Patran 2008 R1 Thermal User's Guide Volume 2: Viewfactor Analysis

  • Uploaded by: Kevin
  • 0
  • 0
  • May 2020
  • PDF

This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA


Overview

Download & View Patran 2008 R1 Thermal User's Guide Volume 2: Viewfactor Analysis as PDF for free.

More details

  • Words: 48,633
  • Pages: 206
Patran 2008 r1 Thermal User’s Guide Volume 2: Viewfactor Analysis

Main Index

Corporate

Europe

Asia Pacific

MSC.Software Corporation 2 MacArthur Place Santa Ana, CA 92707 USA Telephone: (800) 345-2078 Fax: (714) 784-4056

MSC.Software GmbH Am Moosfeld 13 81829 Munich, Germany Telephone: (49) (89) 43 19 87 0 Fax: (49) (89) 43 61 71 6

MSC.Software Japan Ltd. Shinjuku First West 8F 23-7 Nishi Shinjuku 1-Chome, Shinjuku-Ku Tokyo 160-0023, JAPAN Telephone: (81) (3)-6911-1200 Fax: (81) (3)-6911-1201

Worldwide Web www.mscsoftware.com

Disclaimer This documentation, as well as the software described in it, is furnished under license and may be used only in accordance with the terms of such license. MSC.Software Corporation reserves the right to make changes in specifications and other information contained in this document without prior notice. The concepts, methods, and examples presented in this text are for illustrative and educational purposes only, and are not intended to be exhaustive or to apply to any particular engineering problem or design. MSC.Software Corporation assumes no liability or responsibility to any person or company for direct or indirect damages resulting from the use of any information contained herein. User Documentation: Copyright ©2008 MSC.Software Corporation. Printed in U.S.A. All Rights Reserved. This notice shall be marked on any reproduction of this documentation, in whole or in part. Any reproduction or distribution of this document, in whole or in part, without the prior written consent of MSC.Software Corporation is prohibited. The software described herein may contain certain third-party software that is protected by copyright and licensed from MSC.Software suppliers. Contains IBM XL Fortran for AIX V8.1, Runtime Modules, (c) Copyright IBM Corporation 1990-2002, All Rights Reserved. MSC, MSC/, MSC Nastran, MD Nastran, MSC Fatigue, Marc, Patran, Dytran, and Laminate Modeler are trademarks or registered trademarks of MSC.Software Corporation in the United States and/or other countries. NASTRAN is a registered trademark of NASA. PAM-CRASH is a trademark or registered trademark of ESI Group. SAMCEF is a trademark or registered trademark of Samtech SA. LS-DYNA is a trademark or registered trademark of Livermore Software Technology Corporation. ANSYS is a registered trademark of SAS IP, Inc., a wholly owned subsidiary of ANSYS Inc. ACIS is a registered trademark of Spatial Technology, Inc. ABAQUS, and CATIA are registered trademark of Dassault Systemes, SA. EUCLID is a registered trademark of Matra Datavision Corporation. FLEXlm is a registered trademark of Macrovision Corporation. HPGL is a trademark of Hewlett Packard. PostScript is a registered trademark of Adobe Systems, Inc. PTC, CADDS and Pro/ENGINEER are trademarks or registered trademarks of Parametric Technology Corporation or its subsidiaries in the United States and/or other countries. Unigraphics, Parasolid and I-DEAS are registered trademarks of UGS Corp. a Siemens Group Company. All other brand names, product names or trademarks belong to their respective owners.

P3:V2008R1:Z:VWFTR:Z:DC-USR-PDF

Main Index

Contents Viewfactor Analysis

1

Introduction About the Viewfactor Program Features and Benefits About this Guide

3

5

Using this Guide 6 Assumptions About the User Other Pertinent Documents Guide Organization

2

6 6

7

Overview of Viewfactor Analysis

8

Nomenclature 9 Conventions 9 Font Types and Typefaces 9 Filenames 9 Units 9

2

Overview Purpose

12

Relationship of Viewfactor to Patran and Patran Thermal Description 13 Viewfactor Data and Program Flow Description 16

13

16

Summary of the Analysis Cycle for a Thermal Radiation Problem Problem Definition 20 General Preprocessing 20 General Patran Thermal Preparation 20 Thermal Radiation Specific Preprocessing 20 Preparation for Viewfactor Analysis 21 Viewfactor Analysis 21 Post Viewfactor Analysis 21 Patran Thermal Analysis 21 Postprocessing 22 Refinements 22

3

Model Creation for a Thermal Radiation Problem Purpose

24

Radiation Enclosure Concept Definition of Enclosure 25 The Enclosure ID 25

Main Index

25

20

ii Viewfactor Analysis ==

Wavebands and Enclosures 25 list2+s of the Use of Enclosures 26 Surface Orientation in Patran 31 The Importance of Surface Orientation 31 Determining Surface Orientation 34 Correcting Improper Surface Orientations 36 Suggested Practices for Creating Properly Oriented Surfaces

37

Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary Condition) 38 Purpose of the Viewfactor Form 38 Form for the VFAC LBC 38 Input Data for the Viewfactor Form 38 Requirement for Oriented 2-D Surfaces Related to the VFAC LBC 40 Neutral File Data Packet Created from the VFAC LBC 40 Advanced Features of the VFAC Boundary Condition 42 Referencing Participating Media Radiation Nodes 42 Role of Radiation Participating Media in Patran Thermal 42 Participating Media Resistor Networks 42 Defining the Participating Media Node 44 Referencing Ambient or Space Radiation Nodes 44 Role of Ambient Radiation Nodes 44 Suggested Practices to Improve Model Accuracy 44 Ambient Node Resistor Networks in Patran Thermal 45 Defining the Ambient Node 46 Identifying a Surface as Being Convex 47 Significance of Convex Surfaces to Viewfactor Calculations 47 Benefits of Identifying Convex Surfaces to Viewfactor Execution Time 47 Caveats Regarding Convex Surfaces in Axisymmetric Models 48 Identifying a Surface as Not Obstructing the View Between Other Surface Pairs 49 Benefits of Identifying Nonobstructing Surfaces to Viewfactor Execution Time 49 Examples with Nonobstructing Surfaces Identified 49 Caveats Regarding Nonobstructing Surfaces in Axisymmetric Models 49 Relationship of VFAC LBC Data to VFINDAT File Data

51

Patran Thermal TEMPLATEDAT Files for Surface Property Description 52 Thermal Radiation Wavebands as Used in MSC Patran Thermal User’s Guide 52 Radiation Resistor Types Used in Patran Thermal 54 Radiation Resistor Subtypes 55 Patran Thermal MPIDs (Material Property IDs) 56 Patran Thermal=Material Property Definition 57 VFAC Template Format 58 Examples of TEMPLATEDAT Files for Thermal Radiation 61 Compatibility Requirements for Model and VFAC Templates Origin of the Problem 65 Suggested Procedures to Avoid Compatibility Problems 65

65

Symmetry as Applied to the Model and Viewfactor Radiation Exchange The Purpose of Symmetry in Viewfactor 67 Caveats Concerning the Use of Symmetry in Thermal Radiation Modeling Symmetry Operations Supported in Viewfactor 67 Symmetry in 2-D XY Space 68 Symmetric Reflection About a Line 68 Symmetric Rotation About an Axis 68

Main Index

67 67

CONTENTS iii

Combining Symmetry Operations 70 Symmetry in 3-D XYZ Space 70 Symmetric Reflection About a Plane 70 Symmetric Rotation About an Axis 71 Combining Symmetry Operations 72 RZ (Axisymmetric) Space 73 Symmetric Reflection About a Line 73 list2+ of the Use of Symmetry in Thermal Radiation Modeling list2+ Which Appears Symmetric, But in Fact Is Not Symmetric Entering Viewfactor Symmetry Operations in the Patran Model

4

74 74 75

Preparation for Analysis Introduction

78

Viewfactor Execution From Patran Thermal

79

PATQ Translation from the Patran Neutral File to the VFINDAT File Spawning From Patran vs. Stand-Alone Execution 80 Step-by-Step Procedure (Stand-Alone Execution) 80 VFCTL, the Viewfactor Program Execution Control File 85 Philosophy and Structure of the VFCTL File 85 Example VFCTL File 86 Keywords in the VFCTL File 88 $PATH: pathname 88 $MESSAGE_FILE: filename 88 $STATUS_FILE: filename 89 $DIAGNOSTIC_FILE: filename 89 $TITLE: title 89 $RESTART_FILE: filename 89 $IN_DATA: filename 89 $TEMPLATE_FILE: filename 90 $RAW_DATA: filename 90 $OUT_DATA: filename 90 $RAD_NODE_FILE: filename 90 $RUN_CONTROL: value 91 $RESTART_FLAG: value 92 $CONVERGE: value 92 $ZERO: value 92 $APPROX_CURVE: value 93 $GAUSS_ORDER: contours double_area weighting 94 $AXISYM_SURFACE: minimum maximum 95 $EOF: 95

5

Analysis Submitting a Viewfactor Job for Analysis Review the Viewfactor Control/Parameters Review Directory for Required Files 98 $RUN_CONTROL: 0 98 $RUN_CONTROL: 1 99 $RUN_CONTROL: 2 99 The Viewfactor Command Line 99 Output Created by a Viewfactor Execution

Main Index

98 98

101

80

iv Viewfactor Analysis ==

VFMSG 101 VFDIAG 101 VFRAWDAT 102 VFRESDAT 102 VFNODEDAT 102 Reviewing the Viewfactor Output 103 VFMSG, the Viewfactor Message File 103 VFDIAG, the Viewfactor Diagnostic Data File

6

106

Post-Analysis Introduction

110

Interface From Viewfactor to Patran Thermal 111 Viewfactor VFRESDAT and VFNODEDAT Files as Input to Patran Thermal’s QTRAN Translating Binary Resistor File VFRESDAT to a Text File, VFRESTXT 111 Procedure in Patran Thermal’s PATQ to Translate Binary Files to Text 111 The VFRESTXT Resistor Text File 115 noteplains on Resistor Values THERMAL Analysis

116

117

THERMAL Results Postprocessing

7

118

Changing the Surface Template Data After Viewfactors are Calculated Introduction

120

Compatible VFAC LBC and Template Data

121

New Resistors from Raw Viewfactor Data 122 Changing TEMPLATEDAT VFAC Templates 122 Changing Patran Thermal Material Definitions 122 Changing=VFCTL 122 Submitting the New Viewfactor Job 122

8

Theory and Computational Limitations Introduction Viewfactor

124 125

Mean Beam Length Obstructions

126

128

Computational Limitations 129 Grazing Incidence of the Intersurface Ray with the Surface Spatial Resolution 129 Extreme Scales 129

9

Data File Specifications Introduction

Main Index

132

129

111

CONTENTS v

VFINDAT (Input Data File) 133 Examples 133 Example 1 134 Example 2 135 Detailed Descriptions 136 $TITLE 136 $SIZE 136 $SYM and $ENDSYM 136 $NODES and $ENDNODES 137 $EOF 139 VFRAWDAT (Raw Viewfactor Data) VFRESDAT (Resistor Data)

140

144

VFDIAG (Diagnostic Data) 145 Introduction 145 Examples 146 Detailed Descriptions 147 $TITLE 147 $ENCL and $ENDENCL 147 $EOF 148 TEMPLATEDAT (Surface Pointer Data) VFNODEDAT (Radiosity Node Lists)

10

149 151

Rules for Radiation Resistors Introduction

154

General Rules for Radiation Resistors Rules for Emissivity Resistors Rules for Radiosity Resistors

A

8

10

Example Thermal Radiation Problems Purpose

Main Index

4

Machine-Specific File Names for Viewfactor Purpose

E

2

Memory Requirements for Viewfactor Execution Purpose

D

157

Quick Reference Guide to Viewfactor Purpose

C

156

Typical Errors and Probable Causes for Viewfactor Errors Purpose

B

155

12

vi Viewfactor Analysis ==

Problem 1 - Steady-State Radiative Boundary Conditions Objectives 13 Model Description 13 Exercise Procedure 14 Temperature Boundary Conditions: 18 Viewfactor Boundary Condition: 21 Problem 2 - Parallel Semi-Infinite Plates $QTRAN 34 Problem 3 - Heated Reaction Chamber

Main Index

32 35

13

Chapter 1: Introduction Patran Viewfactor Analysis

1

Main Index

Introduction



About the Viewfactor Program



Features and Benefits



About this Guide

5



Using this Guide

6



Guide Organization



Overview of Viewfactor Analysis



Nomenclature

9

2

3

7 8

2

Patran Viewfactor Analysis About the Viewfactor Program

1.1

About the Viewfactor Program The Viewfactor program in the Patran system is designed to facilitate the generation of finite element based radiation viewfactors. These viewfactor calculations are intended to be used as input into the Patran Thermal analysis code, although there should be nothing to stop it from being used in conjunction with other similar thermal analysis codes. It was primarily designed to fill a void in the thermal capabilities of the Patran system and to further expand the potential application of the Patran Thermal module.

Main Index

Chapter 1: Introduction 3 Features and Benefits

1.2

Features and Benefits Viewfactor provides support for finite element analysis of thermal radiation phenomena in the Patran Thermal module by providing the viewfactors and thermal network resistors between thermally radiating surfaces. Presently, diffuse surfaces (surfaces whose radiative properties are independent of direction) are supported. The surface properties are permitted to have spectral dependence in the form of an arbitrary number of piecewise gray wavebands. Thus, the Viewfactor program provides a practical means of modeling diffuse spectral surfaces. Viewfactor provides support for radiation interchange between the following element faces: • Linear quadrilateral faces of elements such as the Patran HEX, WEDGE, and QUAD elements; • Linear triangular faces of elements such as the Patran TET, WEDGE, and TRI elements; • Linear bar edges for 2-D elements such as Patran QUAD, TRI, and BAR elements; • Linear bar edges for axisymmetic elements such as Patran QUAD, TRI, and BAR elements;

These elements faces represent all of the elements faces usually needed to model thermally radiating surfaces. Viewfactor provides support for multiple symmetries in the model. For 3-D geometries, Viewfactor provides support for reflection across a plane and rotation of “n” times by “x” degrees about an arbitrary axis. For 2-D XY geometries, it provides for reflections across a line in the XY plane and rotation of “n” times by “x” degrees about an arbitrary axis perpendicular to the XY plane. For 2-D axisymmetric geometries, it provides for reflections across a line perpendicular to the Z-axis and in the RZ plane. Four separate symmetry objects and symmetry operations may be combined in the same model. Viewfactor does not check for the validity of any symmetry operations. These symmetry operators provide a convenient way to deal with thermal models which are symmetric in all respects except for the radiation component of the problem. Viewfactor is closely coupled with Patran and Patran Thermal and is specifically designed to work well with them. Patran is responsible for the generation of the thermal model and boundary conditions. The Patran Thermal PATQ interface program handles all data translation between the Patran neutral file and the Viewfactor input file. Viewfactor reads the input file and the Patran Thermal template file and then generates a file containing radiation resistors for QTRAN analysis processing. Viewfactor provides support for all of the commonly used model coordinate systems. These are 2-D XY, 2-D RZ axisymmetric, and 3-D XYZ coordinate frames. Coupled with the capabilities in the Patran Thermal module Viewfactor provides support for time and temperaturedependent material properties such as surface emissivities and participating media transmissivities. Viewfactor provides for efficient obstruction checking at the element level. Multiple enclosures are modeled in both Viewfactor and Patran Thermal. This feature, along with the multiple wavebands for spectral surfaces, allows for the modeling of phenomenon such as partially transmitting windows. No fixed problem size limit exists for Viewfactor. Memory for the particular model being analyzed is allocated during run time. Problem size is limited only by available virtual memory, available CPU resources, and storage space for the output data. Core memory requirements are linear functions of the model size, not higher order functions as is true for some viewfactor analysis programs. Through use of the Patran Thermal template file, Viewfactor provides support for optically thin participating media. The geometric calculations involved in determining obstructions and viewfactors are saved as an intermediate result and can be reused with different material properties in the same model. These CPU intensive calculations need not be repeated when material properties change. Viewfactor checks for convergence of its numerical integration algorithms and increases or decreases the integration order as appropriate. This provides excellent performance as measured by the product of accuracy and speed. Diagnostic data is provided to aid in verifying the accuracy of an analysis. Viewfactor provides support for radiation to an ambient or space environment node. Convex surfaces may be flagged to help reduce execution time.

Main Index

4

Patran Viewfactor Analysis Features and Benefits

Nonobstructing surfaces may be flagged in order to reduce the time required to check for obstructed views.

Main Index

Chapter 1: Introduction 5 About this Guide

1.3

About this Guide This Guide contains a complete description of Patran Thermal’s Viewfactor code. The thermal phenomena that this module is designed to model are technically complex. Great care has been taken to present the material contained herein in a manner that is clear and easy to understand. Numerous examples throughout the text illustrate the subject matter.

Main Index

6

Patran Viewfactor Analysis Using this Guide

1.4

Using this Guide The material in this document is presented with certain assumptions about the knowledge and abilities of the user. It is recommended that the user obtain and become familiar with the other pertinent documents listed at the end of this section. This Guide makes frequent reference to these documents, or to material described more fully in them.

Assumptions About the User This document was written under the following assumptions: • The user is familiar with Patran and can make finite element models in Patran. If not sufficiently versed in the use

of Patran, the user may refer to the Patran Reference Manual or attend an MSC Institute course. • The user is familiar with MSC.Software Corporation (MSC) product Patran Thermal and is able to perform

thermal analysis using Patran and Patran Thermal. If not, the user may wish to refer to the Patran Thermal documentation and/or attend the MSC Institute course on Patran Thermal or obtain the video cassette course on Patran Thermal. • The user is familiar with the computer environment in which this software will be used, and is able to manipulate

files, manipulate directories, and edit files. • The user has experience and/or education equivalent to a BS in engineering with an emphasis on thermal analysis.

If the user is a novice thermal analyst, it is suggested that course work or self-directed study be undertaken. It is not the purpose of this Guide to make the user of Viewfactor a thermal analyst. The ideas presented and used in this Guide will not be easily understood by one who does not understand thermal analysis. • Some of the material in this document is aimed at the thermal analysis expert.

Other Pertinent Documents • Patran Reference Manual, Volumes 1 through 3. • MSC Patran Thermal User’s Guide. • Gebhart, B. Heat Transfer, 2nd edition, McGraw-Hill, 1971. • Howell, J.R. A Catalog of Radiation Configuration Factors, McGraw-Hill, 1982. • Siegel, R., and Howell, J.R. Thermal Radiation Heat Transfer, 2nd edition, McGraw-Hill, 1981.

Main Index

Chapter 1: Introduction 7 Guide Organization

1.5

Guide Organization The contents of this Guide are organized into four subject groupings. Each group is described in the following paragraphs, along with the chapters or appendices associated with each group. The four groups are: • Introduction and Overview • The Analysis Cycle • Theory and Specifications • Examples

In addition, there are appendices at the end of the document which provide supplemental information on a number of topics. The Guide also has a comprehensive index to aid the reader in locating information on particular topics.

