bstrRunningSessionName); returns information about the interface of the named solver session.
116 Release 14.5 - SAS IP, Inc. All rights reserved. - Contains proprietary and confidential information of ANSYS, Inc. and its subsidiaries and affiliates.

FLUENT as a Server SDK getRunningSessionNameByIndex(VARIANT &p_variantRunningSessionIndex); returns the name of the solver session with index &p_variantRunningSessionIndex. ResuscitateSession(LPCTSTR p_bstrApplicationName, LPCTSTR p_bstrLocation); similar to StartSession, but instead of creating a new working folder, ResuscitateSession will start the application in the remote folder designated by p_bstrLocation.

Note
p_bstrLocation should be either the absolute path of an existing folder on the remote machine or a path relative to the folder in which ANSYS Session Manager is running. StartSession(LPCTSTR p_bstrApplicationName); starts a session of the application p_bstrApplicationName. StartSessionWithAttributes(LPCTSTR p_bstrApplicationName, VARIANT &p_variantAttributeNames, VARIANT &p_variantAttributeValues); starts a session of the application p_bstrApplicationName with attributes set to the specified values.

17.3.3. Registering the COM Connectors

If you are using a Windows platform you can register the COM connectors by using the following commands:
C:\>cd <ANSYS Installation>\ANSYS Inc\v145\fluent\fluent14.5.0\addons\corba\%ARCH%\ C:\>regsvr32 ComCoFluentUnit.dll C:\>regsvr32 ComAnsysSessionManager.dll

Note
You must have Administrator privileges to register the COM objects

117

118

Chapter 18: Population Balance

This chapter contains information relating to turbulence models available as beta features in ANSYS FLUENT 14.5. 18.1. Coulaloglou and Tavlarides Breakage

18.1. Coulaloglou and Tavlarides Breakage

The breakage frequency is give by

+ +

(18.1)

where and !" are constants, # is the dissipation rate, $ is the parent diameter, % is the surface tension, & is the volume fraction of the dispersed phase, ' is the dissipation of the primary phase, and () is the density of the dispersed phase [1].

Important
Make sure you first enable beta feature access, as described in Introduction (p. 1). Please refer to Section 3.3.1 in the Population Balance Module Manual to learn how to enable the population balance model. Enable the Breakage Kernel option and select tavlarides-model from the and Frequency drop-down list. Enter the desired Surface Tension in the Surface Tension for Population Balance dialog box.

18.1.1. References
1. Coulaloglou, C. A. and Tavlarides, L. L., Description of Interaction Processes in Agitated Liquid-Liquid Dispersions, Chem. Eng. Sci., 32 (1977) 1289-1297.