Main Index

8

Patran Viewfactor Analysis Overview of Viewfactor Analysis

1.6

Overview of Viewfactor Analysis Chapter 1 - Introduction, contains mostly nontechnical information about this Guide, the Viewfactor module, and the relationship between the Viewfactor code, Patran Thermal and Patran. If you are unfamiliar with the conventions and structure of the Patran System product documentation, you should read Introduction (Ch. 1) before attempting to use the other chapters of this Guide. Chapter 2 - Overview, describes, at a high level of abstraction, the program structure, data flow, and program execution sequence for the Viewfactor product. An understanding of this overview will provide a reference frame within which to organize and associate the technical information contained in this document and to relate it to Patran and P⁄THERMAL. The last section of this chapter reviews the analysis cycle for a thermal radiation problem using the Patran System products:Patran and Patran Thermal. The next group of chapters, The Analysis Cycle, is summarized here. The Analysis Cycle is fairly long and complicated, and therefore the new user may find the Summary of the Analysis Cycle for a Thermal Radiation Problem, 20 to be an excellent introduction and overview of Chapter 3 through Chapter 7. If you are unfamiliar with the overall thermal analysis cycle using the Patran System products Patran and Patran Thermal, you should first read Overview (Ch. 2). Chapters 3-7 - The Analysis Cycle, deal with the details of thermal analysis as it relates to the thermal radiation boundary condition and its subsequent modeling and analysis using P⁄THERMAL,and Patran. Other aspects of the analysis cycle, such as general model creation, other thermal boundary conditions, material properties, thermal network analysis, and thermal results postprocessing, are described in more detail in Volume 1 of the MSC Patran Thermal User’s Guide and the Patran Reference Manual. Please refer to these other documents for information relating to those parts of the thermal analysis cycle not pertaining directly to the thermal radiation boundary condition. Unfortunately, the analysis of complex thermal phenomena is itself complex. The software described here was designed to model complex phenomena, such as temperature and/or time dependent emissivities and transmissivities. Therefore, the analysis cycle is more complex than it would be if only simple phenomena were being modeled. This complexity is the price for the richness of the thermal modeling and highly nonlinear system solution capabilities available in this software. This complexity will not go away. For the analyst wishing to model the simpler phenomena, the complexity generally reduces to simple steps in the analysis cycle. With this in mind, it is advisable to obtain a thorough understanding of the cycle presented here and then become proficient at using the parts applicable to your thermal analysis needs. Chapters 8-10 - Theory and Specifications, deal with the following topics: • Formulae and methods used to compute the radiation viewfactors. • Computational limitations brought on by the finite precision of computers. • Rules governing the generation of Patran Thermal radiation resistors. • Format specifications for the data files (VFINDAT, VFRAWDAT, VFRESDAT, TEMPLATEDAT, VFDIAG, and

VFNODEDAT). Generally, this information will only be needed by the engineer concerned with the limitations of the computer algorithms and who wishes to obtain a more complete understanding of what is being done in the program, or wishes to interface the Viewfactor data files to other software. MSC.Software Corporation does not guarantee that the data file formats will remain unchanged. Appendices - Examples, Appendix E contains examples of thermal radiation analysis problems presented as complete analysis cycles. These examples were designed to present the important features and capabilities of the Viewfactor code, as well as some of the advanced features of Patran Thermal as they pertain to analysis of thermal radiation problems. The problems are generally simple in other aspects and thus are easy to model and not too time-consuming. Once you have gained sufficient understanding of the analysis cycle, a quick review of the example problems will refresh your memory after you’ve been away from this software for a period of time. Some of the examples were also designed to demonstrate the correctness of the thermal analysis and to build confidence in the use of the Viewfactor code. Other topics covered in the appendices are error conditions, memory requirements, and technical support.

Main Index

Chapter 1: Introduction 9 Nomenclature

1.7

Nomenclature Certain documentation conventions (such as different typefaces having meaning), special characters with specific meaning (such as slashes in Patran commands), technical definitions, and symbols used in this document are defined or described in this section.

Conventions Font Types and Typefaces Computer messages or responses are printed in plain character format: SYSTEM RESPONSES ARE IN A PLAIN LETTER STYLE Commands (or queries) that you enter are printed in bold typeface which is darker and heavier than normal or plain text: commands entered by the user are in bold type Some fields, or parts of a surface, may be optional. Fields contained within brackets [ ] are optional for data input. Generally, optional data fields that do not receive input will default to a predetermined value. VFAC,TID[,NBANDS] Filenames A generic file naming convention is used in this guide. This reduces confusion for users on the various computer platforms and operating systems supported by MSC.Software Corporation. Generic file names contain no delimiters. A file referred to as “filenameELS” in this Guide would appear as “filename.ELS” for VAX, Apollo, Celerity, Hewlett-Packard, Data General, SGI, Prime and Cray. It would appear as “filename ELS” for the IBM VM/CMS, and as “filename_ELS” for the CDC NOS/VE.

Units As you are proceeding with your modeling tasks in Patran and Patran Thermal, remember that they are unit-less or dimensionless. That is to say they will accept as input any number and it is your responsibility to make sure that the units you are using are consistent. Typically, the type of units to be used is defined or locked in when you choose your material properties. These units for your material properties must be consistent with the other dimensioned quantities within the model, such as length.

Main Index

10

Patran Viewfactor Analysis Nomenclature

Main Index

Chapter 2: Overview Patran Viewfactor Analysis

2

Main Index

Overview



Purpose



Relationship of Viewfactor to Patran and Patran Thermal



Viewfactor Data and Program Flow



Summary of the Analysis Cycle for a Thermal Radiation Problem

12 13

16 20

12

Patran Viewfactor Analysis Purpose

2.1

Purpose This chapter provides an overview of performing a Viewfactor analysis. It describes how the Viewfactor code is related to the other Patran products (Patran and Volume 1 of Patran Thermal) showing the high level data and program flow in Viewfactor. It summarizes the typical analysis cycle for a thermal radiation problem using Patran.

Main Index

Chapter 2: Overview 13 Relationship of Viewfactor to Patran and Patran Thermal

2.2

Relationship of Viewfactor to Patran and Patran Thermal The Viewfactor code was designed primarily to support and enhance the thermal analysis capability in P⁄THERMAL. Although Viewfactor is a stand-alone executable, we will be primarily concerned with its use in conjunction with Patran and Patran Thermal. Viewfactor was designed to work closely with the Patran Thermal module, and thus uses many of the same files as Patran Thermal.

Description The relationship of the Viewfactor code to Patran and Patran Thermal is shown schematically in Figure 2-1. Patran Thermal’s PATQ takes the geometric and boundary condition data from a Patran neutral file and converts it to data about the thermally radiating surfaces. The data is then output to the VFINDAT file. The TEMPLATEDAT file is the Patran Thermal template file, with the addition of a new template, called VFAC (Viewfactor). The template data is necessary

Main Index

14

Patran Viewfactor Analysis Relationship of Viewfactor to Patran and Patran Thermal

for Viewfactor to make thermal network resistors for Patran Thermal. The Viewfactor run is a noninteractive process whose execution is controlled by data in the VFCTL file.

USER INPUT

Patran CREATE MODEL

DISPLAY RESULTS

NEUTRAL FILE

NODAL RESULTS QOUTDAT

Patran Thermal VFRESTXT

PATQ

QTRAN

OTHER THERMAL FILES

TEMPLATEDAT

VFRESDAT

VFINDAT VFNODEDAT

VFCTL

Viewfactor Calculate Viewfactors

Make Resistors VFRAWDAT

VFDIAG Figure 2-1

VFMSG

Viewfactor Relationship to Patran Thermal and Patran

Viewfactor transforms the data in the VFINDAT file into viewfactor data, which is output to the intermediate file VFRAWDAT. This intermediate file permits you to change the surface properties in the TEMPLATEDAT file and generate new thermal network resistors without having to redo the computationally expensive viewfactor calculations. The intermediate raw viewfactor data file also makes possible the use of Viewfactor to generate just viewfactor information for use other than interfacing directly to Patran Thermal. Next, the raw viewfactor data in VFRAWDAT and the information in the surface template file, TEMPLATEDAT will be combined by Viewfactor to make thermal network resistors and radiosity nodes for Patran Thermal. The resistors will be output in the binary file VFRESDAT and the radiosity nodes will be put out in the VFNODEDAT file. These two files will

Main Index

Chapter 2: Overview 15 Relationship of Viewfactor to Patran and Patran Thermal

be input by Patran Thermal and used in the thermal analysis of the problem. The results of the thermal analysis are output by Patran Thermal in the file QOUTDAT and nodal results files. These files may be displayed along with the geometric model using the postprocessing capabilities of Patran. There are several other files shown in Figure 2-1 which we have not yet discussed. These are VFRESTXT, VFDIAG, and VFMSG. These files do not participate in the computer analysis of the problem. They are provided to assist you in determining that the problem is correctly modeled and the analysis has been correctly performed. In the event you have an error, they will be helpful in finding and correcting it. VFRESTXT is a text version of VFRESDAT. Since VFRESDAT is stored in binary form, it cannot be read by most computer file editors. The capability to translate the binary VFRESDAT file into a text file, VFRESTXT, is provided in Patran Thermal’s PATQ. You may then examine the thermal network resistors generated by Viewfactor using the text editor of your choice. VFRESDAT files tend to be large and are best implemented in binary form, which is more compact than text form. You will find, in most cases, that the VFRESTXT file is too large to be examined in detail. The files VFDIAG and VFMSG contain information useful in evaluating a Viewfactor execution. VFMSG predominately contains text information concerning the progress of the Viewfactor program execution and reports of any errors which were detected. You are strongly advised to examine the VFMSG file for error messages, since this is the only way to know if errors occurred. The VFDIAG file contains predominately numerical data relating to the sums of viewfactors to each surface. This data can often be compared to expected values for the sums of viewfactors and thus used to judge the correctness of the viewfactor analysis.

Main Index

16

Patran Viewfactor Analysis Viewfactor Data and Program Flow

2.3

Viewfactor Data and Program Flow This section describes in a very general manner the internal workings of the Viewfactor code. Strictly speaking, it is not necessary to know this information in order to use Viewfactor. However, by knowing something about Viewfactor’s internal data flow it is easier to understand some of the information needed in order to use Viewfactor. Therefore, while it is not necessary to dwell on the details of this section, it is good to be familiar with it. You may also wish to refer to this section from time to time to refresh your memory or to facilitate your understanding of how some detail of the viewfactor analysis fits into the overall Patran System Thermal Analysis scheme.

Description Figure 2-2 shows a high level abstraction of the program structure contained in Viewfactor. The program functions are

described in outline form, with the level of indentation representing the level of nesting in the program. Data files are shown in ovals with arrows to the general portion of the program where the data is input or output. The file VFMSG receives output throughout the program execution and thus does not have an arrow from a specific portion of the program.

Main Index

Chapter 2: Overview 17 Viewfactor Data and Program Flow

I. Initialize III. Make Patran Thermal Resistors Input the Control Data Input VFAC Template Data from the TEMPLATEDAT File VFCTL Validate the Control Data Validate the Template Data Initialize Internal Data Sort the Template Data Open the Appropriate Files Input Raw Viewfactor Data; Make Resisto TEMPLATEDAT II. Calculate Viewfactors as We Proceed Input the Model Data and Input the Title, Size, and Symmetry Dat Calculate Viewfactors as We Input and Process the Node Data Proceed For Each Enclosure Input Title Data VFINDAT Input the Enclosure Data Input Size Data Associate the Surface User ID (UID Input Symmetry Data with a Template ID (TID) Input and Process Node Data Make the Radiosity Nodes For each Enclosure Make and Output the Emissivity VFRAWDAT Input Enclosure Data Resistors Process Enclosure Data For Each Surface Pair in this Enclo For Each Surface Pair in this Input Raw Viewfactor Data for th Enclosure Surface Pair Check Self Shadowing VFRESDAT Check Consistency and Check for Obstructed View Compatibility of the Calculate Viewfactors and Surface Pair Data Mean Beam Distances Make the Radiation Resistors Output the Raw Viewfactor End of Surface Pair for Loop Data for this Surface Pair VFNODEDAT Sort, Merge and Output the Resisto End of Surface Pair End of Enclosure for Loop for Loop Output the Radiosity Node Data to End of Enclosure for Loop VFNODEDAT File Close the VFINDAT File VFDIAG Close the Data Files IV. Exit If this is an Abnormal Termination, Attemp VFMSG Clean up and Exit Gracefully Close the VFMSG File Set the Status Flag Stop Figure 2-2

High Level Data and Program Flow for Viewfactor

The execution of Viewfactor is controlled by parameters in the control file VFCTL. These parameters serve three general purposes: 1. Set program parameters, such as the value of the parameter used for convergence checking.

Main Index

18

Patran Viewfactor Analysis Viewfactor Data and Program Flow

2. Specify file names other than the default names for the data files. 3. Control which parts of the Viewfactor program are executed and subsequently which input data files are required and which output data files are created. The third function of program control is described here. All of these functions are described in more detail in Viewfactor Execution From Patran Thermal, 79. Referring to Figure 2-2, you should be able to identify the following program parts: Part I

Initialize.

Part II

Calculate Viewfactors.

Part III

Make Patran Thermal Resistors.

Part IV

Exit from the Viewfactor program structure.

Parts I and IV are always executed. Through the use of a parameter in the VFCONTROL file you may cause any one of three execution modes to occur. MODE 1

Main Index

In the first mode, all of the Parts I through IV are executed. The required data input is a VFCTL file, a VFINDAT file, and a TEMPLATEDAT file. The output produced is a VFRESDAT file, a VFNODEDAT file, a VFRAWDAT file, a VFDIAG file, and a VFMSG file. This mode takes in the geometric description of the radiating surfaces and their Patran Thermal surface template data and creates as output resistor network data for Patran Thermal. Also created as output for possible later use (see the description of the third mode below) is the raw viewfactor data. Diagnostic data is also output.

Chapter 2: Overview 19 Viewfactor Data and Program Flow

MODE 2

In the second mode, only Parts I, II, and IV are executed. Part III (Make Patran Thermal Resistors) is not executed and thus no thermal network data is generated. The TEMPLATEDAT file is not required for this mode, although no harm will be caused by its presence. The required data input is a VFCTL file and a VFINDAT file. The output produced is a raw viewfactor file, VFRAWDAT, and the diagnostic files VFDIAG and VFMSG. The VFRAWDAT file produced here may be used in the third mode described in the next paragraph. The second mode is useful if you only want to generate viewfactor data and do not care about the Patran Thermal network resistors. It is also useful if you do not yet have the TEMPLATEDAT file describing the surface properties and wish to begin the viewfactor calculations. You must take care to make sure that the thermal radiation problem described in the VFINDAT file and the property data identified in the yet to be created TEMPLATEDAT file are compatible. This is described in more detail in Compatibility Requirements for Model and VFAC Templates, 65 and Introduction, 120. The intermediate file VFRAWDAT and the TEMPLATEDAT file may be combined to make the thermal network resistors at some later time by using the third mode.

MODE 3

In the third mode, only Parts I, III, and IV are executed. Part II, (Calculate Viewfactors) is not executed and thus there must already be in existence and available to the program a data file of raw viewfactor data, VFRAWDAT. This mode also requires a TEMPLATEDAT file of surface data for the Patran Thermal resistors that will be created. The VFCTL file is also input in this mode. The output created here is the thermal resistor network for the radiating surfaces, contained in the files VFRESDAT and VFNODEDAT, and the diagnostic data contained in the files VFDIAG and VFMSG. This third mode allows you to change the surface property definition by changing the information contained in the TEMPLATEDAT file. Then run this mode of the Viewfactor program again. Note that in this way you may generate a new and different thermal resistor network simply by changing the TEMPLATEDAT file. You do not have to rerun the computationally expensive viewfactor calculations which were already performed in the first or second modes described above. This provides great savings of computer time in cases where the geometry does not change, but you wish to run two or more thermal analyses using different radiative surface properties. It is also useful for performing initial analysis using simpler material properties (e.g., constant properties). Once the analyst is satisfied that the problem is correctly modeled, the material properties may be changed to more closely represent reality (e.g., temperature dependent properties). By submitting simpler, computationally faster models for preliminary analysis the analyst can optimize the use of available computer resources and improve overall performance. When using this mode you must take special care to define the radiating surfaces in such a way that they are capable of supporting all of the various material property definitions you plan to attach to each surface in the future. This method is described in more detail in Patran Thermal TEMPLATEDAT Files for Surface Property Description, 52 and Introduction, 120.

Main Index

20

Patran Viewfactor Analysis Summary of the Analysis Cycle for a Thermal Radiation Problem

2.4

Summary of the Analysis Cycle for a Thermal Radiation Problem Thermal phenomena tend to be complex physical processes and the analysis of these processes by digital computer is equally complex. The tools provided by MSC.Software Corporation in the form of Patran and P⁄THERMAL provide advanced capabilities to model some very complex thermal problems. While great effort has been taken to make the tools simple and easy to use, the very complexity of the thermal analysis problem necessitates some degree of complexity in the software provided to perform the analysis. This section provides an overview of the generic thermal analysis cycle using Patran and Patran Thermal. Since this Guide deals with Viewfactor analysis, this summary emphasizes parts of the cycle peculiar to the analysis of a problem containing thermal radiation.

Problem Definition The first step is to define the problem. This includes identifying the geometry, boundary conditions, materials, material properties and approximations to be used in the analysis. You may find it useful and efficient to outline the entire analysis procedure as it pertains to the problem at hand. This should help avoid unpleasant surprises later on. It also ensures that all the steps in a complex process are followed.

General Preprocessing This step involves creating or inputting the geometric model, creating a finite element mesh on the model, and assigning boundary conditions and material identifications, in Patran. This step requires close coordination with the next step, General Patran Thermal Preparation, so that the boundary conditions and material properties identified in Patran correspond to material definitions and boundary conditions in the supporting Patran Thermal files. These activities and entities are described more fully in the MSC Patran Thermal User’s Guide. Planning at this stage of the analysis is important if you wish to be able to easily change boundary conditions and/or material property definitions in the future.

General Patran Thermal Preparation The model of the thermal analysis problem built with the Patran preprocessor generally has only identification numbers attached to boundary conditions and material properties. These identification numbers are used by the Patran Thermal module to point into various databases and files for the actual data describing the boundary condition or material property. If these files do not already exist, they must be created. Details concerning these files are contained in the MSC Patran Thermal User’s Guide.

Thermal Radiation Specific Preprocessing Thermal radiation problems analyzed with the Patran System of products require some specific preprocessing in Patran. You must identify the material surfaces which are participating in the radiation interchange and identify the radiative properties of these material surfaces. The VFAC LBC form has been introduced into Patran specifically to facilitate the modeling of thermal radiation problems. The VFAC form provides support for basic thermal radiation boundary conditions, for participating absorbing and emitting media between surfaces, for identifying convex surfaces which cannot radiate to themselves directly, for identifying surfaces that are not obstructions, and for radiation to an ambient node. This Patran form is described in detail in Advanced Features of the VFAC Boundary Condition, 42. In support of these modeling capabilities, you will also need to enter data into the Patran Thermal files for the surface emissivity properties and the participating media (if any) extinction or transmissivity properties, including in both cases waveband data if applicable. This data is typically entered into the Patran Thermal TEMPLATEDAT, MATDAT, and MICRODAT files. These files and data relevant to a thermal radiation problem will be briefly described in Patran Thermal TEMPLATEDAT Files for Surface Property Description, 52. For full details on these files, refer to the MSC Patran Thermal User’s Guide. Facilities have also been programmed into Patran Thermal and Viewfactor to accept and process information about symmetry occurring in the thermal analysis problem. Problem symmetry is also input in Patran at this stage of the analysis cycle. The use of symmetry in thermal radiation problems is described in more detail in Symmetry as Applied to the Model and Viewfactor Radiation Exchange, 67.

Main Index

Chapter 2: Overview 21 Summary of the Analysis Cycle for a Thermal Radiation Problem

Preparation for Viewfactor Analysis After the model of the problem to be analyzed has been prepared in Patran and the required supporting Patran Thermal files have been created, there are two steps to prepare the problem for processing by Viewfactor. These are: 1. Translate the model description contained in the Patran neutral file to the data and form required for the Viewfactor input file VFINDAT, and 2. Create a VFCTL file which will direct the execution of Viewfactor. The translation of the Patran neutral file is done using a menu pick from Patran Thermal’s PATQ menu and is described in detail in Preparation for Analysis (Ch. 4). The VFCTL file is typically created using your editor. It is about 20 lines of identifying keywords and associated parameter values. This file is described in Viewfactor Execution From Patran Thermal, 79. Note:

The VFCTL file is automatically created when the analysis is submitted from the Analysis form in Patran.

Viewfactor Analysis Viewfactor will usually be executed as a noninteractive batch process. Merely invoke the command procedure to submit Viewfactor and its control file, VFCTL, for execution, or select “Execute Viewfactor Analysis” in the Analysis / Submit Options form. Since Viewfactor analysis tends to be computationally expensive, review all aspects of the model carefully before beginning the viewfactor analysis. This will help to minimize the number of Viewfactor analyses submitted with incorrect or incomplete data. Viewfactor has some data checking and error detection capabilities, but it cannot detect all user errors. The procedure for submitting a Viewfactor job is described in detail in Submitting a Viewfactor Job for Analysis, 98. Viewfactor will create a number of output files. The files created depend on some parameters in the VFCTL file. The various files created as Viewfactor output are described in Viewfactor Data and Program Flow, 16 and Output Created by a Viewfactor Execution, 101. When the viewfactor analysis is completed, the Viewfactor diagnostic files, VFDIAG and VFMSG, should be reviewed for acceptable diagnostic data values and possible error messages, as described in Reviewing the Viewfactor Output, 103.

Post Viewfactor Analysis After the Viewfactor analysis is complete and the rest of the Patran Thermal input files are complete, the user is ready to perform the thermal network analysis using Patran Thermal. Viewfactor will have created two files to which Patran Thermal must be given access. These are the radiation resistor file, VFRESDAT, and the radiosity node file, VFNODEDAT. This access is usually provided by giving Patran Thermal the names of these files through the Patran Thermal QINDAT file. The QINDAT file is described in the MSC Patran Thermal User’s Guide. The particular aspects of the QINDAT file relevant to the viewfactor analysis and the Viewfactor files VFRESDAT and VFNODEDAT are described in Interface From Viewfactor to Patran Thermal, 111. At this point in the analysis cycle you may also translate the binary radiation resistor data file, VFRESDAT, into a text file which you may examine. This capability is provided by a menu pick in Patran Thermal’s PATQ and is described in Interface From Viewfactor to Patran Thermal, 111. This text file has no other purpose in the analysis. It is provided merely for your convenience. The viewfactor portion of the analysis is now complete. The remaining steps in the analysis cycle all concern general thermal analysis.

Patran Thermal Analysis The procedure for submitting a Patran Thermal analysis is described in the MSC Patran Thermal User’s Guide. Briefly, the process involves generating some FORTRAN source code for the particular problem, compiling the new source code, linking with the Patran Thermal QTRAN run-time library, and submitting the job for execution. The results of the thermal

Main Index

22

Patran Viewfactor Analysis Summary of the Analysis Cycle for a Thermal Radiation Problem

analysis will be contained in the Patran Thermal QOUTDAT file and in nodal results files for use with the Patran postprocessing tools.

Postprocessing The capabilities of Patran and the Patran Thermal interface permit analysis results to be displayed and examined quickly and efficiently. For more information about thermal results postprocessing, refer to the Patran Thermal User’s Guide. Refer to the Patran Reference Manual for general postprocessing information.

Refinements After examining the analysis results, you may be satisfied with the analysis, in which case this analysis cycle terminates. You may wish to refine or modify the computer model of the problem and perform the analysis again, in which case the analysis cycle starts over and repeats itself as applicable.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem Patran Viewfactor Analysis

3

Main Index

Model Creation for a Thermal Radiation Problem 

Purpose



Radiation Enclosure Concept

25



Surface Orientation in Patran

31



Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary Condition) 38



Advanced Features of the VFAC Boundary Condition



Relationship of VFAC LBC Data to VFINDAT File Data



Patran Thermal TEMPLATEDAT Files for Surface Property Description



Compatibility Requirements for Model and VFAC Templates



Symmetry as Applied to the Model and Viewfactor Radiation Exchange

24

42 51 52

65 67

24

Patran Viewfactor Analysis Purpose

3.1

Purpose This chapter presents the concepts, processes, and commands that describe the thermal radiation specific attributes of heat transfer analysis problems being modeled in the Patran System. Most of this chapter deals with preprocess model building in Patran. The concept of radiation enclosure is specific to thermal radiation analysis and is described in Radiation Enclosure Concept, 25. The concept of surface orientation, while not unique to thermal radiation analysis, has not been introduced previously in MSC.Software Corporation thermal analysis tools, and is described in Surface Orientation in Patran, 31. Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary Condition), 38 and Advanced Features of the VFAC Boundary Condition, 42 deal with identifying the radiative boundary conditions and other associated information using the Patran Radiation Boundary condition. Relationship of VFAC LBC Data to VFINDAT File Data, 51 explains the relationship of the Viewfactor LBC form to the data in the VFINDAT input data file for Viewfactor. Patran Thermal TEMPLATEDAT Files for Surface Property Description, 52 and Compatibility Requirements for Model and VFAC Templates, 65 discuss the Patran Thermal TEMPLATEDAT file and the associated VFAC template data. In the last section of this chapter, page 67, the role of symmetry in thermal radiation problems is discussed. The method for specifying the existence and type of symmetry in a problem is presented.

Some sophisticated thermal analysis tools have been provided and thus this chapter contains a large amount of information. Take the time to understand all of the capabilities of the tools available here. This will enable you to make informed decisions regarding how to model the thermal radiation phenomena at hand and choose the appropriate tools for the analysis. You must understand the concept of radiation enclosure, Radiation Enclosure Concept, 25. If you wish to model materials or media with wavelength dependent properties, then Surface Orientation in Patran, 31 must be understood. Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary Condition), 38 deals with basic radiative boundary conditions, while Advanced Features of the VFAC Boundary Condition, 42 deals with move advanced features that enable modeling of participating media, radiation to ambient nodes, and methods for reducing CPU time required for a Viewfactor analysis. If you are also responsible for the accompanying Patran Thermal analysis, then Patran Thermal TEMPLATEDAT Files for Surface Property Description, 52 andCompatibility Requirements for Model and VFAC Templates, 65 regarding the TEMPLATEDAT files and VFAC templates are strongly recommended. Be cautious using symmetry in thermal radiation problems, since the thermal radiation boundary conditions have subtle ways of making what appears to be a symmetric problem actually nonsymmetric. However, if you must make use of symmetry, Symmetry as Applied to the Model and Viewfactor Radiation Exchange, 67 should be thoroughly mastered.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 25 Radiation Enclosure Concept

3.2

Radiation Enclosure Concept The radiation enclosure concept is fundamental to the analysis of thermal radiation problems and also to the techniques used to model the problems in Viewfactor and Patran Thermal. Therefore, it is important to understand the concept, not only as it is classically applied to thermal radiation problems, but also as it is used in creating the computer model of the thermal radiation phenomena.

Definition of Enclosure For our purposes, an enclosure is a collection of thermally radiating surfaces which have the potential to see each other (radiate to each other), along with open areas which can potentially be seen by the surfaces and participating media or ambient nodes associated with these surfaces. From this definition you may infer that there are a large number of enclosures possible in even a simple model. It is up to you to select appropriate enclosures for the particular thermal analysis problem at hand. In most cases, appropriate choices of enclosures are natural and obvious from the model geometry. Surfaces in different enclosures do not have the potential to radiate to each other. In addition, a surface in one enclosure does not have the ability to obstruct the view between a pair of surfaces in another enclosure. These properties of enclosures are exploited in the Viewfactor program to reduce the CPU time required to analyze the viewfactor problem. Surfaces not in an enclosure need not be considered as potential obstructions for that enclosure. Surface pairs that are not in the same enclosure need not have calculations done for them. The enclosures are also used for defining portions of the model over which the viewfactors from one surface to all other surfaces it sees are summed. These sums have two uses: 1. For diagnostic purposes, and 2. To determine the viewfactor to that portion of the enclosure that is not represented by real surfaces, but instead is open to space.

The Enclosure ID Enclosures are made distinct by giving each different one a unique identification number. This ID number is associated with all of the surfaces in its enclosure. The enclosure ID is assigned in the VFAC LBC form as described in Viewfactor (p. 125) in the Patran Thermal User’s Guide Volume 1: Thermal/Hydraulic Analysis.

Wavebands and Enclosures Wavebands are used when the thermal radiative material properties depend on the wavelength of the radiation (they may also depend on time and temperature). The concept of wavebands and their proper use in modeling the thermal analysis problem are explained in Patran Thermal TEMPLATEDAT Files for Surface Property Description, 52. If you do not need to use spectrally dependent material properties to adequately model the problem, then this subsection need not be understood. Note, however that you will need to understand this subsection in order to understand all of the examples in the next subsection, list2+s of the Use of Enclosures, 26. Enclosures and wavebands (see page 52) have a special relationship. The wavebands associated with each surface in an enclosure must match exactly the wavebands of every other surface in that enclosure which the first surface can see. This holds true when every surface in the enclosure has the same wavebands. It is recommended that all surfaces within an enclosure have the same wavebands. For two surfaces in an enclosure, if the surfaces can see each other and the wavebands are not identical, then a fatal error will occur when you attempt to make radiation resistors for the model. This error cannot be detected until after the viewfactors are calculated (the most CPU intensive part of the Viewfactor analysis). It is a mistake that you should avoid so that you do not have to redo the viewfactor calculations. For the wavebands associated with two surfaces to be identical is meant: • The number of wavebands for each surface must be the same, and • The lower limit of each waveband on each surface must be the same and

Main Index

26

Patran Viewfactor Analysis Radiation Enclosure Concept

• The upper limit of each waveband on each surface must be the same and • The wavebands for each surface must be input in the same order in the TEMPLATEDAT file.

list2+s of the Use of Enclosures Figure 3-1 through Figure 3-6 show schematically some examples of enclosures and their use in modeling thermal radiation problems. Figure 3-1 shows a rectangular cross section of either a hollow torus in axisymmetric space or of a long tube in Euclidean

space. The section is hollow on the inside and the interior surfaces are thermally radiating, as indicated by the arrows attached to these surfaces. These surfaces are in the same physical enclosure. It is best to use the naturally occurring enclosure as the enclosure for the computer model. It does not make sense to divide these interior surfaces into two enclosures. Some pairs of surfaces would then be in different enclosures and no viewfactors would be calculated for them. This would not be correct, since in this model all of the interior surfaces can see each other.

Enclosure 1

Denotes Thermally Radiating Surface. Figure 3-1

Solid with Hollow Interior and Thermal Radiation in the Interior

Figure 3-2 shows a cross section with two cavities. Each cavity is filled with a different participating media, and thus there

is a different participating media node for each cavity. Two enclosures are needed to keep separate the two different media. Different enclosure ID numbers are assigned to each cavity. If the media in each cavity had been the same (or if there had been no participating media), it would have been acceptable to give both cavities the same enclosure ID. This is not preferred, because the cavity surfaces naturally fall into two groups.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 27 Radiation Enclosure Concept

Any surface in one group cannot see any other surface in the other group. By identifying these group through different enclosure IDs the CPU time required to analyze the viewfactors will be reduced.

Enclosure 1

• Media Node 1

Enclosure 2

• Media Node 2

Denotes Thermally Radiating Surface. Different Arrowheads are Associated with Different Enclosures. Figure 3-2

Example Showing Model with Two Cavities which Naturally Correspond to Two Enclosures

Figure 3-3 shows an object which is exposed to ambient radiation nodes above the object and below. Due to fortuitous geometric circumstances in the model, this object could actually be modeled as one enclosure. This is not recommended. A better approach would be to divide the model into two enclosures. One consists of the upward facing cavity and surfaces and the second consists of the downward facing cavity and surfaces. The astute reader may also observe that the model can be properly divided into even more enclosures. Recognize that the horizontal surfaces not in the cavities see nothing but an ambient node. Thus each of these surfaces could be identified as a unique enclosure. Doing this in a model where each of these surfaces was divided into many elements would result in substantial saving of CPU time to perform the viewfactor analysis.

• Ambient Node 1

Enclosure 1

Denotes Thermally Radiating Surface. Different Arrowheads are Associated with Different Enclosures.

Main Index

Enclosure 2

• Ambient Node 2

28

Patran Viewfactor Analysis Radiation Enclosure Concept

Figure 3-3

Example Showing Model with Two Open Cavities which Naturally Correspond to Two Enclosures

Figure 3-4 shows an example where three enclosures have been identified. Other groupings of the surfaces, media node,

and ambient nodes are possible. The reader who wishes to master the art of identifying enclosures in the thermal analysis model is urged to devise some other groupings of the surfaces and nodes into enclosures and then ascertain their correctness for modeling this problem. For the correct enclosure groupings, what are the advantages and disadvantages?

Enclosure 1

• Media Node

Ambient Node 2



Enclosure 3

Enclosure 2 Ambient Node 1



Denotes Thermally Radiating Surface. Different Arrowheads are Associated with Different Enclosures.

Figure 3-4

Example Showing the Use of Three Enclosures to Group Surfaces, Media and Ambient Nodes

Figure 3-5 shows an object made of two different materials, each having different wavelength dependent surface emissivity

properties. This cavity is correctly modeled as one enclosure, but the wavebands in the enclosure must be the composite of all the waveband transitions for the various materials’ wavebands. Thus, there are six wavebands in this enclosure and

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 29 Radiation Enclosure Concept

each material will have to be described in terms of these six wavebands, not in terms of the two wavebands of surface one, for example.

Material 1

Material 2 Denotes Thermally Radiating Surface

1

Wavebands for Material 1

ε1,λ 0 λ0 1

λ3

Wavebands for Material 2

ε2,λ 0 λ0

λ1 λ2

λ4

λ5

λ6 = ∞ Wavebands for Enclosure

λ0 Figure 3-5

λ1 λ2 λ3

λ4

λ5

λ6 = ∞

Example Showing the Wavebands in an Enclosure Having Materials with Different Wavebands

Figure 3-6 is an introduction to the kinds of complex problems which may be modeled with enclosures and wavebands. It consists of a cavity surrounded by two different materials, each with different wavelength dependent emissivities, and a partition vertically through the center of the cavity and at the boundary between the different wall materials. This partition is transparent to some wavelengths and opaque to others as shown in the transmissivity graph. For the spectral region where the partition is transparent, the cavity may be modeled as one enclosure, since the radiant interchange is unimpeded by the partition which is transparent to radiation in this spectral region. For the spectral region where the partition is opaque, the right and left halves of the cavity cannot see each other, but the surfaces in each half can see the right and left faces of the partition, respectively. Thus we model the right and left half cavities along with their associated face of the partition each as an enclosure. The wavebands for these enclosures are shown at the lower left of the Figure 3-6.

Main Index

30

Patran Viewfactor Analysis Radiation Enclosure Concept

Note that the wavebands for enclosure 1 end at lambda sub 4, and the wavebands for enclosures 2 and 3 begin at lambda sub 4, the transition wavelength between transparency and opaquecy for the partition material. Semi-Transparent Partition Material 1

Material 2

Denotes Thermal radiating surface in: Enclosure 1 Enclosure 2 Enclosure 3 τ3,λ Transmissivity of Partition Material λ4

ε1,λ

Wavebands for Material 1 λ0

λ3

ε2,λ

Wavebands for Material 2

λ0

λ5

λ2

λ6

λ7 = ∞ Wavebands for Enclosure 1

λ0 λ1 λ2 λ3 λ4 Wavebands for Enclosure 2 λ7 = ∞

λ4

Wavebands for Enclosure 3 λ4 Figure 3-6

Main Index

λ5

λ6

λ7 = ∞

Example Showing the Use of Enclosures and Wavebands to Model a Cavity with a Partition Transparent in One Waveband and Opaque in Another Waveband

Chapter 3: Model Creation for a Thermal Radiation Problem 31 Surface Orientation in Patran

3.3

Surface Orientation in Patran Patran does not check for consistently oriented surfaces in two-dimensional entities such as patches and quadrilateral elements. Properly oriented surfaces are required for correct modeling of 2-D XY and 2-D axisymmetric thermal radiation models. Since Patran does not ensure consistent surface orientation, the user must assume this responsibility. Failure to do so will result in erroneous models.

The Importance of Surface Orientation The computation of viewfactors requires knowledge of the normal to the surfaces for which the viewfactors are being computed. In addition, we must also know from which side of the surface the radiation is emanating. In 3-D models this is not a problem since Patran consistently uses only one orientation of the parametric axes, called the right hand orientation. However, for 2-D models, Patran allows either clockwise or counterclockwise or a combination of both orientations for surfaces. Note that this orientation is with respect to the global coordinate system, and not the local element coordinate system. Viewfactor uses information gained from an assumed orientation of the surfaces to determine from which side of a surface the radiation emanates. Surfaces in two and three dimensions are shown in Figure 3-7 with different surface normal orientations.

n

n

n

n Figure 3-7

Two and Three Dimensional Surfaces with Different Surface Normal Orientations

Viewfactor assumes that all coordinate systems are oriented in a right hand (also known as counterclockwise) arrangement (i.e., if you place your right hand so that the fingers point from the first coordinate axis to the second coordinate axis, then the raised thumb points up out of the paper in the positive direction of the third coordinate axis). To apply this test to cylindrical coordinate systems, the fingers of the right hand should be pointed in the direction of increasing polar angle. The term “counterclockwise” comes from the fact that for right-handed systems viewed from the

Main Index

32

Patran Viewfactor Analysis Surface Orientation in Patran

front, the direction of rotation from the first axis (e.g., x-axis), to the second axis (e.g., y-axis), is counterclockwise. Sketches illustrating these concepts are shown in Figure 3-8. z

9

10

11

12 1

8

2

7 6

5

3

4

y

x

Rotation from x to y same as direction of right hand fingers and thumb points in direction of z. Rotation from x to y is counterclockwise with clock face pointing in z direction.

z (second axis)

θ (polar angle)

r (first axis)

Right hand direction from r rotated to z by Right Hand Rule. Figure 3-8

Orientation of Coordinate System

InViewfactorthe direction of the surface normal for 3-D surfaces is determined by the right hand rule as you point your fingers in the direction determined by the order in which the surface corner nodes are given. Thus, if the normal to a surface is pointing toward you, the order of the nodes describing that surface will appear counterclockwise. For 3-D models, Patran

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 33 Surface Orientation in Patran

automatically takes care of this and you do not need to be concerned with it. The relationship of 3-D surface orientation and node ordering is illustrated in Figure 3-9. Third Node

n

First Node

Second Node

First Node

n Second Node

Third Node Third Node

n

Second Node Fourth Node

First Node Figure 3-9

Relationship of Node Order to Surface Normal for 3-D Surfaces

For 2-D (either Cartesian or axisymmetric) models, you must take care to correctly orient the surfaces. The term “surface” here generically means a boundary of the model. For objects modeled in 2-D space surfaces are represented as lines. The orientation of these lines in Viewfactor is determined by the order in which their beginning and ending points are given. You imagine yourself walking in the plane of the model, feet on the plane and on the side determined by the right-hand orientation. Then as you walk from the beginning point to the ending point of the line, the principle or positive

Main Index

34

Patran Viewfactor Analysis Surface Orientation in Patran

normal direction is that which the right arm points when extended horizontally. The relationship of node order to normal direction and this method of determining normal direction for 2-D surfaces (lines) is illustrated in Figure 3-10.