119

120

ANSYS Fluent Tutorial Guide 2020 R2 PDF
77% (13)
ANSYS Fluent Tutorial Guide 2020 R2 PDF
1,056 pages
Ansys Scripting in Mechanical Guide
No ratings yet
Ansys Scripting in Mechanical Guide
352 pages
ANSYS Polyflow Tutorial Guide
No ratings yet
ANSYS Polyflow Tutorial Guide
420 pages
ANSYS
100% (1)
ANSYS
12 pages
Fluent-FSI 14.5 Lect-00 Intro
50% (2)
Fluent-FSI 14.5 Lect-00 Intro
16 pages
Fluent Tips and Tricks
No ratings yet
Fluent Tips and Tricks
137 pages
Laminar Pipe Flow
No ratings yet
Laminar Pipe Flow
77 pages
Numerical Grid Generation
100% (2)
Numerical Grid Generation
336 pages
Prius Motor - Maxwell 2D Transient To FLUENT 3D Steady State Maxwell 2D Transient To FLUENT 3D Steady State Coupling
100% (1)
Prius Motor - Maxwell 2D Transient To FLUENT 3D Steady State Maxwell 2D Transient To FLUENT 3D Steady State Coupling
11 pages
Step by Step NX Open C++ Introduction - Cad Cam Development PDF
No ratings yet
Step by Step NX Open C++ Introduction - Cad Cam Development PDF
6 pages
American Cinematographer July 2013 US
100% (1)
American Cinematographer July 2013 US
92 pages
Oracle Net Suite PBC S Sync
No ratings yet
Oracle Net Suite PBC S Sync
13 pages
Tut 6 With Answer
No ratings yet
Tut 6 With Answer
7 pages
Fluent Tutorial Guide 182
100% (1)
Fluent Tutorial Guide 182
1,038 pages
ANSYS Mechanical APDL Modeling and Meshing Guide PDF
100% (1)
ANSYS Mechanical APDL Modeling and Meshing Guide PDF
242 pages
System Coupling Tutorials
No ratings yet
System Coupling Tutorials
236 pages
Macro Heat Exchanger Model Fluent Tutorial
100% (5)
Macro Heat Exchanger Model Fluent Tutorial
12 pages
Lecture 1: Overview of FSI in Workbench: ANSYS Fluent Fluid Structure Interaction (FSI) With ANSYS Mechanical
No ratings yet
Lecture 1: Overview of FSI in Workbench: ANSYS Fluent Fluid Structure Interaction (FSI) With ANSYS Mechanical
38 pages
Introduction To Fluent
No ratings yet
Introduction To Fluent
1 page
Turbulent Tutorial ANSYS Fluent
No ratings yet
Turbulent Tutorial ANSYS Fluent
23 pages
ANSYS Polyflow in ANSYS Workbench Users Guide 18.2
No ratings yet
ANSYS Polyflow in ANSYS Workbench Users Guide 18.2
60 pages
Workbench Users Guide
No ratings yet
Workbench Users Guide
436 pages
Module 02: Setting Up Domain: Introduction To ANSYS Fluent
No ratings yet
Module 02: Setting Up Domain: Introduction To ANSYS Fluent
13 pages
Teaching Room Modes and Diffraction Using Comsol Multiphysics
No ratings yet
Teaching Room Modes and Diffraction Using Comsol Multiphysics
15 pages
Iberian COMSOL Conference
No ratings yet
Iberian COMSOL Conference
112 pages
ANSYS Mechanical APDL Fluids Analysis Guide
No ratings yet
ANSYS Mechanical APDL Fluids Analysis Guide
44 pages
ANSYS Running ANSYS Fluent Using A Load Manager 18.2
No ratings yet
ANSYS Running ANSYS Fluent Using A Load Manager 18.2
54 pages
Discrete Phase Modelling - Combustion
No ratings yet
Discrete Phase Modelling - Combustion
63 pages
Mesh-Intro 15.0 WS 07e Manifold
No ratings yet
Mesh-Intro 15.0 WS 07e Manifold
25 pages
Fluent Tutorial 02 Modeling Periodic Flow and Heat Transfer
No ratings yet
Fluent Tutorial 02 Modeling Periodic Flow and Heat Transfer
26 pages
ICEM CFD Ansys Tutorial
No ratings yet
ICEM CFD Ansys Tutorial
30 pages
Introduction To ANSYS FLUENT
No ratings yet
Introduction To ANSYS FLUENT
64 pages
Boundary Conditions Available in ANSYS Fluent
100% (1)
Boundary Conditions Available in ANSYS Fluent
3 pages
Chapter 4: Use of UDF in Moving Deforming Mesh: 4.1. Preparation
No ratings yet
Chapter 4: Use of UDF in Moving Deforming Mesh: 4.1. Preparation
26 pages
Boundary Layer Thickness PDF
No ratings yet
Boundary Layer Thickness PDF
6 pages
6 Dof
No ratings yet
6 Dof
108 pages
Mesh-Intro 16.0 WS 04.1 ANSYS Meshing Methods PDF
No ratings yet
Mesh-Intro 16.0 WS 04.1 ANSYS Meshing Methods PDF
35 pages
14.5 Ansys CFX Tutorials
0% (1)
14.5 Ansys CFX Tutorials
792 pages
Symbolic TB
No ratings yet
Symbolic TB
1,498 pages
06 Udf Flow
No ratings yet
06 Udf Flow
19 pages
Turbulence Modeling - Beta
No ratings yet
Turbulence Modeling - Beta
88 pages
Introduction To ComSol MultiPhysics
No ratings yet
Introduction To ComSol MultiPhysics
184 pages
Fluent UDF File
100% (1)
Fluent UDF File
1 page
ANSYS Fluent Tutorial Part 1
No ratings yet
ANSYS Fluent Tutorial Part 1
24 pages
Moving and Repeating Load in APDL PDF
No ratings yet
Moving and Repeating Load in APDL PDF
4 pages
Mesh-Intro 16.0 WS 07d Assembly Meshing PDF
No ratings yet
Mesh-Intro 16.0 WS 07d Assembly Meshing PDF
26 pages
Fluent Lecture Fluid - Flow
No ratings yet
Fluent Lecture Fluid - Flow
42 pages
FLUENT Periodic Boundary Condition SOLVED
No ratings yet
FLUENT Periodic Boundary Condition SOLVED
10 pages
ANSYS 12.1 Tutorial
No ratings yet
ANSYS 12.1 Tutorial
2 pages
Fluent-Intro 16.0 AL2 Transient PDF
0% (1)
Fluent-Intro 16.0 AL2 Transient PDF
30 pages
SALOME Tutorial
0% (1)
SALOME Tutorial
82 pages
Discrete Phase Modeling: Srikanth Thorlikonda
No ratings yet
Discrete Phase Modeling: Srikanth Thorlikonda
41 pages
Geometric Parameter Optimization of A Tuning Fork: Created in COMSOL Multiphysics 5.5
No ratings yet
Geometric Parameter Optimization of A Tuning Fork: Created in COMSOL Multiphysics 5.5
10 pages
ANSYS Fluent Tutorial Part 2
No ratings yet
ANSYS Fluent Tutorial Part 2
10 pages
The Basics CFD
No ratings yet
The Basics CFD
63 pages
Meshing Tutorial Ans - Sys
No ratings yet
Meshing Tutorial Ans - Sys
350 pages
ANSYS CFX Introduction
No ratings yet
ANSYS CFX Introduction
68 pages
Fluent 15.0 Beta Features Manual
No ratings yet
Fluent 15.0 Beta Features Manual
174 pages
Fluent_Beta_Features_Manual
No ratings yet
Fluent_Beta_Features_Manual
612 pages
Fluent 14.5 Tutorial
100% (2)
Fluent 14.5 Tutorial
1,146 pages
Ansys 14 - Tutorial
100% (2)
Ansys 14 - Tutorial
82 pages
ANSYS FLUENT Getting Started Guide
No ratings yet
ANSYS FLUENT Getting Started Guide
26 pages
Flu Mig
No ratings yet
Flu Mig
34 pages
Using FLUENT 12.0 in Workbench: January 2009
No ratings yet
Using FLUENT 12.0 in Workbench: January 2009
66 pages
ANSYS Fluent Tutorial Guide
No ratings yet
ANSYS Fluent Tutorial Guide
11 pages
Pyrhonen Squirrel Cage Motor Calculation Mathcad13 PDF
No ratings yet
Pyrhonen Squirrel Cage Motor Calculation Mathcad13 PDF
35 pages
Transient Behaviour of Grounding Grids
No ratings yet
Transient Behaviour of Grounding Grids
115 pages
3 F 55
No ratings yet
3 F 55
5 pages
EV&T July2013 Article
No ratings yet
EV&T July2013 Article
1 page
Factors Influency The Design of Large High-Voltage Power Transformers 2ProceedingsofJan1968
No ratings yet
Factors Influency The Design of Large High-Voltage Power Transformers 2ProceedingsofJan1968
28 pages
Maxwell ANSYS Tutorial For Simulating Conductors For Inductance/Resistance Measurements Includes Analysis of Mesh Setup For ANSYS Adaptive Solutions
No ratings yet
Maxwell ANSYS Tutorial For Simulating Conductors For Inductance/Resistance Measurements Includes Analysis of Mesh Setup For ANSYS Adaptive Solutions
14 pages
TSW
No ratings yet
TSW
15 pages
Ansys Tutorial
No ratings yet
Ansys Tutorial
17 pages
Flux & FluxMotor New Features 12.2
No ratings yet
Flux & FluxMotor New Features 12.2
88 pages
Overcoming The Challenges of Hybrid/Electric Vehicle Traction Motor Design
No ratings yet
Overcoming The Challenges of Hybrid/Electric Vehicle Traction Motor Design
12 pages