(Z)Y Boundary or Surface of Object Beginning Node of Left Boundary Right Arm Extended

Direction of Walk Object

n

Ending Node of Left Boundary

X(R)

Figure 3-10

Normals for Boundaries of 2-D Objects, Node Ordering and the “Right Arm Rule”

Patran allows you to arbitrarily mix left handed and right handed oriented systems for 2-D entities such as patches and two-dimensional elements. This is not allowed when using Viewfactor. Since facilities to automatically manage orientation of 2-D entities are not currently available, you must take care of this task. Failure to do so will result in erroneous 2-D models. The following three subsections contain information to help manage this task.

Determining Surface Orientation There are several ways to show the orientation of patches and two-dimensional elements in Patran. Keep in mind that these tests for determining orientation are for a view from above or in front of the plane containing the model. “Above” or “in front of” means with respect to a right-handed coordinate system for the plane. The parametrization of a geometric entity is not displayed as default. The parametrization of geometric entities can be displayed by clicking on display of the top level menu bar and then selecting geometric and turning on the parametric direction button. Several patches with their C1 parametric directions are shown in Figure 3-11, along with an indication of right-handed (desired) or left-handed (undesired) orientation. Basically, if viewed from above the plane containing the model (above

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 35 Surface Orientation in Patran

with respect to a right-handed system), the C1 parametric direction should point in the counterclockwise direction around the perimeter of the patch for properly oriented patches.

Right

Left

Right Right

Right

Left Left

Left

Figure 3-11

Right and Left Handed Oriented Patches and their C1 Parametric Directions

The next test may be used to determine a patch’s orientation. They are not recommended because they do not conform to the engineering and mathematical standards for surface orientation. These tests are based on the order of the corner grid IDs for the patch and on the order of the edges of the patch. Patran specifies the corner grid and edge order of right hand oriented patches to be clockwise, whereas most users will be familiar with the usual counterclockwise orientation. The patch corner grid ordering may be shown by clicking on geometry in Patran and selecting Action: show, Object: surface, Method: attribute and selecting the surface. When the spreadsheet comes up, click on the vertices button to see the corner grids and their order. Look at the graphics window to see if the grids in this order go clockwise around the patch. If they do, then this is a properly oriented patch.

Main Index

36

Patran Viewfactor Analysis Surface Orientation in Patran

The ordering of corner grids and edges is shown in Figure 3-12 for various right-handed (properly) and left-handed (improperly) oriented paths. 2

2

3

4

4

3

1

3

2 1

3

RIGHT

RIGHT

4

3

1

RIGHT

2 4

1 1

4

4 1 3

4

3

3

2

LEFT

2 1

2 3

RIGHT

1

4

4

4 1

1

1 4

3

4

4

4

3 2

4

2

LEFT

3

LEFT

1

3

3 2 2 3

2

2

2

RIGHT

3

2

1

1

2

2 3 1

4

2

3

1 LEFT

4

4

x

Grid Numbers

x

Edge Numbers

1

C1 Parametric Directions Figure 3-12

Corner Grid and Patch Edge Ordering for Right- and Left-Hand Oriented Patches

The orientation of two-dimensional elements (i.e., quadrilaterals and triangles), may be determined by using the element verification menu. Since elements are typically much more numerous than patches, take care to properly orient all patches before beginning to generate elements to ensure that there are no improperly oriented elements. The commands listed above for determining the element orientation are described fully in the Patran Reference Manual. Note, however, that the ordering for nodes on a properly oriented element is counterclockwise as is customary in engineering analysis, and that this is opposite of the ordering of corner grids on a properly oriented patch.

Correcting Improper Surface Orientations Improperly oriented patches may be reversed with the modify action on the geometry menu.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 37 Surface Orientation in Patran

The patch may also be deleted and a new properly oriented patch created in its place, or the patch may be overwritten with a properly oriented patch with the same patch ID. Please see the Patran Reference Manual for information on the various patch menus. Elements may be reversed with the AUTOREVERSE option of the NORMALS submenu of the VERIFY action on finite elements form. These forms are all described in the Patran Reference Manual. You must exercise caution when reversing elements since Patran may not correctly transform the boundary conditions associated with an element edge when the element is reversed. Other data associated with the element may not be transformed correctly either. Note:

You should not reverse any elements which have LBC element properties associated with it.

In most cases, if you have left-handed elements, it is recommended that the Patran finite element entities be deleted. Then the orientation problems should be corrected at the patch level before any finite element entities are generated.

Suggested Practices for Creating Properly Oriented Surfaces It is important that elements in two-dimensional models (Cartesian and axisymmetric) are correctly oriented. Be aware of the problem and plan carefully to avoid it. Then, as each patch is made, its orientation should be checked and verified to be correct. Improperly oriented patches should be reversed. This may be done with the modify action of geometry menu. Carefully check to make sure all patches are properly oriented before beginning to generate the finite elements, finite element properties, and LBCs. Note:

Axisymmetric models only Since handedness (left- or right-hand rule for orientation) is determined relative to the direction (from the back or from the front) you view the model, righted handedness will appear left handed when viewed from the back. Some commonly used axisymmetric coordinate systems present a view from the back. These systems will require that you use left-handed oriented patches and elements instead of right-handed ones as explained in The Enclosure ID, 25, Determining Surface Orientation. To determine which one to use for your axisymmetric coordinate system, perform the following test. Form the cross product of the r-axis with the z-axis. Use the right-hand rule with your fingers pointing from the raxis to the z-axis and observe the direction of your thumb. Your thumb will point in the direction of the cross product, either out of the screen or into the screen. If the direction is out of the screen, use right-handed patches and elements. If the direction is into the screen, use left-handed patches and elements.

Main Index

38

Patran Viewfactor Analysis Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary Condition)

3.4

Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary Condition) The Patran Viewfactor Loads/BCs option was specifically implemented in Patran, under the Patran Thermal preference, to provide support for Viewfactor thermal radiation problems. Basic features of the form are described in this section. Advanced features are described in the following section.

Purpose of the Viewfactor Form The Viewfactor form is used: 1. To identify surfaces in the model which will participate in thermal radiation interchange; 2. For these surfaces to identify certain properties relevant to thermal radiation transfer; 3. To provide information on radiation enclosures, participating media nodes, ambient nodes, convex surfaces, and nonobstructing surfaces. The surface properties are identified only by a pointer to a VFAC template ID, on the VFAC LBC input data form. This template ID points to material property data in the Patran Thermal data files. It must correspond to a template ID number, TID, in the template data file, TEMPLATEDAT.

Form for the VFAC LBC This is a technical description of the Viewfactor form. For more information, refer to Viewfactor (p. 125) in the Patran Thermal User’s Guide Volume 1: Thermal/Hydraulic Analysis. Input Data for the Viewfactor Form Input Data VFAC TEMPLATE ID

Description The user function ID, UID, identifies the VFAC template in P⁄THERMAL’s TEMPLATEDAT file which will be used to identify the material properties associated with this surface. These properties include surface emissivity and participating media transmissivity data. This parameter is required and is entered as an integer. In Patran Thermal and Viewfactor only positive UIDs are valid. Patran does not check for nonpositive UIDs and so it is up to you to observe this restriction. A nonpositive UID causes an error in Patran Thermal and Viewfactor.

MEDNOD

This parameter identifies the participating media (if any) node by its node ID number. The default value is 0 (zero), indicating that no media node is present. If this referenced node ID number changes as a result of optimization, equivalencing, or node renumbering in Patran, the corresponding reference in the VFAC record will not be automatically updated to match this change. So it is up to you to make sure that the node ID for the media node has not changed.

AMBIENT

This parameter identifies the ambient or space node (if any) by its node ID number. The default value is 0 (zero), indicating that no ambient node is present. This is the node which radiation escaping the enclosure will reach and which will represent the ambient radiation from space for this surface. If this referenced node ID number changes as a result of optimization, equivalencing, or node renumbering in Patran, then the corresponding reference in the VFAC record will not be automatically updated to match this change.

NODE ID

CONVEX SURFACE ID

Main Index

The convex surface ID, CNVSID, is used to identify convex surfaces in an enclosure. This is used to reduce computer time for the Viewfactor raw viewfactor calculations. A convex surface is one for which no point on the surface has a direct line of sight view of any other point on the surface. For our purposes, plane surfaces may be considered convex. Note that the scope of the CNVSID is confined to the present enclosure and thus the user may reuse convex surface ID numbers in different enclosures without adverse effects. The default value is 0 (zero), indicating that this is not a convex surface. Special care must be taken in axisymmetric and 3-D models to make sure that saddle-like surfaces are not mistakenly thought to be convex.

Chapter 3: Model Creation for a Thermal Radiation Problem 39 Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary

The value of 1, causes the nonobstruction flag to be set. This means to Viewfactor that this surface is not capable of obstructing the view between any other pair of surfaces in this enclosure, including the view between this surface and other surfaces. This facility provides the option to reduce Viewfactor calculation time by identifying the nonobstructing surfaces in an enclosure.

OBSTRUCTION FLAG

TOP/BOTTOM FLAG

The bottom surface flag, “1”, is used when applying VFAC boundary conditions to the bottom surface (not edges) of quadrilateral, triangular, or bar elements. In these cases, VFAC DFEGs can be applied either to the top or bottom, the default being the top. The bottom is selected when the character “1”is present in the TOP/BOTTOM FLAG data box. The bottom surface flag has no meaning for solid elements or for the edges of quadrilateral and triangular elements, and is ignored. The top of a quadrilateral or triangular element is defined by the right-hand rule. For bar elements, the beam orientation is used. The top of bars points towards the beam orientation, the bottom points away. See the Patran Reference Manual for more information on beam orientation. Entering the correct enclosure ID is critical to the proper performance of Viewfactor. There is no way for Patran, Patran Thermal or Viewfactor to check the correctness of the enclosure ID. Exercise extreme care in this regard.

ENCLOSURE ID

Note: This is required even if there is only one enclosure. Examples of a Viewfactor LBCs applied to a 2-Dand 3-D model are shown in Figure 3-13 and Figure 3-14, respectively. 13

14

15 8

7

U777,0,0,0

U79,0,0,0 10

11

12 6

5

U777,0,0,0

U79,0,0,0 7

8

9 4

3

U777,0,0,0

U79,0,0,0 4

5

6 2

1

U777,0,0,0

U79,0,0,0 Y 1

Z

Figure 3-13

Main Index

2

X

Example VFAC LBC on Edges of a Patch

3

40

Patran Viewfactor Analysis Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary Condition)

Y

X

Z

Figure 3-14

U308,0,0,0

list2+ VFAC DFEGs on a Hyperpatch

Requirement for Oriented 2-D Surfaces Related to the VFAC LBC Patch and element orientation is important for two-dimensional planar and axisymmetric models with regards to the VFAC and surface orientation for calculating viewfactors. Surface orientation and its importance to the thermal radiation model is explained in detail in the previous section of this chapter. This is a very important aspect of two-dimensional models in Patran, and if you are not familiar with the material in Surface Orientation in Patran, 31 and plan to model two-dimensional planar or axisymmetric models, then you should study Surface Orientation in Patran, 31.

Neutral File Data Packet Created from the VFAC LBC This information is provided to assist you should you desire to examine the model data in the Patran neutral file or wish to interface to this data in some other manner. If you do not wish to do this, then this subsection need not be read. VFAC LBC data will be output in the neutral file on packet 19. The sequence of the packet 19s will be the same as for all other data packets (see the Patran Reference Manual). The form of packet 19 is as follows:

Header Card ITC

Main Index

Format ID

IV

(I2,8I8) KC

ITC

19

ID

Element ID

IV

Enclosure ID

KC

1

N2

UID

N2

Chapter 3: Model Creation for a Thermal Radiation Problem 41 Specifying Radiation Boundary Conditions Using Patran Reference Manual (VFAC Boundary

Data Card 1 MEDNOD

Main Index

Format AMBNOD

CNVSID

(6I8, 2X, 8I1) OBSTR DYN

SURF NODE(8)

MEDNOD

Participating media node

AMBNOD

Ambient node

CNVSID

Obstruction Flag (0 or 1)

DYN

Dynamic Flag (0 or 1)

SURF

Surface (0=Top, 1=Bottom)

NODE (8)

8 element node flags (0 or 1)

42

Patran Viewfactor Analysis Advanced Features of the VFAC Boundary Condition

3.5

Advanced Features of the VFAC Boundary Condition The Viewfactor boundary condition in Patran was designed to support modeling a media participating in the thermal radiation interchange. This interchange could take place between two surfaces or between a surface and an ambient or space node representing the background thermal radiation conditions. The participating medium is assumed to be isothermal and gray within each waveband and to be diffusely emitting and absorbing without photoactive effects. The VFAC boundary condition also provides support for some capabilities in Viewfactor which are used to reduce the CPU time required to calculate the viewfactors. This support comes in the form of the CNVSID and nonobstruction FLAG parameters in the VFAC LBC form which are used to identify convex surfaces and nonobstructing surfaces, respectively, in Patran. These features of the VFAC boundary condition are described in this section.

Referencing Participating Media Radiation Nodes The participating medium is assumed to be at a uniform temperature, gray in a waveband, and diffuse. It is assumed to be weakly absorbing (i.e., the exact exponential absorption function may be accurately replaced with a linear approximation). It is assumed that the medium does not obscure itself or the other surfaces. Role of Radiation Participating Media in Patran Thermal The Patran Thermal user is free to assign the temperature or heat flux to the participating media node as a simple constant or in more complicated situations as functions of time and temperatures of other nodes. The participating media node will be treated as any other node in the thermal resistor network. Participating Media Resistor Networks When the participating media node is present, the Viewfactor code makes three resistors instead of the usual one resistor between the surface subareas. One of the resistors will still be between the surface subareas. But it will have been modified to account for the portion of the radiant energy which will be absorbed as it traverses between the two surfaces. The second and third resistors will be between the surface subareas for each surface and the participating media node. They will account for the radiant interchange between the surfaces and the participating media. Refer to the MSC Patran Thermal User’s Guide for an explanation of resistor networks and subareas. Two resistor networks are shown in Figure 3-15, one without a participating medium and one with a medium.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 43 Advanced Features of the VFAC Boundary Condition

Node 2

1- ε2 ε2 A2

Radiosity Nodes (typ)

1- ε4 ε4 A4

1 F24 A2

Node 4

1

Surface 2

Surface 1

F23 A2

1 F14 A1

1- ε1

1- ε3 ε3 A3

1 F13 A1

ε1 A1 Node 1

Node 3 Thermal Network Resistor (typ)

Network Without Participating Media 1

1

1

F23 A2 τ23

F14 A1 τ14

1

Surface 1

F24 A2 (1-τ24)

1- ε1 ε1 A1

Node 1

Media Node

Main Index

Node 4

1 F42 A4 (1-τ42)

1

1

F23 A2 (1-τ23)

F41 A4 (1-τ41)

1

1

F14 A1 (1-τ14)

F32 A3 (1-τ32)

1

1

F13 A1 (1-τ13)

F31 A3 (1-τ31)

1 F13 A1 τ13)

Network With Participating Media Figure 3-15

1- ε4 ε4 A4

Thermal Resistor Networks With and Without Participating Media

Surface 2

Node 2

1- ε2 ε2 A2

F24 A2 τ24

1- ε3 ε3 A3

Node 3

44

Patran Viewfactor Analysis Advanced Features of the VFAC Boundary Condition

Defining the Participating Media Node The participating media node must be defined in the Patran model before it can be referenced in the VFAC LBC form. If the participating media node does not already exist, it may be created with the Patran Finite Element form:

Finite Element Action:

Create

Object:

Node

Method:

Edit

In the node location list box, point to a convenient grid point. Note:

The exact location of this node is not important. For clarity, you may want to use a conspicuous node ID number such as 1000, 9999.