Appendix C: ANSYS TEG Tutorial
No ratings yet
Appendix C: ANSYS TEG Tutorial
11 pages
8 Nvidia PDF
No ratings yet
8 Nvidia PDF
48 pages
Simcenter 3D LFEM - Tutorial Power Transformer
No ratings yet
Simcenter 3D LFEM - Tutorial Power Transformer
40 pages
Keywords: Powerful Electrical Transformer, Losses, Heating, Three-Dimensional Modeling
No ratings yet
Keywords: Powerful Electrical Transformer, Losses, Heating, Three-Dimensional Modeling
9 pages
Presentation of Cigre SC A2 Transformers'
No ratings yet
Presentation of Cigre SC A2 Transformers'
7 pages
Analysis of Ac Contactors Combining Electric Circuits, Time-Harmonic Finite Element Simulations and Experimental Work
No ratings yet
Analysis of Ac Contactors Combining Electric Circuits, Time-Harmonic Finite Element Simulations and Experimental Work
14 pages
Cigre Trafo
No ratings yet
Cigre Trafo
11 pages
Performance DebuggingTools
No ratings yet
Performance DebuggingTools
21 pages
Power Transformer
No ratings yet
Power Transformer
5 pages
Prace Autumn School 2013 Ponzini Cineca Sailddeslecture
No ratings yet
Prace Autumn School 2013 Ponzini Cineca Sailddeslecture
20 pages
8Qlyhuvlgdggh1Dyduud: 1Diduurdnr8Qlehuwvlwdwhd
No ratings yet
8Qlyhuvlgdggh1Dyduud: 1Diduurdnr8Qlehuwvlwdwhd
19 pages
Indian Customs EDI System
No ratings yet
Indian Customs EDI System
10 pages
St-Module1-Note 4
No ratings yet
St-Module1-Note 4
19 pages
The Murmurator: A Flocking Simulation-Driven Multi-Channel Software Instrument For Collaborative Improvisation
100% (1)
The Murmurator: A Flocking Simulation-Driven Multi-Channel Software Instrument For Collaborative Improvisation
6 pages
Word Exercises
100% (1)
Word Exercises
8 pages
FMEA_Nasa
No ratings yet
FMEA_Nasa
12 pages
Mohammed Rafi
No ratings yet
Mohammed Rafi
4 pages
Informatica Functions Explanation: IIF: Syntax
No ratings yet
Informatica Functions Explanation: IIF: Syntax
4 pages
TwinCAT PLC PDF
No ratings yet
TwinCAT PLC PDF
2 pages
WEIGHING SCALE Service Guide by Jose Gellado IV
No ratings yet
WEIGHING SCALE Service Guide by Jose Gellado IV
9 pages
Atomlab 500 Quick Reference Procedures: Iodex
No ratings yet
Atomlab 500 Quick Reference Procedures: Iodex
4 pages
Oomd QB
No ratings yet
Oomd QB
5 pages
Circular 20241126140554 Seating Plan 27.11.2024 Group-Ii
No ratings yet
Circular 20241126140554 Seating Plan 27.11.2024 Group-Ii
21 pages
Document
No ratings yet
Document
517 pages
FC-MX-TAX-188 - Create in SAP A Withholding Tax Report For Mexico v2
100% (1)
FC-MX-TAX-188 - Create in SAP A Withholding Tax Report For Mexico v2
14 pages
VLAN Tricks With NICs - Teaming & Hyper-V
No ratings yet
VLAN Tricks With NICs - Teaming & Hyper-V
4 pages
What Is The History of Artificial Intelligence (AI) - Tableau
No ratings yet
What Is The History of Artificial Intelligence (AI) - Tableau
10 pages
Asd 2012
No ratings yet
Asd 2012
3 pages
Making The Game in p5.js: Our Player Character
No ratings yet
Making The Game in p5.js: Our Player Character
11 pages
MPMC Unit-1-Part-B PDF
No ratings yet
MPMC Unit-1-Part-B PDF
23 pages
Princy K Patel - Resume
No ratings yet
Princy K Patel - Resume
2 pages
M6S Datasheet English
No ratings yet
M6S Datasheet English
2 pages
Chapter 2: Objects and Primitive Data: Solutions
100% (1)
Chapter 2: Objects and Primitive Data: Solutions
13 pages
CSE 105: Discrete Mathematics
No ratings yet
CSE 105: Discrete Mathematics
42 pages
AR5516 Brochure
No ratings yet
AR5516 Brochure
2 pages
MagArrow Spec Sheet
No ratings yet
MagArrow Spec Sheet
2 pages
Assignment - 1: Shree V.J. Modha College of Information Technology
No ratings yet
Assignment - 1: Shree V.J. Modha College of Information Technology
15 pages