Referencing Ambient or Space Radiation Nodes Frequently in thermal radiation analysis we encounter a problem which has an enclosure with an opening to space or to the ambient environment. In order to facilitate modeling this situation, Viewfactor and Patran Thermal provide for ambient thermal radiation nodes. The Patran Viewfactor boundary condition provides support for entering this part of the model through the Patran preprocessing. Role of Ambient Radiation Nodes In the present analysis scheme for a given enclosure, each surface may see at most one ambient node. This implies that each surface in the enclosure sees an isothermal ambient environment. If there is a participating media present, it is assumed to exchange radiant energy with the ambient environment also. The distance from the ambient node to the participating media node is not well defined. Therefore, this combination cannot be used with Patran Thermal resistor types which calculate the media absorption and emittance from Beer’s Law of Absorption (i.e., resistor subtypes 7, 8, 11, and 12). Suggested Practices to Improve Model Accuracy There is a way to produce a more accurate model of the thermal radiation interchange between a surface and the ambient radiation through an opening in the enclosure. This is done by covering the openings in the enclosure with surfaces having the temperature of the ambient environment. The viewfactor from the surface to the ambient environment is calculated by taking one minus the sum of the viewfactors to all other surfaces that the first surface can see. This summing process introduces the possibility of cumulative errors. By filling the enclosure openings with ambient surfaces, the error is only the error occurring in the viewfactor calculation from the first surface to the ambient surface. This method does in general require more CPU time, but this increase in CPU time will be relatively small if the ambient surfaces needed to close the open enclosure are few compared to the total number of surfaces in the enclosure. This method also has another advantage. There are no restrictions on the use of Patran Thermal radiation resistors which use Beer’s Law to calculate the interaction with the participating media (see Role of Ambient Radiation Nodes). This method allows you to have nonuniform ambient temperatures and to model the radiative surface properties of the ambient environment. Some examples of open enclosures and how they might be closed by ambient surfaces are shown in Figure 3-16.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 45 Advanced Features of the VFAC Boundary Condition

Surface at Ambient Temperature

Ambient Node

Enclosure Enclosure

Enclosure 2 Ambient Node 2

Figure 3-16

Ambient Node 1

This Region at Node 1 Temperature

Enclosure 1

This Region at Node 2 Temperature

Ambient Surface Enclosing the Model

Converting Open Enclosures into Closed Enclosures

Ambient Node Resistor Networks in Patran Thermal Whenever there is an ambient node present in an enclosure and the sum of the viewfactors from a surface to all other surfaces it can see in that enclosure is not one, then an Patran Thermal radiation resistor is created from that surface to the ambient node. In addition, if there is a participating media node and the previous resistor to the ambient node was made, then a resistor will be made from the participating media node to the ambient node. A simple example network is shown in Figure 3-17 for an enclosure with participating media and an ambient node. Note the presence of the media transmissivity, τ , in the resistors to the ambient node. Since the distance to the ambient node is not well defined, this

Main Index

46

Patran Viewfactor Analysis Advanced Features of the VFAC Boundary Condition

transmissivity cannot be calculated from Beer’s Law and thus Patran Thermal radiative resistor subtypes 7, 8, 11, and 12 are not permitted in an enclosure with both a participating medium and an ambient node. 1 (1-ΣjF2j) A2 τ2j

1

1

F24 A2(1-τ24)

(1-ΣjF2j) A2(1-τ2j)

Media Node

1- ε1 ε1 A1

Node 1

1- ε4 ε4 A4

1 F24 A4 τ24

Node 4

Ambient Node

Surface 2

1- ε2 ε2 A2

Surface 1

Node 2

1- ε3 ε3 A3

Node 3

Radiosity Node (typ)

Figure 3-17

Only Typical Resistor Values are Shown

A Simple Resistor Network for an Enclosure with Participating Media and an Ambient Node

Defining the Ambient Node The ambient node must be defined in the Patran model before it can be referenced in the VFAC boundary condition. If the ambient node does not already exist, it may be defined with the Patran finite element create menu. See example in Figure 3-18.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 47 Advanced Features of the VFAC Boundary Condition

Upper Plate

Ambient Node

Fixed Temperature Ambient Surface

Lower Plate Figure 3-18

Schematic Diagram of an Example of Ambient Node

Identifying a Surface as Being Convex Since the calculation of viewfactors is computationally expensive, we have provided the user with the ability to identify convex surfaces in the thermal radiation portion of the model. A convex surface is one for which no pair of points on the surface can see each other by direct line of sight. Thus the exterior of a sphere and the exterior of a cylinder are convex surfaces. Plane surfaces are also considered to be convex. Thus, for example, the faces of a cube are all convex, and the entire exterior of a cube is convex, while the interior of the cube has six separate convex surfaces, one for each face. Caution must be exercised when specifying convex surfaces in axisymmetric models (see Caveats Regarding Convex Surfaces in Axisymmetric Models, 48. Significance of Convex Surfaces to Viewfactor Calculations To calculate viewfactors, each surface must be paired with every other surface in the enclosure and a determination made regarding whether there is a direct line of sight view between pairs of points on these two surfaces, one point on each surface. By the nature of convex surfaces, we know a priori that a pair of surfaces, element faces in our case, on a convex surface do not have a direct line of sight view of each other. Benefits of Identifying Convex Surfaces to Viewfactor Execution Time This determination of whether one surface has a line of sight view of another surface from geometrical considerations is computationally significant. If the determination can be made simply by comparing convex surface IDs (a computationally trivial test) a significant amount of CPU time can be saved on problems which have a large number of element faces on convex surfaces in an enclosure. This requires that the convex surfaces be identified by the user before they are submitted to Viewfactor for analysis. This of course is not required in order to run Viewfactor. You will have to decide whether it is an economical use of your time to identify convex surfaces in order to save some computational time. People seem to be more efficient than machines at

Main Index

48

Patran Viewfactor Analysis Advanced Features of the VFAC Boundary Condition

identifying convex surfaces and are able to group large numbers of element faces into convex surfaces merely by looking briefly at the model. Caveats Regarding Convex Surfaces in Axisymmetric Models In certain situations, you must take special care not to identify as convex surfaces which are not really convex. A simple example of this is the exterior of a torus, which appears convex when the axisymmetric model is drawn, but in reality, due to its double curvature, is partially not convex. This example is illustrated in Figure 3-20. Such situations are common in axisymmetric models and also can occur in three-dimensional models. Viewfactor has no way to check the correctness of convex surface identification and so the user must take care not to make mistakes. When in doubt, it is best not to use the convex surface ID. Convex Surface

Main Index

Figure 3-19

Two Concentric Spherical Shells, Axisymmetric Model

Figure 3-20

Torus and its Axisymmetric Model with Nonconvex Outer Surface Shaded

Chapter 3: Model Creation for a Thermal Radiation Problem 49 Advanced Features of the VFAC Boundary Condition

Identifying a Surface as Not Obstructing the View Between Other Surface Pairs In calculating viewfactors, we must check to see if the view between two surfaces is obstructed or blocked by one or more other surfaces in the enclosure. In the worst case, this requires checking all of the surfaces in the enclosure as potential obstructions between every pair of surfaces in the enclosure. Anything that can be done to reduce the number of surfaces that must be considered as potential obstructions of the view between surface pairs is of interest to the user who has a limited amount of CPU time available. Benefits of Identifying Nonobstructing Surfaces to Viewfactor Execution Time By setting the nonobstruction flag, you can indicate that a surface cannot obstruct the view between any pair of surfaces in the enclosure. Viewfactor checks the nonobstruction flag. If the flag is set for a surface, that surface is not included in the potential obstruction list. Thus the number of potential obstructions to be checked is reduced and CPU time is conserved. Examples with Nonobstructing Surfaces Identified In the example shown in Figure 3-18, none of the surfaces can obstruct the view between any other pair of surfaces. Thus all of the surfaces with VFAC boundary condition may have their nonobstruction flag set for the lower plate. For the example shown in Figure 3-19, the inner surface of the outer spherical shell does not obstruct the view of any other pair of surfaces in the enclosure and we may set the nonobstruction flag for the VFAC boundary condition on element faces on this surface. Caveats Regarding Nonobstructing Surfaces in Axisymmetric Models Nonobstructing surfaces may be particularly difficult to identify in axisymmetric models. You may wish to forgo the use of the nonobstruction flag for axisymmetric models. The nature of the difficulty is illustrated in Figure 3-21 with a simple example. The object being modeled is a solid cylinder surrounded by an annulus and a larger, hollow cylinder. Referring to the axisymmetric model of the object in the figure, it appears at first inspection that there are not obstructing surfaces in the model. However, a top view of the object, as seen in the figure, reveals that the solid cylinder does indeed obstruct the view between parts of the outer cylinder.

Main Index

50

Patran Viewfactor Analysis Advanced Features of the VFAC Boundary Condition

View from 1 to 2 is obstructed by the solid cylinder.

2

1

Axisymmetric Model Figure 3-21

Top View of Actual Objects

Solid Cylinder Inside Hollow Cylinder with Annular Space

This type of situation can be difficult to correctly identify in more complicated axisymmetric models. You are urged Surface to exercise caution.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 51 Relationship of VFAC LBC Data to VFINDAT File Data

3.6

Relationship of VFAC LBC Data to VFINDAT File Data The Viewfactor boundary condition data is output from Patran in the neutral file packet 19s. The neutral file is read by Patran Thermal’s PATQ and translated into the Viewfactor input file VFINDAT. The VFINDAT file is described in detail in Chapter 9. The relationship of data in the neutral file to data in the VFINDAT file is described in general terms. The VFINDAT file begins with title data for the problem, followed by data about the size of the problem. Next is symmetry data which is discussed in Compatibility Requirements for Model and VFAC Templates, 65. All of the nodes in the Patran model are then copied to the VFINDAT file, including their IDs and coordinates. The surfaces with VFAC boundary conditions are then grouped into their respective enclosures and put out to the VFINDAT file grouped by enclosure. Some of the data in the VFAC record is copied directly. This data is the enclosure ID, the user function ID or TID, the participating media node or MEDNOD, the ambient node or AMBNOD, and the convex surface ID or CNVSID. The nonobstruction flag is translated to 0 or 1 and copied. The element ID with which each surface is associated is in the packet 19 and also copied to VFINDAT. The VFINDAT file also requires other information about the surface. This is developed by PATQ from data about the model from other parts of the neutral file. This data includes information about the shape of the surface, the order of interpolation functions needed to describe the surface, the number of nodes associated with the surface, an ID for the surface, an identification of the element face represented by this surface, the node IDs associated with this surface, and possibly some other data associated with the surface.

Main Index

52

Patran Viewfactor Analysis Patran Thermal TEMPLATEDAT Files for Surface Property Description

3.7

Patran Thermal TEMPLATEDAT Files for Surface Property Description Patran Thermal supports an arbitrary finite number of wavebands for spectral dependent material properties for thermal radiation interchange. Within each waveband the material properties are assumed to be independent of wavelength (gray) and diffuse. Completely gray surfaces are also supported without the added work on your part to specify the waveband as the entire spectrum. This section describes the data needed by Viewfactor in the form of the Patran Thermal TEMPLATEDAT file and VFAC template to make the Patran Thermal thermal network resistors for modeling thermal radiation interchange.

Thermal Radiation Wavebands as Used in MSC Patran Thermal User’s Guide If you only plan to use gray surfaces, then the information on thermal radiation wavebands and wavelength dependent network resistors may be ignored. If the wavelength dependence is not specified by entering the wavebands, then Viewfactor assumes that the enclosure is gray and you need not be concerned with wavebands. Thermal radiation wavebands are defined in the Viewfactor code for use by P⁄THERMAL in terms of their beginning and ending wavelengths in units of microns. The spectrum begins at zero wavelength and extends to infinity. Since infinity is not a convenient quantity with which to work in computers, we will use some finite, but large number to represent the upper end of the spectrum, for example 1.0E10 microns. In general, the thermal radiation above this wavelength is completely negligible for engineering problems. (This may not be true in certain physics problems.) Patran Thermal evaluates the black body function in each waveband and at the temperature of each surface. This is an improvement over other methods which use some mean temperature between the surface pairs (e.g., geometric mean, for the black body temperature). The heat flow between two surfaces 1 and 2 as represented by the Patran Thermal network equations is

Q

(t)

1⇒2

Z

nbands σ F ( λ i Ó 1, λi, T 1 ( t ), t ) T 41 ( t ) Ó F ⎛⎜ λi Ó 1, λi, T ( t ), t ) T 42 ( t ) 2 ⎝ ∑ -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------R ( λi Ó 1, λi, T 1 ( t ), T 2 ( t ), t ) iZ1

(3-1)

where: i

=

Waveband index

s

=

Stefan-Boltsmann constant

nbands

=

Number of wavebands

F

=

Black body function from

T1,T2

=

Temperatures of surfaces 1 and 2, respectively

R

=

Effective radiative resistance between surfaces 1 and 2, taking into account possibly time, temperature, and waveband

t

=

Time

l

=

Wave length

λ i Ó 1 to λ i

at temperature T

An example of spectrally-dependent surface emissivities for two surfaces is plotted in Figure 3-22. Also shown are the approximate wavebands and constant waveband properties that might be used to represent the surface properties of these

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 53 Patran Thermal TEMPLATEDAT Files for Surface Property Description

two surfaces. The approximating properties are shown as thin dashed lines. Refer to Wavebands and Enclosures, 25 for a discussion on the need for a consistent set of wavebands throughout an enclosure.

1

ε1,λ

0

λ

1

ε2,λ

0 λ0

λ λ1

λ2

λ3

λ4

λ5

λ6 = ∞

Actual Properties Waveband Approximation Six bands are used here to model these emissivities. The same bands must be used throughout an enclosure. Figure 3-22

Modeling Spectrally Dependent Properties

Patran Thermal’s QTRAN will accept overlapping wavebands and/or inactive or missing regions in the spectrum. This is both a blessing and a curse. It gives you the latitude to model surface properties with piecewise constant basis functions and leave out inactive regions of the spectrum from the analysis, but no checking is performed for a nonoverlapping and/or incomplete spectrum. Thus you must be responsible for the correctness of the waveband model and data.

Main Index

54

Patran Viewfactor Analysis Patran Thermal TEMPLATEDAT Files for Surface Property Description

Radiation Resistor Types Used in Patran Thermal This Guide deals specifically with radiation resistors. Refer to the Introduction (Ch. 1) in the Patran Thermal User’s Guide Volume 1: Thermal/Hydraulic Analysis for a complete discussion of Patran Thermal resistors. Patran Thermal’s radiation resistors are classified and identified according to their type and subtype. Two types are allowed: being gray (no spectral dependence) and waveband dependent (spectral dependencies as described in the previous section). The gray resistors are identified by the letter R and the waveband dependent resistors are identified by the letter W. Each type has a number of possible subtypes, the set of possible subtypes being identical for each type. This being the case, the subtypes are described without reference to the types and you may then associate the subtypes with the various types as desired. The different subtypes are identified by integer IDs. Viewfactor creates unformatted data records for the Patran Thermal radiation resistors it creates. These records are in the VFRESDAT file. Since the data is unformatted, it is not easily read by the user. The unformatted form is used to save space. For our discussions here, the data used to describe the resistors is represented in a form readable by the user. A Patran Thermal radiation resistor record consists of 11 pieces of data, not all of which are required for each resistor type and subtype. A radiation resistor is described by: RESTYP, SUBTYP, NODE1, NODE2, NODE3, MPID, DATA1, DATA2, DATA3, LAMBDA1, LAMBDA2 Input Data Description

Main Index

RESTYP

Resistor type, character, R or W.

SUBTYP

Resistor subtype, integer, 1 through 12.

NODE1

First node in the resistor record, integer.

NODE2

Second node in the resistor record, integer.

NODE3

Third node in the resistor record, integer (not always used, but must be present as a place holder).

MPID

Material property ID for Patran Thermal, integer.

DATA1

Data for use in calculating the resistor value, real.

DATA2

Data for use in calculating the resistor value, real (not always used, but must be present as a place holder).

DATA3

Data for use in calculating the resistor value, real (not always used, but must be present as a place holder).

LAMBDA1

Beginning wavelength of waveband, real (not used for type R resistors, but must be present as a place holder).

LAMBDA2

Ending wavelength of waveband, real (not used for type R resistors, but must be present as a place holder).

Chapter 3: Model Creation for a Thermal Radiation Problem 55 Patran Thermal TEMPLATEDAT Files for Surface Property Description

Radiation Resistor Subtypes Subtype 1

R = (1.0 - EPSILON) / (EPSILON * AREA)

This resistor is used between a gray surface and a radiosity node, with an emissivity, EPSILON, which is evaluated from a material property, MPID. If the material property is temperature dependent, it will be evaluated at the temperature of NODE1. Typically, NODE1 is the surface node and NODE2 is the radiosity node. The variable, AREA, is the area (nodal subarea) associated with NODE1. NODE3 is not used for this resistor. Subtype 2

R = 1.0 / ( FF * AREA * TAU )

This resistor is used between radiosity nodes and has a participating media transmissivity, TAU, evaluated directly from a material property, MPID. If the transmissivity material property is temperature dependent, it will be evaluated at the temperature of NODE3. Typically, NODE1 and NODE2 represent the radiosity nodes and NODE3 the participating media node. DATA1 contains the area, AREA, associated with NODE2. DATA2 contains the viewfactor, FF, from NODE2 to NODE1. If one of the radiosity nodes is an AMBNOD, then it will typically be NODE1. Currently, this subtype is not created by Viewfactor since subtype 9 is adequate for the present requirements and requires less computational time. Subtype 3

R = 1.0 / ( FF * AREA * ( 1.0 - TAU ) )

This resistor is similar to subtype 2, except this subtype is used to represent the radiant interchange between a radiosity node, NODE2, and a participating media node, NODE1. Typically, NODE3 and NODE1 are the same, but this is not required. Currently, this subtype is not created by Viewfactor since subtype 10 is adequate for the present requirements and requires less computational time. Subtype 4

R = 1.0 / ( DATA1 * DATA2 )

This is a general purpose resistor between NODE1 and NODE2 whose value is determined by multiplying two constants, for example the viewfactor and the area for a case with no participating media. NODE3 is not used for this resistor. Currently, this subtype is not created by Viewfactor since subtype 5 is adequate for the present requirements and requires less computational time. Subtype 5

R = 1 / DATA1

This is another general purpose resistor between NODE1 and NODE2 and is the simplest and computationally fastest resistor. It is useful whenever material properties are known constants and do not require access to the Patran Thermal material property data. Two typical uses are (1) between two radiosity nodes with DATA1 = FF * AREA * TAU, or (2) as an emissivity resistor between a non-black surface node and a radiosity node with DATA 1 = (EPSILON * AREA ) / ( 1.0 - EPSILON ). NODE3 is not used for this resistor. Subtype 6

R = ( 1.0 - DATA2 ) / ( DATA2 * DATA1 )

This is the constant known property version of subtype 1 where DATA2 is typically EPSILON and DATA1 is AREA. NODE3 is not used for this resistor. Presently, this subtype is not created by Viewfactor since subtype 5 is adequate for the present requirements and requires less computational time. Subtype 7

R = 1.0 / ( FF * AREA * TAU )

This resistor is used between radiosity nodes and has a participating media transmissivity, TAU, evaluated using Beer's law ( TAU = EXP( - KAPPA * DISTANCE ) ) and an extinction coefficient from a material property, MPID. If the extinction coefficient material property is temperature dependent, it will be evaluated at the temperature of NODE3. Typically, NODE1 and NODE2 represent the radiosity nodes and NODE3 the participating media node. DATA1 contains the area, AREA, associated with NODE2. DATA2 contains the viewfactor, FF, from NODE2 to NODE1. DATA3 contains the mean beam length from the surface associated with NODE2 to the surface associated with NODE1. If one of the radiosity nodes is an AMBNOD, then it will typically be NODE1. Currently, this subtype is not created by Viewfactor since subtype 11 is adequate for the present requirements and requires less computational time. Subtype 8

R = 1.0 / ( FF * AREA * ( 1.0 - TAU ) )

This resistor is similar to subtype 7, except this subtype is used to represent the radiant interchange between a radiosity node, NODE2, and a participating media node, NODE1. Typically, NODE3 and NODE1 are the same, but this is not required. Currently, this subtype is not created by Viewfactor since subtype 12 is adequate for the present requirements and requires less computational time. Subtype 9

Main Index

R = 1.0 / ( FA * TAU )

56

Patran Viewfactor Analysis Patran Thermal TEMPLATEDAT Files for Surface Property Description

This resistor is similar to subtype 2, but the FF and AREA have been combined by multiplication into FA and stored in DATA1. DATA2 is not used. Subtype 10

R = 1.0 / ( FA * ( 1.0 - TAU ) )

This resistor is similar to subtype 3. It is to subtype 3 what subtype 9 is to subtype 2. Subtype 11

R = 1.0 / ( FA * TAU )

This resistor is similar to subtype 7, but the FF and AREA have been combined by multiplication into FA and stored in DATA1. DATA2 in not used. Subtype 12

R = 1.0 / ( FA * ( 1.0 - TAU ) )

This resistor is similar to subtype 8. It is to subtype 8 what subtype 11 is to subtype 7.

Patran Thermal MPIDs (Material Property IDs) Refer to the Introduction (Ch. 1) in the Patran Thermal User’s Guide Volume 1: Thermal/Hydraulic Analysis for a complete discussion of MPIDs (material property IDs). If you choose to use MPIDs for the thermal radiation material properties, then these properties must either already be available in the material property data file or you will need to create the appropriate Patran Thermal material property specifications. Some examples are shown and discussed below as they would appear in the ‘mat.dat’ file.

***************************************************** MPID4000LCI_TABLEKELVIN 1.0 LINEAR CONSTAT INTERVAL TABLE FOR EMISSIVITY *SEE FIGURE 3-23 FOR A GRAPH OF THIS PROPERTY MDATA100.0 MDATA500.0 MDATA0.1 MDATA0.1 MDATA0.9 MDATA0.9 / ******************************************************

***************************************************** MPID3000TABLETIME1.0 TABLE FOR TIME DEPENDENT EMISSIVITY *SEE FIGURE 3-24 FOR A GRAPH OF THIS PROPERTY MDATA0.00.5 MDATA0.10.5 MDATA100.00.95 MDATA200.00.95 / ******************************************************

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 57 Patran Thermal TEMPLATEDAT Files for Surface Property Description

1.0

EMISSIVITY

0.8 0.6 0.4 0.2 0

0

400

800

1200

1600

TEMPERATURE Figure 3-23

Temperature Dependent Emissivity

1.0

EMISSIVITY

0.8 0.6 0.4 0.2 0

0

50

100

150

200

TIME Figure 3-24

Time Dependent Emissivity

Patran Thermal Material Property Definition The various ways to define material properties are described in the MSC Patran Thermal User’s Guide. Note that by referencing the MPID as a negative number in the TEMPLATEDAT template the material property will be evaluated as a function of time instead of as a function of temperature. In Patran, the material properties in Patran Thermal MPIDs (Material Property IDs), 56 are defined under fields:

Finite Element

Main Index

Action:

Create

Object:

Material Property

Method:

General

58

Patran Viewfactor Analysis Patran Thermal TEMPLATEDAT Files for Surface Property Description

Click on Input Data, then the desired Patran Thermal material function (e.g., "mpid_linr_tabl" will bring up the input data for a tabular input). Note:

Make sure you have selected Patran Thermal as the analysis preference.

VFAC Template Format The complete specification of the VFAC template is given in TEMPLATEDAT (Surface Pointer Data), 149. Unless the reader plans to interface to the Viewfactor module formally through another computer code, the description of the VFAC template in this section will be all that is needed. The VFAC template consists of a header line and one or more following data lines. Data on the header line will determine how many data lines must follow. Some of the data is optional, or optional depending on what other data is entered. All optional data defaults to standard values. If the value desired is the default value, then you need not enter the optional default values, as they will be assigned automatically. KEYWORD TID nbands The first line of the VFAC template, or header line, consists of three fields. These fields are: 1. The keyword VFAC, 2. Followed by the template ID number, TID, and finally 3. The number of wavebands and nbands used to define the thermal radiation properties. The first two fields are required and the third field is optional and defaults to zero. The TID and nbands must be integers. The fields should be separated by one or more spaces or by commas. TID

Valid TIDs are positive integers. TIDs will be associated with UIDs from the Patran Viewfactor LBC form.

nbands

Valid values of nbands are non-negative integers. If nbands is zero (the default), then this template will cause R-type resistors to be created; otherwise, W-type resistors will be created when this template is referenced.

The number of data lines following the header line must be exactly the number of nbands, except that if the nbands field is blank or zero there must be exactly one data line following. Comments may also be placed at the end of header or data lines by placing a semicolon after the last data field on the line. Some example header lines are: VFAC

99

which specifies that this is VFAC template 99 and represents a gray material property, and VFAC

1001

6

which specifies that this is VFAC template 1001 and represents a material with 6 wavebands. The data lines contain eight fields each. The fields are separated by one or more spaces or by commas. The first field, CONSTANT_EPSILON, is the only required field. Trailing blank fields may be omitted. Blank fields between nonblank fields are not permitted and will be ignored, thus corrupting the user’s data. The fields are: CONSTANT_EPSILON, CONSTANT_TAU, EMPID, TMPID, LAMBDA1, LAMBDA2, KFLAG, COLLAPSE

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 59 Patran Thermal TEMPLATEDAT Files for Surface Property Description

Input Data

Description

CONSTANT_EPSILON

This field is required and must be a real number. It is the value of the surface’s constant emissivity, or if the emissivity is not constant, then it must have the value 0.0. Valid values of emissivity are greater than 0.0 and less than or equal to 1.0. If the emissivity is not constant, then the material property ID which describes the property must be given in the EMPID field.

CONSTANT_TAU

This field is optional and defaults to 1.0. It is the value of the constant transmissivity of the participating media if any (the transmissivity of a vacuum is 1.0, hence the default value). If this value is not constant, then it must be given the value 0.0 here. Valid values of constant transmissivity are greater than 0.0 and less than or equal to 1.0. If the KFLAG is set to 1, then this value will represent the extinction coefficient for absorption according to Beer's Law and valid values are greater than 0.0. If the extinction coefficient is not constant, then this value must be set to zero. If this property, either transmissivity or extinction coefficient, is not constant, the material property ID which describes the property must be given in TMPID field.

EMPID

This field is an optional integer MPID (material property ID) and defaults to 0. It must assume its default value if a constant emissivity is specified by CONSTANT_EPSILON. If the emissivity is not constant, as indicated by the value 0.0, then EMPID must be nonzero. Positive values of EMPID denote temperature dependence, while negative values of EMPID will be evaluated as functions of time.

TMPID

This field is an optional integer MPID (material property ID) and defaults to 0. It must assume its default value if a constant transmissivity or extinction coefficient is specified by CONSTANT_TAU. If a constant is not specified, as indicated by its value of 0.0, then the TMPID must be nonzero. Positive values of TMPID denote temperature dependence, while negative values of TMPID will be evaluated as functions of time. The TMPID will evaluate to a transmissivity if the KFLAG is zero and to an extinction coefficient if the KFLAG is one, just as the CONSTANT_TAU does.

LAMBDA1, LAMBDA2

These are optional real fields, but either they both must be present or both not present. They are the beginning and ending wavelengths for the present waveband. Note that the wavebands do not necessarily have to be in order of increasing wavelength but must be in the same order for every surface in an enclosure. These fields default to 0.0, the value for the case when nbands is 0. If nbands is greater than 0 (i.e., the properties have spectral dependence), then LAMBDA2 should be greater than LAMBDA1, which should be greater than 0.0. The units used for wavelength are microns or micrometers.

Main Index

60

Patran Viewfactor Analysis Patran Thermal TEMPLATEDAT Files for Surface Property Description

KFLAG

This optional field signals whether the transmissivity is evaluated directly (KFLAG = 0), either from the constant value or from the MPID referenced in TMPID, or the transmissivity is evaluated using Beer’s Law and an extinction coefficient evaluated from either the constant value or from the MPID referenced by TMPID. The default KFLAG value is 0 and the data must be integer. Beer’s Law may be stated as: Transmissivity = EXP( - Extinction_Coefficient * Distance )

COLLAPSE

This optional field signals whether radiosity nodes associated with a given surface node should be collapsed into a single radiosity node. If COLLAPSE is zero (the default value), then the radiosity nodes will not be collapsed. The COLLAPSE_ID associated with a nodal subarea surface is transferred to that nodal subarea’s radiosity node. Then radiosity nodes connected to the same surface node by way of emissivity resistors and having the same nonzero COLLAPSE_ID will be collapsed into one node. The resulting parallel emissivity resistors will be merged if possible. The COLLAPSE_ID must be a non-negative integer. The main advantage of using COLLAPSE to collapse radiosity nodes is that this will result in a much smaller number of radiation resistors in the model. The effect of using COLLAPSE for small resistor networks is shown in Figure 3-25. The effect is more pronounced for larger networks or for 3-D networks. A smaller number of resistors usually means that the thermal analysis will proceed faster. In the best cases, the number of radiation resistors may be reduced by about a factor of four for 2-D Cartesian or axisymmetric models and by about a factor of 16 for 3-D models. The main advantage of using COLLAPSE to collapse radiosity nodes is that this will result in a much smaller number of radiation resistors in the model. The effect of using COLLAPSE for small resistor networks is shown in Figure 3-25. The effect is more pronounced for larger networks or for 3-D networks. A smaller number of resistors usually means that the thermal analysis will proceed faster. In the best cases, the number of radiation resistors may be reduced by about a factor of 4 for 2-D Cartesian or axisymmetric models and by about a factor of 16 for 3-D models. Our experience is that the loss of accuracy is quite small for fine meshes and lower temperatures. The user may wish to try the examples in Example Thermal Radiation Problems, 11, using the COLLAPSE field modeling techniques. Other existing models may also be rerun using the new COLLAPSE flag. Then the results can be compared with previous results and provide the user with a basis for deciding when to use or not use the COLLAPSE feature.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 61 Patran Thermal TEMPLATEDAT Files for Surface Property Description

Resistor Network without using COLLAPSE.

Resistor Network using COLLAPSE.

Figure 3-25

Surface Nodes

Radiosity Node

Surface

Radiation Resistor

Effect of COLLAPSE on Radiation Resistor Network

The COLLAPSE is applied on a template by template basis and is applied separately to each individual waveband in the template. This versatility gives the user full control over which surface will have their corresponding radiosity nodes collapsed. In order to collapse the radiosity nodes on one surface, but not on another surface of the same material, the user will assign two different template IDs (TIDs), to the two surfaces. Then in the VFAC templates, specifying the COLLAPSE_ID for each surface will determine whether the radiosity nodes made for that surface are collapsed.

Examples of TEMPLATEDAT Files for Thermal Radiation The following examples of VFAC templates range from simple to complex. Each example is described briefly.

Main Index

62

Patran Viewfactor Analysis Patran Thermal TEMPLATEDAT Files for Surface Property Description

This VFAC template will be referenced whenever a surface which was VFAC LBCed with UID 99 is referenced. This template defines a gray surface (nbands defaults to 0) and a constant emissivity of 0.89 and a transparent media (default value of 1.0 for TAU).

************************************************************************ *Example VFAC Template 1 * *KeywordTIDnbands * VFAC99 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE * 0.89 * ************************************************************************

The following template also describes a gray surface, but this time there is a participating media also with a constant transmissivity of 0.95.

************************************************************************ *Example VFAC Template 2 * *KeywordTIDnbands * VFAC10010 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE * 0.890.95 * ************************************************************************

This is a slightly more complicated template. It is still for a gray surface, but now the emissivity is defined by the material property with MPID 4000.

************************************************************************ *Example VFAC Template 3 * *KeywordTIDnbands * VFAC117 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE * 0.00.954000 * ************************************************************************

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 63 Patran Thermal TEMPLATEDAT Files for Surface Property Description

The following template defines both the emissivity and transmissivity in terms of material property data records identified by MPIDs (EMPID and TMPID). Since the TMPID is negative, it will be evaluated as a function of time.

************************************************************************ *Example VFAC Template 4 * *KeywordTIDnbands * VFAC1098 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE * 0.00.04000-3000 * ************************************************************************

The next template has the KFLAG set to one, and thus needs all of the preceding data on the line defined (recall that embedded blank fields are not allowed). This template is similar to the previous one, except that the transmissivity will be evaluated by Beer’s Law and the extinction coefficient will be determined from the time dependent material property definition in MPID 3000.

************************************************************************ *Example VFAC Template 5 * *KeywordTIDnbands * VFAC897 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE * 0.00.04000-30000.00.01 * ************************************************************************

The next example shows the previous example without the supporting comments and spaces. Clearly the previous example is easier to read and we recommend that some standard and clear format be followed by the user.

************************************************************************ *Example VFAC Template 6 * VFAC897 0.0,0.0,4000,-3000,0.0,0.0,1 * ************************************************************************

Main Index

64

Patran Viewfactor Analysis Patran Thermal TEMPLATEDAT Files for Surface Property Description

The following template is for a surface with three wavebands, the first from 0.0 to 2.0 microns, the second from 2.0 to 5.0 microns, and the third from 5.0 to 1.0E6 microns (approximately infinity for wavelengths). Each waveband defines the properties in a different way and serves to exemplify the versatility of the wavebands and VFAC templates.

************************************************************************ *Example VFAC Template 7 * *KeywordTIDnbands VFAC8973 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE 0.890.00-30000.02.0 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE 0.0.0.04000-30002.05.01 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE 0.981.0005.01.0E6 * ************************************************************************

The following templates show the use of the COLLAPSE_ID. Note that when a COLLAPSE_ID is given, the other fields of the template must be filled in with appropriate or default values as placeholders.

************************************************************************ *Example VFAC Template 8 * *KeywordTIDnbands VFAC392 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE 0.000.0000.03.000 * *EPSILONTAUEMPIDTMPIDLAMBDA1LAMBDA2KFLAGCOLLAPSE 0.000.0003.01.0E602 * ************************************************************************

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 65 Compatibility Requirements for Model and VFAC Templates

3.8

Compatibility Requirements for Model and VFAC Templates The model of the problem and its VFAC boundary condition data must satisfy certain compatibility requirements or else an error will occur in Viewfactor while attempting to combine the viewfactor data and the VFAC template data to make resistors. Should this occur, you may need to change the VFAC boundary condition and rerun the viewfactor analysis. The best way to avoid this is careful planning.

Origin of the Problem The compatibility requirements primarily originate from the fact that thermal radiation interchange is between pairs of surfaces and the requirement in Patran Thermal that the radiation resistors be symmetric. Thus, the resistor that would be made from surface one to surface two must be the same as would be made from surface two to surface one. Which resistors are made is determined by the VFAC template data and VFAC boundary condition data for each surface. Incompatibility results, for example, when one surface is gray (nbands = 0) and the other surface has spectral wavebands (nbands > 0). Compatibility problems can also arise just from a single surface and its associated VFAC data. An example of this is a VFAC boundary condition with an AMBNOD specified and a VFAC template with KFLAG set to one. The problem here is that the KFLAG setting will require a distance for calculating the transmissivity from Beer’s Law and the distance from the surface to the ambient environment (AMBNOD) is not well defined or known. The specific compatibility requirements are: • All UIDs referenced in the VFAC boundary condition, input data form must be available TIDs in the VFAC

templates. • All VFAC templates referenced by the VFAC boundary condition UIDs in an enclosure must have the same

number of wavebands. This means that all surfaces in an enclosure must either be gray (nbands = 0) or have the same spectral wavebands (nbands > 0). • Surfaces with an AMBNOD in their VFAC LBC are incompatible with VFAC templates that have their KFLAG

set to one. If the user wishes to model the problem with KFLAG set to one, then the ambient environment must be modeled as surfaces at the ambient temperature. • If the surface has not been given a participating media node, MEDNOD, in its VFAC boundary condition and

the constant transmissivity, CONSTANT_TAU, is not set to 1.0 in the VFAC template identified by the UID for that surface, then an error will occur. This is because no participating media has been defined, yet the VFAC template says that some of the radiant energy from the surface is absorbed by the medium (since the transmissivity is not 1.0). • For every pair of surfaces in an enclosure that can see each other, the following quantities for each must be

equal: MEDNOD, CONSTANT_TAU, TMPID, LAMBDA1, LAMBDA2, and KFLAG. • If there are multiple wavebands, then this condition must hold for each waveband. Note also that by the previous

requirement on wavebands in the enclosure that the number of wavebands for every pair of surfaces which can see each other is equal.

Suggested Procedures to Avoid Compatibility Problems Careful planning is the key to avoiding compatibility problems between the model’s VFAC boundary condition data and the VFAC templates in the TEMPLATEDAT file. Plan out the modeling strategy, paying particular attention to the most complex model of the problem. If for example, a simple model does not include a participating media node even though it is not used, the viewfactor calculations will have to be redone if in the future a participating media needs to be modeled. If, on the other hand, the participating media node had been included in the model, even though it will not be used in the simple model when it is needed in the more complex model with the participating media, it will be there and the viewfactors will not have to be recalculated. Only the VFAC templates will need to be changed and new Patran Thermal radiation resistors made.

Main Index

66

Patran Viewfactor Analysis Compatibility Requirements for Model and VFAC Templates

If you plan or want to be able to use the KFLAG set equal to one option, then do not use ambient environment nodes, AMBNOD, in the model. Instead, model the ambient environment as surfaces at the ambient temperature. If you plan or want to be able to model participating media, include the participating media node, MEDNOD, in the model. There is no harm if it is not used, but it will be there when it is needed. Take care when planning the enclosures and participating media nodes, MEDNOD, to ensure that all surface pairs in an enclosure which can see each other have the same MEDNOD. The above suggestions are the most important, as failure to follow them may require that the viewfactors be recalculated and this is the most computationally expensive part of the Viewfactor analysis. The requirements for the equality of the CONSTANT_TAU, TMPID, LAMBDA1, LAMBDA2, and KFLAG between surface pairs which can see each other in an enclosure is not so troublesome if violated. It can be corrected by changing the VFAC templates but does not in general require recalculating the viewfactors. Still, careful planning is the best prevention. If there are two or more enclosures, the user may wish to use different VFAC templates for the same material in the different enclosures. This will give the user greater latitude in modeling different, but simultaneous, phenomena in the different enclosures. If a constant value for transmissivity, CONSTANT_TAU, is used, it must be the same for all surfaces in an enclosure which can see each other. The TMPID must be the same for all surfaces in an enclosure which can see each other. The KFLAG must be the same for all surfaces in an enclosure which can see each other. If the KFLAG is zero, then the CONSTANT_TAU or TMPID data must be for transmissivity data directly. If the KFLAG is one, then the CONSTANT_TAU or TMPID data must be for extinction coefficient data to be used in Beer’s Law. The beginning and ending wavelengths for each waveband in the enclosure must be in the same order and must be equal for all surfaces in an enclosure which can see each other. Given these restrictions on the data in an enclosure, the user may wish to model the same material in each different enclosure with different UIDs so that for example it may be represented with different VFAC templates in each different enclosure. This involves some additional work, but greatly extends the flexibility and capabilities of the model.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 67 Symmetry as Applied to the Model and Viewfactor Radiation Exchange

3.9

Symmetry as Applied to the Model and Viewfactor Radiation Exchange Symmetry as it relates to viewfactors and thermal radiation analysis is not a very easy topic to understand. If the user does not plan to make use of symmetry in modeling the problem, then this section may be omitted from study. Also, the user may wish to come back to this section at a later time, after mastering other aspects of viewfactor analysis, since this material is not in the mainstream of the rest of this document. The symmetry which we refer to here is not the symmetry of axisymmetry or the symmetry which allows us to reduce a three-dimensional model to a two-dimensional model. Rather it is the symmetry which pertains to reflections about lines or planes and discrete rotations about an axis.