Fluent Tutorial

Uploaded by

Fluent Tutorial

Uploaded by

FLUENT 14.

5 Beta Features Manual

Release 14.5 October 2012

Beta Features Manual Copyright and Trademark Information

U.S. Government Rights

3.2. Meshing Mode Access

Chapter 4: Boundary Conditions

Boundary Conditions Figure 4.1: The Operating Conditions Dialog Box

4.2. Non-Reflecting Boundary Conditions

Chapter 5: Heat Exchangers

5.2. Alternate Formulation for the Dual Cell Heat Exchanger

6.1. Explicit Algebraic Reynolds Stress Model

@ is given by: DFE DF =

G TU = HYITU + H` P TV P VU QQWR TU + Ha ITV P VU P TV IVU + Hb ITV P VXP XU + P TV P VXI XU

Explicit Algebraic Reynolds Stress Model

term) is neglected in Equation 6.5 (p. 14). Second,

standard coefficients are used.

6.1.1. Accessing the WJ-BSL-EARSM Model

6.2. Wall-Modeled LES (WMLES) S-Omega Formulation

6.2.1. Accessing the WMLES S-Omega Model

Turbulence Figure 6.3: The Fluid Dialog Box

6.3. Near Wall Treatment for the Porous Media Interface

or dissipation rate wall values are pre-

6.3.1. Accessing the Turbulent Wall Treatment Option

6.4. Near Wall Treatment for Models

: fully turbulent region

! : viscosity-affected near wall region

6.4.1.1. Momentum Equations

(6.8) and   are friction velocities. Both the friction

The blending for the friction velocity, 5 6, is:

"# )01 = + % 234 ( &' $

6.4.1.2. H I Turbulence Models

Near Wall Treatment for Y ` Models

(6.13) , 576 = , 89 = , @A = , and BC = .

6.4.1.3. Iteration Improvements

Turbulence Figure 6.8: Residual history using new iterative procedure

Figure 6.9: Residual history using standard procedure

6.4.2. User Interface

between wall and inlet.

Figure 6.11: Zero pressure gradient flat plate boundary conditions

Turbulence Figure 6.13: Stanton number

6.5. Buoyancy Effects on Omega-Based Turbulence Models

I-based differential Reynolds stress models

6.5.2. User Interface

7.2. Number of Species in Reacting Flows

8.1. Coal Derived Soot

 =  

xy = d efghihjfkhlm tuv = notuvp w xyz {|}~| =

(8.5) (8.6) (8.7)

Coal Derived Soot

6.77e+05 ( /kg-s) 9.77e+10 (1/s) 5.02e+08 (1/s)

52,300 286,900 198,900

8.1.1. Using the Coal Derived Soot Model

is recommended for the Mean density of soot particle.

Mercury Pollutant Formation

8.2. Atomic Balance for Sulfur

8.3. Mercury Pollutant Formation

8.3.1. Mercury Speciation in Coal Flames

8.3.1.2. Governing Equations for Mercury Transport

Mercury Pollutant Formation

(8.10) (8.11) (8.12)

 +  "&'() + %&'()

and vwxy are to be determined depending on the form of fuel

(8.13) (8.14) (8.15) (8.16)

), where i = Hg, HgCl2,..., HCl,

) from fuel, where i = Hg, HCl

) from gas phase reactions,

where i = Hg, HgCl2,..., HCl, Cl,...

= source of i ( ) due to capture into solid

8.3.1.3. Mercury Speciation Model

8.3.1.3.1. One Step Mechanism

8.3.1.3.2. Two Step Mechanism

Mercury Pollutant Formation

8.3.1.3.3. Detailed (Wilcox) Mechanism

Figure 8.3: Chlorine Only Reaction Subset

Wilcox [5] mercury chlorination/oxidation model consists of 12 species (

(6.8) and are friction velocities. Both the friction

=

xy = d efghihjfkhlm tuv = notuvp w xyz {|}~| =

+ "&'() + %&'()

= volatile release (devolatilization) rate from coal

$ !"# = mass fraction of j" in volatiles, where j = Hg or % = cell volume (&' )

"#$$%& '! ()0 8 12345637 and 9@A H BCADEFAG respectively) are