The Purpose of Symmetry in Viewfactor When modeling thermal problems the analyst may look for and make use of symmetry in the model geometry, materials, and boundary conditions. This is typically done to reduce the overall size of the computer descriptions of the problem and reduce the time required to analyze it. There are many existing models for particular thermal problems and many of them have made use of symmetry in describing the problem. It is advantageous, to the greatest extent reasonably possible, to be able to use these existing models for thermal radiation analysis also. Unfortunately, this is frequently not possible because the radiative boundary condition imposes a much stricter symmetry requirement on the model than do the other boundary conditions, material properties or geometry. The other boundary conditions typically involve only one surface area at a time, whereas the radiative energy interchange at the boundary involves pairs of surface areas and complicated relationships between the surface’s normals, angles between these normals and the intersurface ray, and the distance between the surfaces. In order for the model to possess true symmetry, all of these attributes must exhibit the same symmetry and this is usually not the case. It is desirable not to make a full model of an otherwise symmetric model just to deal with the nonsymmetry imposed by the radiative boundary conditions. To this end, Viewfactor provides some capabilities for specifying symmetry for the viewfactor analysis that is only needed for the Viewfactor analysis and not needed for the remainder of the thermal analysis. This will allow the use of existing models, in some cases, with only minor modification (addition of the symmetry information), and allows the thermal network analysis to be performed on the symmetric model while the viewfactor calculations are performed on the complete model.

Caveats Concerning the Use of Symmetry in Thermal Radiation Modeling Patran and Viewfactor cannot check the validity of the user’s symmetry description of the model. When ascertaining the nature of the symmetry present (or not present as the case may be) for thermal radiation interchange and viewfactors, there is ample opportunity for error. These errors are often difficult to detect either by inspection of the model or from detecting erroneous analysis results. We are generally not used to thinking in terms of the symmetry requirements for viewfactors between pairs of surfaces and so this symmetry, or lack thereof, is not intuitively obvious to us as is say geometric symmetry. Frequently we work with problems so complex that we are unable to known whether the analysis is correct or not and thus cannot rely on the detection of erroneous results to detect symmetry errors. Some examples are given later in this section to illustrate the difficulty of correctly modeling symmetry for radiative boundary conditions. The use of symmetry may also in some cases result in computer numerical errors accumulating in such a way that they do not cancel each other out. Thus it is possible to have significantly larger errors in the viewfactors for a model making use of symmetry than in one which does not use symmetry.

Symmetry Operations Supported in Viewfactor The symmetry supported in Viewfactor will be referred to as symmetry operations, for example discrete rotation and reflection operations. These operations are different depending on the coordinate system in which the model is described. In Viewfactor and Patran Thermal, the user may use any one of three coordinate systems, these being two-dimensional Cartesian coordinates, three-dimensional Cartesian coordinates, and two-dimensional axisymmetric coordinates. The symmetry operations supported by Viewfactor and Patran Thermal for Viewfactor analysis will now be described for each of these coordinate systems.

Main Index

68

Patran Viewfactor Analysis Symmetry as Applied to the Model and Viewfactor Radiation Exchange

Symmetry in 2-D XY Space There are two symmetry operations supported in 2-D XY space (two-dimensional Cartesian coordinate system). These are reflection about a straight line in the coordinate axes plane and some number of rotations by some number of degrees about an axis perpendicular to the coordinate plane. Symmetric Reflection About a Line This is analogous to reflections in a mirror. The line must be straight and in the coordinate system plane. An example is shown in Figure 3-26. Symmetric Rotation About an Axis This symmetry operation causes the entire model to be replicated by rotating it as a unit about an axis. The axis must be perpendicular to the coordinate plane. The number of degrees of rotation must be specified as well as the number of times the rotation is to be repeated. For repeated rotation, the object will be recreated at each step in the series of rotations. An example is shown in Figure 3-26.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 69 Symmetry as Applied to the Model and Viewfactor Radiation Exchange

Symmetry Line

Y

X

Y

Axis

X

Rotation of 72

Figure 3-26

Main Index

repeated 4 times about an axis

Symmetry Operations in 2-D XY Space

70

Patran Viewfactor Analysis Symmetry as Applied to the Model and Viewfactor Radiation Exchange

Combining Symmetry Operations Up to two reflections about lines and one rotation about an axis may be combined in 2-D XY models. Successive symmetry operations operate on the model and its symmetric images present at the beginning of the symmetry operation. An example showing combined symmetry operations is illustrated in Figure 3-27.

Symmetric Images After 2 Rotations of 120

Symmetric Image After First Reflection Original Model

Rotation Axis

Y Symmetric Image After Second Reflection

Second Symmetry Line

First Symmetry Line X Figure 3-27

Combined Symmetry Operations in 2-D XY Space

Symmetry in 3-D XYZ Space There are two symmetry operations supported in 3-D XYZ space (three-dimensional Cartesian coordinate system). These are reflection about a plane and some number of rotations by some number of degrees about an axis of rotation. Symmetric Reflection About a Plane This is analogous to reflections in a mirror. The mirror and the plane representing it must be planar. Its orientation in space does not matter. An example is shown in Figure 3-28.

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 71 Symmetry as Applied to the Model and Viewfactor Radiation Exchange

Symmetric Rotation About an Axis This symmetry operation causes the entire model to be replicated by rotating it as a unit about an axis. The number of degrees of rotation must be specified as well as the number of times the rotation is to be repeated. For repeated rotation, the object will be recreated at each step in the series of rotations. An example is shown in Figure 3-28.

Symmetric Image

2

6

Reflection Plane

4

6 2

Object

Rotation Axis

Original Object

Symmetric Images 2 Rotations of 120

Figure 3-28

Main Index

Symmetry Operations in 3-D XYZ Space

About the Rotation

72

Patran Viewfactor Analysis Symmetry as Applied to the Model and Viewfactor Radiation Exchange

Combining Symmetry Operations Up to three reflections about planes and one rotation about an axis may be combined in 3-D XYZ models. Successive symmetry operations operate on the model and its symmetric images present at the beginning of the symmetry operation. An example showing combined symmetry operations is illustrated in Figure 3-29.

Symmetric Image After First Reflection

Second Reflection Plane First Reflection Plane Symmetric Images After Second Reflection

Original Obje

Figure 3-29

Main Index

Combined Symmetry Operations in 3-D XYZ Space

Chapter 3: Model Creation for a Thermal Radiation Problem 73 Symmetry as Applied to the Model and Viewfactor Radiation Exchange

RZ (Axisymmetric) Space Symmetric Reflection About a Line There is only one symmetry operation supported in 2-D axisymmetric RZ space. It is reflection across a straight line perpendicular to the Z axis. The straight line must be defined in the RZ plane. No combination of symmetry operations is allowed in this coordinate system. An example is shown in Figure 3-30.

Center Line

Object

Z Symmetry Line R

Symmetric Image

Figure 3-30

Main Index

Symmetry Operations in Axisymmetric RZ Space

74

Patran Viewfactor Analysis Symmetry as Applied to the Model and Viewfactor Radiation Exchange

list2+ of the Use of Symmetry in Thermal Radiation Modeling Figure 3-31 shows a two-dimensional model of a long solid cylinder surrounded by an annulus and a concentric hollow cylinder. The cylinders are assumed to be long enough that end effects can be neglected and are modeled in two dimensions.

Symmetry Line

3'

3

2'

1'

2

1'

Nonuniform Heat Flux Figure 3-31

Example of the Use of Symmetry

The lower semicircular outer boundary of the outer hollow cylinder receives a nonuniform heat flux as shown by the length of the arrows. Energy is transferred from the inner surface of the hollow cylinder to the solid cylinder by thermal radiation. Although the geometry is axisymmetric, the heat flux boundary condition is not. The heat flux boundary condition and the geometry together are symmetric about a vertical line through the center of the model. Unfortunately, the radiation boundary conditions are not symmetric about this line as can be seen by examining the view from point 1 to points 2 and 2' (or many other pairs of points). Normally, this lack of symmetry either would not be noticed by the analyst and an incorrect model created or the entire object in 2-D would be modeled. Viewfactor, through its symmetry operator, will allow the user to model this object as its symmetric right or left half, thus reducing the size of the Patran Thermal analysis by half. To handle the nonsymmetry of the radiative boundary conditions, the user must tell Viewfactor, through use of the symmetry operators, to take into account the symmetric image of the model when calculating viewfactors and making Patran Thermal radiation resistors. In this case, this would be done by specifying a symmetric reflection about a line coincident with the vertical symmetry line. Instructions on how to enter the symmetry operators into the Patran model will be given at the end of this section. See Entering Viewfactor Symmetry Operations in the Patran Model, 75.

list2+ Which Appears Symmetric, But in Fact Is Not Symmetric This example is similar to the previous example except the heat flux boundary condition has been changed to be symmetric on a quarter section of the model. One way to model this problem appears to be to take the upper right quarter section and

Main Index

Chapter 3: Model Creation for a Thermal Radiation Problem 75 Symmetry as Applied to the Model and Viewfactor Radiation Exchange

replicate it by rotating about the center point by 90 degrees and then reflecting the resulting model and image about a horizontal line through the cylinder’s center as shown in Figure 3-32.

1'

2 3'

2'

1 3 3'' 1'' 2'''

3''' 2''' 1'''

Figure 3-32

Incorrect Use of Symmetry Operations in Viewfactor

The geometry and boundary conditions seemingly appear as we think they should. You might be tempted to define a Viewfactor rotational symmetry operator and reflection symmetry operator for this problem and proceed with the thermal analysis using the quarter section model. This, however, would be erroneous. Upon careful examination of the model the reader can see that, for example, the view from point 1 is not the same as the view from point 1''. Thus this is not a correct model of the radiative interchange in this model. Such errors are subtle and difficult to detect. In general, we recommend that symmetry not be used in thermal radiation models in order to preclude the possibility of such problems.

Entering Viewfactor Symmetry Operations in the Patran Model The symmetry operators for Viewfactor are entered into the model through Patran as special elements. Patran Thermal’s PATQ then translates these special elements into Viewfactor symmetry operators. The symmetry operators, their relationship to the thermal model, and the images of the model they cause to be created in the Viewfactor program are described in The Purpose of Symmetry in Viewfactor, 67 through Symmetry Operations Supported in Viewfactor, 67. This section describes the mechanics of entering the special elements used to describe the symmetry operators in Patran. The meaning and proper use of the symmetry operators is explained in the previous sections. The symmetry operators and their corresponding special elements are: Reflection about a Plane

Radiation Symmetry Triangle Element.

Reflection about a Line

Radiation Symmetry Bar Element--Reflection.

Rotation about an Axis

Radiation Symmetry Bar Element Rotation with element property data.

Main Index

76

Patran Viewfactor Analysis Symmetry as Applied to the Model and Viewfactor Radiation Exchange

The element property data for rotation about an axis contains the number of times the rotation is to be repeated and the angle of rotation in degrees. The element input property form is found under Element Properties looks like Get the Input Properties form under Element Properties.

Finite Element Action:

Create

Object:

ID

Type:

Rad Sym Bar Rotation

The nodes belonging to the radiation symmetry elements have to be declared as type "I" nodes. To do so one must first create OD elements at the location of the nodes belonging to the radiation symmetry elements. When the OD elements are created go to the element properties menu and select dimension: OD, type: Node type and click on the ‘input data’ button and select ‘information node’ in the ‘node type’ data box. Then select the OD elements created as the application region and click on ‘apply’. Nodes of type I (ignore) will not be translated by PATQ into the Patran Thermal nodes, but will only be used by PATQ to generate the Viewfactor symmetry operators.

Main Index

Chapter 4: Preparation for Analysis Patran Viewfactor Analysis

4

Main Index

Preparation for Analysis



Introduction



Viewfactor Execution From Patran Thermal



PATQ Translation from the Patran Neutral File to the VFINDAT File



VFCTL, the Viewfactor Program Execution Control File

78 79

85

80

78

Patran Viewfactor Analysis Introduction

4.1

Introduction After the model has been created in Patran and output to a neutral file, you can proceed with the next step in the analysis cycle. This chapter addresses two topics: 1. Translating the Patran neutral file into a file, VFINDAT, which can be read by Viewfactor for analysis, and 2. Creating the Viewfactor program execution control file, VFCTL.

Main Index

Chapter 4: Preparation for Analysis 79 Viewfactor Execution From Patran Thermal

4.2

Viewfactor Execution From Patran Thermal The Viewfactor run is submitted from the Patran Analysis form. The analysis preference must be set to Patran Thermal. To submit the run: 1. Make sure action is set to analyze in the Analysis form. 2. Under solution type, click Perform Viewfactor Analysis, OK to enter. 3. Under Solution Parameters, the Run Control Parameters and Viewfactor File Names may be selected. Default values have been assigned which should work for the majority of cases. (See Keywords in the VFCTL File, 88 for the parameter definition.) 4. Under Submit Options, make sure the Execute Viewfactor and Spawn Batch Job toggle are set (defaults). After clicking, Apply on the Analysis form, the appropriate files will be created and a background job will be submitted. Note:

Main Index

If both Execute Viewfactor Analysis and Execute Thermal Analysis toggles are set, Patran Thermal will wait until the Viewfactor solution is done before proceeding with the thermal analysis.

80

Patran Viewfactor Analysis PATQ Translation from the Patran Neutral File to the VFINDAT File

4.3

PATQ Translation from the Patran Neutral File to the VFINDAT File The MSC Patran Thermal User’s Guide provides more complete information on the Patran Thermal program PATQ. This Guide is only concerned with the aspects of PATQ which relate to translating a neutral file to a VFINDAT file.

Spawning From Patran vs. Stand-Alone Execution Typically, Patran Thermal jobs will be spawned directly from Patran, in which case the job control will have been defined based on the solution type and submit options selected in the Analysis form. There are occasions (for example, if the thermal solver resides on a different workstation or on a mainframe) where a stand-alone run is required. The following provides the steps in the translation program (PATQ) for a stand-alone run. The translation process is fairly simple and PATQ is an interactive menu-driven program with prompts for user input when required. Many of the prompts have default values that will be used if just a carriage return is entered. PATQ also generates diagnostic error messages if it detects any problems with the neutral file data.

Step-by-Step Procedure (Stand-Alone Execution) Translating a neutral file using PATQ is a fairly simple procedure. Bold text represents user input. Plain capitalized text is used to show computer response. Commands are entered by pressing the RETURN key. The following example details the steps needed to do the translation. These sessions were run on a VAX/VMS installation. If you have a different system, or have altered the standard installation, your sessions may be slightly different. Refer to the Operations Guide for your computer for the correct commands. Note:

Main Index

The steps below are automatically performed as a background job when Patran Thermal execution is spawned from Patran.

Chapter 4: Preparation for Analysis 81 PATQ Translation from the Patran Neutral File to the VFINDAT File

Main Index

82

Patran Viewfactor Analysis PATQ Translation from the Patran Neutral File to the VFINDAT File

$PATQ

At the system level, run PATQ.

PATRAN <--> Q/TRAN <--> Viewfactor Preference Module Version 2.3 Release Date: JULY 1, 1988 Please 1 --> 2 --> 3 -->

Enter the Desired Option: Quit Access the Material Property Data Base Utilities Read a PATRAN Neutral File and Generate Q/TRAN and Viewfactor Input Data File Segments 4 --> Generate a new QTRAN.FOR 5 --> Convert a Q/TRAN Output File to PATRAN Nodal Results File(s) 6 --> Convert a Q/TRAN Output File to PATRAN Neutral Files 7 --> Generate Temperature vs. Time Plot Files 8 --> Convert CONDUC.DAT, VFRES.DAT, or CAP.DAT from binary to text 9 --> Map Temperatures from one Neutral File to Another 10 --> Convert a Nodal Results File to a Neutral File

>3

Choose menu pick number 3 to translate a neutral file to QTRAN and Viewfactor input data files.

Please enter the number corresponding to the dimensionality of the problem, where: 2 --> X-Y -2 --> R-Z 3 --> X-Y-Z (Default)

>2 Neutral File Name?

>

Enter the appropriate dimensionality code for your problem. (Default = patran.out) (or type QUIT to Quit)

Press RETURN (or type the name of the desired file).

Do you wish to convert the nodal coordinates to another system of units? (Y/N, Default = N)

>

Press RETURN.

Virtual me1ppropriate file (Y or N, Default = Y)

>

Main Index

Press RETURN.

Chapter 4: Preparation for Analysis 83 PATQ Translation from the Patran Neutral File to the VFINDAT File

Neutral File Input Complete. Reading the TEMPLATE file. Searching THERMAL$LIB for a TEMPLATE.BIN file. TEMPLATE.BIN file in THERMAL$LIB successfully opened. Loading of default MID templates proceeding. Default MID Template Files Loaded. TEMPLATE.DAT Input Complete. Generating VFIN.DAT. Do you wish to generate a VF input data file (VFIN.DAT)? (Y or N, Default = Y)

>

Press RETURN to generate a VFINDAT file.

Total number of Radiating Surfaces = 18 Total number of Enclosures = 1. VF Input File Successfully Completed. Generating CONVEC.DAT. CONVEC.DAT Complete. Generating QMACRO.DAT and QBASE.DAT. QMACRO.DAT and QBASE.DAT Complete. Generating TMACRO.DAT, TEMP.DAT, and TFIX.DAT. TMACRO.DAT, TEMP.DAT, and TFIX.DAT Complete. Proceeding With NODE.DAT. NODE.DAT Complete. Proceeding with CONDUC.DAT and CAP.DAT. ***>>> Allocating memory for conductive resistor sort and merge operation<<<*** Releasing 77262 words of virtual memory. Releasing 12000 words of virtual memory. Releasing 12000 words of virtual memory. >>> Beginning Sort & Merge Operation on 108 Conductive Resistors <<< Partitioning of resistor data complete. Beginning final sort operation. Conductive Resistor Sort Completed. Beginning Conductive Resistor Merge Operation. Conductive Resistor Merge Successfully Completed. There is now a total of 93 resistors >>> Beginning Sort & Merge Operation on 72 Capacitors <<< Partitioning of capacitor data is complete. Beginning final sorting operation. Capacitor Data Sort Completed. Beginning Capacitor Merge Operation. >>> Capacitor Merge Successfully Completed. There is now a total of 42 capacitors. Element Data Translation Completed Successfully. Enter to continue.

>

Main Index

Press RETURN

84

Patran Viewfactor Analysis PATQ Translation from the Patran Neutral File to the VFINDAT File

PATRAN <--> Q/TRAN <--> Viewfactor Preference Module Version 2.3 Release Date: JULY 1, 1988 Please 1 --> 2 --> 3 --> 4 5 6 7 8

--> --> --> --> -->

9 --> 10 -->

>1

Enter the Desired Option: Quit Access the Material Property Data Base Utilities Read a PATRAN Neutral File and Generate Q/TRAN and Viewfactor Input Data File Segments Generate a new QTRAN.FOR Convert a Q/TRAN Output File to PATRAN Nodal Results File(s) Convert a Q/TRAN Output File to PATRAN Neutral Files Generate Temperature vs. Time Plot Files Convert CONDUC.DAT, VFRES.DAT, or CAP.DAT from binary to text Map Temperatures from one Neutral File to Another Convert a Nodal Results File to a Neutral File

The translation is now done. Enter 1, menu selection Quit, to exit PATQ.

There are many different options in the translation process and they cannot all be presented here. The above dialogue is typical of a PATQ translation. If errors are detected by PATQ, error messages will be written to the terminal screen. Patran Thermal’s PATQ uses the default file name VFINDAT for the Viewfactor input data file. Viewfactor allows you to specify all relevant file names and so you may, at your discretion, change the name of the VFINDAT file created by PATQ. The new name must also be given in the VFCTL file (see VFCTL, the Viewfactor Program Execution Control File, 85.

Main Index

Chapter 4: Preparation for Analysis 85 VFCTL, the Viewfactor Program Execution Control File

4.4

VFCTL, the Viewfactor Program Execution Control File The Viewfactor program execution control file, VFCTL, contains the information needed by Viewfactor to know which parts of the Viewfactor program should be executed, the names of various files needed by Viewfactor and some data or parameters which will influence the viewfactor calculations. Note:

For execution spawned from Patran Thermal, this file is automatically created based on input data in the Patran Thermal Analysis Forms.

Philosophy and Structure of the VFCTL File The VFCTL file is analogous to a session or command file in that it can be thought of as containing commands which set parameters and filenames in the Viewfactor program. This was done because the Viewfactor program will typically be submitted as a batch job or run as a subroutine call. In both cases, the user is not interacting with the program and the run control commands need to be already stored in the VFCTL file. You may use any nonconflicting legal file name for the VFCTL file. The default name is in the default directory and named VFCTL. A nondefault name must be entered as a parameter on the command line submitting a Viewfactor job (see Analysis (Ch. 5). The file may contain comment lines and data lines. Comment lines begin with an asterisk. Data lines begin with a keyword, followed by some data (sometimes optional data). The valid keywords are defined in Keywords in the VFCTL File, 88. Lines may contain leading blanks. Default values are provided automatically for all run control data. You do not need to enter data lines if the default is the desired value. The default values are given in Keywords in the VFCTL File, 88. The data lines may be placed in any order and only one occurrence of each keyword is allowed. A complete VFCTL file from a VAX/VMS platform is shown below. The filenames have been made generic, but the $PATH is not generic. Invalid keywords cause a fatal error. This prevents a costly Viewfactor execution with invalid data.

Main Index

86

Patran Viewfactor Analysis VFCTL, the Viewfactor Program Execution Control File

Example VFCTL File

************************************************************************** * *BEGINNING OF EXAMPLE VFCTL FILE * *The path will be prefixed onto every other file name given below. * $PATH:[PHILLIPS.VF.FILES.CYLINDER.CASE1] * *The message file will contain text messages from Viewfactor. * $MESSAGE_FILE:VFMSG * *The status file was designed for restarting the Viewfactor program. *This capability is not presently supported. * $STATUS_FILE:VFRESTARTSTAT * *The diagnostic file will contain numerical diagnostic data. * $DIAGNOSTIC_FILE:VFDIAG * *The title will be placed near the top of various output files to aid in later identification. * $TITLE:THIS IS A TEST * *The restart file was designed for restarting the Viewfactor program. *This capability is not presently supported. * $RESTART_FILE:VFRESTARTDAT * *This is the input data file which typically was created by the Patran Thermal PATQ translator. * $IN_DATA:VFINDAT * *This is the file containing the VFAC template data. $TEMPLATE_FILE:TEMPLATEDAT * *This is the file containing the raw viewfactor data.SSSS * $RAW_DATA:VFRAWDAT * *This file contains the Patran Thermal radiation resistors made by Viewfactor. * $OUT_DATA:VFRESDAT * *This file contains the node definition data for Patran Thermal for the radiosity nodes *created by Viewfactor. * $RAD_NODE_FILE:VFNODEDAT * *The value of the run control parameter controls which parts of the Viewfactor program *will be executed. * $RUN_CONTROL:0

Main Index

Chapter 4: Preparation for Analysis 87 VFCTL, the Viewfactor Program Execution Control File

* *This parameter controls the restart capabilities of Viewfactor. *Presently only the value 0 is supported (this is not a restart). * $RESTART_FLAG:0 * *This parameter controls the convergence checking in Viewfactor. The *value of -1.0 causes a default value based on the number of surfaces *in each enclosure to be calculated. * $CONVERGE-1.0 * *Viewfactors below the zero cutoff value will be set to 0.0 * $ZERO:0.0 * *This parameter controls the accuracy of curved surface approximation by *linear surfaces for obstruction checking. * $APPROX_CURVE:0.1 * *These values determine the maximum order of numerical integration *quadrature permitted for integration of various quantities. * $GAUSS_ORDER:888 * *These values determine the minimum and maximum number of subdivisions for *the three-dimensional representation of an axisymmetric surface for the *purpose of checking for obstructions and calculating viewfactors. * $AXISYM_SURFACE:313 * *This keyword signals the end of the VFCTL file. * $EOF: * *THIS IS THE END OF THE VFCTL FILE EXAMPLE. * **********************************************************************

Main Index

88

Patran Viewfactor Analysis VFCTL, the Viewfactor Program Execution Control File

Keywords in the VFCTL File Presently, there are 19 valid keywords available for use in the VFCTL file. All keywords here begin with "$" and end with ":". The valid keywords are: $PATH: $MESSAGE_FILE: $STATUS_FILE: $DIAGNOSTIC_FILE: $TITLE: $RESTART_FILE: $IN_DATA: $TEMPLATE_FILE: $RAW_DATA: $OUT_DATA: $RAD_NODE_FILE: $RUN_CONTROL: $RESTART_FLAG: $CONVERGE: $ZERO: $APPROX_CURVE: $GAUSS_ORDER: $AXISYM_SURFACE: $EOF: Each keyword will be described in detail with the following information given: 1. purpose, 2. required or optional, 3. default value, 4. valid range, and 5. suggested values. The order of the keywords in the VFCTL file does not matter, with the exception that the $EOF: keyword should be last. Data will not be read after an $EOF: keyword is encountered. If a keyword is not present, its data will assume the default values. $PATH: pathname Purpose of the Keyword

"$PATH:" is to name a path for Viewfactor files. The pathname will be prefixed to all other filenames used in this particular execution of Viewfactor. This feature is provided so that long pathnames do not need to be entered with every filename, provided they are common and the same as that given in pathname. If you adopt a standard convention for filenames and group different models and cases into directories or paths, then these may be accessed by Viewfactor merely by changing the pathname in VFCTL. If the pathname is blank, then nothing will be prefixed to the filenames and your current default directory will be used.

Required or Optional

Optional.

Default Value

Default is a blank string.

Valid Range

Valid pathnames are computer system dependent.

Suggested Values

None.

$MESSAGE_FILE: filename Purpose of the Keyword

"$MESSAGE_FILE:" is to name the file that will contain text messages from Viewfactor. See VFMSG, 101 and VFMSG, the Viewfactor Message File, 103

Required or Optional

Optional.

Default Value

VFMSG.

Main Index

Chapter 4: Preparation for Analysis 89 VFCTL, the Viewfactor Program Execution Control File

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

$STATUS_FILE: filename Purpose of the Keyword

"$STATUS_FILE:" is to name the file to be used for restart status data for Viewfactor. This capability is not presently supported.

Required or Optional

Optional.

Default Value

VFRESTARTSTAT.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

$DIAGNOSTIC_FILE: filename Purpose of the Keyword

"$DIAGNOSTIC_FILE:" is to name the file that will contain numerical diagnostic data from Viewfactor. See VFDIAG, 101 and VFDIAG, the Viewfactor Diagnostic Data File, 106.

Required or Optional

Optional.

Default Value

VFDIAG.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

$TITLE: title Purpose of the Keyword

"$TITLE:" is to identify a character string title that will be output to various files in Viewfactor to aid in identifying the analysis case to which a file belongs.

Required or Optional

Optional.

Default Value

Default is a black string.

Valid Range

The character string should have less than 80 characters.

Suggested Values

None.

$RESTART_FILE: filename Purpose of the Keyword

"$RESTART"_FILE:" is to name the file that will contain data for restarting Viewfactor. This capability is not presently available.

Required or Optional

Optional.

Default Value

VFRESTARTDAT.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

$IN_DATA: filename Purpose of the Keyword

"$IN_DATA:" is to name the file that must contain the input data for Viewfactor.

Required or Optional

Optional.

Default Value

VFINDAT.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

Main Index

90

Patran Viewfactor Analysis VFCTL, the Viewfactor Program Execution Control File

$TEMPLATE_FILE: filename Purpose of the Keyword

"$TEMPLATE_FILE:" is to name the file that must contain the VFAC templates for this model.

Required or Optional

Optional.

Default Value

TEMPLATEDAT.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

$RAW_DATA: filename Purpose of the Keyword

"$RAW_DATA:" is to name the file which will contain the raw viewfactor data. Viewfactor creates and outputs raw viewfactor data as well as reads in the raw viewfactor data when making Patran Thermal radiation resistors. See VFRAWDAT, 102.

Required or Optional

Optional.

Default Value

VFRAWDAT.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

$OUT_DATA: filename Purpose of the Keyword

"$OUT_DATA:" is to name the file which will contain the Patran Thermal radiation resistor data created by Viewfactor. See VFRESDAT, 102.

Required or Optional

Optional.

Default Value

VFRESDAT.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

$RAD_NODE_FILE: filename Purpose of the Keyword

"$RAD_NODE_FILE:" is to name the file which will contain the node definition data for Patran Thermal for radiosity nodes created by Viewfactor. See VFNODEDAT, 102.

Required or Optional

Optional.

Default Value

VFNODEDAT.

Valid Range

Valid filenames are computer system dependent.

Suggested Values

None.

Main Index

Chapter 4: Preparation for Analysis 91 VFCTL, the Viewfactor Program Execution Control File

$RUN_CONTROL: value Purpose of the Keyword

"$RUN_CONTROL:" is to set a run control value. The run control value determines the execution sequence for Viewfactor. It is this parameter that allows the user to calculate viewfactors before the VFAC templates are ready (set value to 1), calculate or recalculate the Patran Thermal radiation resistors from existing raw viewfactor data (set value to 2), or calculate viewfactors and make radiation resistors in the same execution (set value to 0). For additional information, see Viewfactor Data and Program Flow, 16 and Changing VFCTL, 122. The valid values of the $RUN_CONTROL: parameter and their effects are: 0

Check data, and if the data checking status = OK, calculate viewfactors and make radiation resistors.

1

Check data, and if the data checking status = OK, calculate viewfactors only.

2

Calculate radiation resistors only, using existing raw viewfactor data. This process involves extensive data checking as the calculations are being done and since no significant CPU time saving is realized, no data checking is done before the calculations.

100

Check the data for making viewfactors and radiation resistors and then stop. Do not calculate viewfactors or make resistors.

101

Check the data for making viewfactors and then stop. Do not calculate viewfactors.

102

Check that the appropriate files are present for making radiation resistors from raw viewfactor data. Do not make the radiation resistors.

200

Check the data for making viewfactors and radiation resistors. If the data checking status = OK or the data only generated warning messages, proceed with the viewfactor calculations and the creation of the radiation resistors. If the data checking generated an error message, stop.

201

Check the data for making viewfactors only. If the data checking status = OK or the data only generated warning messages, proceed with the viewfactor calculations only. If the data checking generated an error message, stop.

202

This option is functionally the same as 2 above.

1000

Skip data checking and proceed directly to calculating the viewfactors and making the radiation resistors.

1001

Skip the data checking and proceed directly to calculating the viewfactors only.

1002

This option is functionally the same as 2 above.

Required or Optional

Optional.

Default Value

0.

Valid Range

Valid values are the integers listed above.

Suggested Values

None.

Main Index

92

Patran Viewfactor Analysis VFCTL, the Viewfactor Program Execution Control File

$RESTART_FLAG: value Purpose of the Keyword

"$RESTART_FLAG:" is to set the Viewfactor restart flag. The restart flag value controls restarting Viewfactor. Presently this capability is not supported.

Required or Optional

Optional.

Default Value

0.

Valid Range

The only valid value is the integer 0.

Suggested Values

None.

$CONVERGE: value Purpose of the Keyword

"$CONVERGE:" is to set the parameter that controls convergence checking in Viewfactor. If the value entered is less than or equal to zero, a convergence criteria is calculated by Viewfactor based on the number of surfaces in each enclosure. This number will be recalculated as each enclosure is processed. The formula for convergence criteria used by Viewfactor to calculate the default value is: converge_value = 0.01/(number of surfaces in enclosure)0.7.

Required or Optional

Optional.

Default Value

-1.0.

Valid Range

Valid values are real numbers

Suggested Values

The suggested value is the default value, -1.0. If a faster execution time is desired, at the cost of lost accuracy, a value around 0.1 may be appropriate. Until you gain some experience with Viewfactor, it is best to use the default value. The convergence criteria is not an absolute requirement. You don’t know how fast the numerical scheme converges. The fact that it appears to have converged over a few iterations is no guarantee that it has converged. The default values are based on previous user's experience. As you gain experience you may be able to improve on these default values.

≤ 1.0

and representable on the computer.

$ZERO: value Purpose of the Keyword

"$ZERO:" is used to establish a cut off value. Viewfactors below the zero cutoff value will be set to zero. This is a convenient way to eliminate viewfactors and their associated radiation resistors for viewfactors which are close to zero. You must exercise discretion here. For example, in a model with 10,000 surfaces, viewfactors smaller than 0.000001 may be very significant. If the value is less than -0.5, then the zero cutoff value will be set equal to the convergence criteria value. (See $CONVERGE: value, 92. The program gives the user the ability to not make radiation resistors if the viewfactor between nodal subareas is less than a user specified cutoff value. The zero cutoff value is set using the $ZERO: parameter in the VFCTL file.

Required or Optional

Optional.

Default Value

0.0.

Main Index

Chapter 4: Preparation for Analysis 93 VFCTL, the Viewfactor Program Execution Control File

Valid Range

Valid values are real numbers less than or equal to 1.0.

Suggested Values

The default value of 0.0 is highly recommended. The use of numbers greater than the default convergence criteria value is definitely not recommended. In general, the use of nonzero cutoff values is not recommended. If you want to use them anyway, set the zero cutoff value to zero during the calculation of raw viewfactor data and then set it to the desired cutoff value during the creation of resistors from the raw viewfactor data. This will give you the flexibility of changing the zero cutoff value without having to redo the viewfactor calculations. This procedure requires that you submit Viewfactor twice, once with "$RUN_CONTROL:" set to 1 and $ZERO: set to 0.0 and again with "$RUN_CONTROL:" set to 2 and "$ZERO:" set to the desired cutoff value.

$APPROX_CURVE: value Purpose of the Keyword

"$APPROX_CURVE:" is to set the parameter that controls the number of subdivisions a curved surface or line will undergo as it is approximated by plane triangles or straight line segments. These linear subdivisions are used for obstruction checking. The smaller the value, the more subdivisions will be required to approximate a curved surface or line. As more subdivisions are created, the accuracy of the calculations will generally increase and the CPU time will increase dramatically. Note: For straight or planar (flat) surfaces this parameter has little effect since they are already well approximated by linear lines or planes.

Required or Optional

Optional.

Default Value

0.1.

Valid Range

Valid values are real numbers less than or equal to 0.5 and greater than or equal to 0.00001 and representable on the computer.

Suggested Values

The default value of 0.1 is the recommended value.

Main Index

94

Patran Viewfactor Analysis VFCTL, the Viewfactor Program Execution Control File

$GAUSS_ORDER: contours Purpose of the Keyword

double_area

weighting

"$GAUSS_ORDER:" is to set the value that will determine the maximum integration order that Viewfactor will attempt to use while trying to satisfy its convergence criteria. The three different values are applied to three different types of integration. Contours applies to contour integration around the border of a surface. Double_area, applies to integration over the surface areas of surface pairs for which partially obstructed viewfactors are being calculated. Weighting, applies to calculations to weight the surface to surface viewfactors according to the finite element interpolation functions. The values of the parameters do not correspond exactly to the integration order. The correspondence is shown in Table 4-1: Table 4-1

Correspondence between Parameter Values and Integration Order

Parameter ValueIntegration Order 33 44 55 66 77 89 99 1010 1112 1216 1320 1424 1532 1640

In general, the larger the parameter value, the more accurate and more CPU costly the Viewfactor analysis will be. The integration orders jump in larger step sizes as they increase because, for example, an increase from 39 to 40 has little effect, whereas an increase from 3 to 4 may have a significant effect. Required or Optional

Optional.

Default Value

The default values are 8, 8, and 8.

Valid Range

Valid values are integers between and including 3 and 16.

Suggested Values

In general, the suggested values are the default values. The maximum integration order is only used when the numerical integration scheme has not converged. In most cases, this happens before the maximum integration order is reached. Thus you only pay for the higher integration orders when they are needed for accuracy. If you wish to do a test run at less expense than the actual analysis run, the CPU time may be cut down by reducing the GAUSS_ORDER parameters. In general, you will make the weighting parameter less than or equal to the other parameters, although this is not required.

Main Index

Chapter 4: Preparation for Analysis 95 VFCTL, the Viewfactor Program Execution Control File

$AXISYM_SURFACE: minimum maximum Purpose of the Keyword

"$AXISYM_SURFACE:" is to set the two parameters that control the minimum and maximum number of effective subdivisions the three-dimensional image of an axisymmetric surface is subdivided into for obstruction and viewfactor calculations. Here again, Viewfactor does convergence checking and stops iterating when convergence has occurred. The actual number of subdivisions is not the same as the value of the parameter. They correspond in the same way as the integration orders in Table 4-1.

Required or Optional

Optional.

Default Value

3 for the minimum, 13 for the maximum.

Valid Range

Valid values are integers

≤ 16 and ≥ 3

, and maximum greater than or equal to

minimum. Suggested Values

These parameters have the greatest effect of any of the parameters on the execution time and accuracy of an axisymmetric model submitted for viewfactor calculation. The execution time will increase approximately as the square of the number of subdivisions. Unfortunately, the accuracy does not improve in a like manner. In general, the recommended values are the default values. However, the user may find with some experience that values of maximum as low as 8 give satisfactory results.

$EOF: Purpose of the Keyword

"$EOF:" is to signal the end of the VFCTL file.

Required or Optional

Optional.

Default Value, Valid Range, Suggested Values

There is no parameter to receive a default value, be checked for validity, or receive a recommended value.

This concludes the description of the VFCTL keywords.

Main Index

96

Patran Viewfactor Analysis VFCTL, the Viewfactor Program Execution Control File

A sample VFCTL file is delivered with the Viewfactor program. This file is reproduced here. *Sample VFCONTROL file. * *Pathname $PATH *Message file name $MESSAGE_FILE:VFMSG *Diagnostic data file name $DIAGNOSTIC_FILE:VFDIAG *h3 $TITLE:'THIS IS A TEST' *Input data file name $IN_DATA:VFINDAT *Template file name $TEMPLATE_FILE:TEMPLATEDAT *Raw viewfactor data file name $RAW_DATA:VFRAWDAT *Radiation resistor file name $OUT_DATA:VFRESDAT *Radiosity node file name $RAD_NODE_FILE:VFNODEDAT * $STATUS_FILE:VFRESTARTSTAT * $RESTART_FILE:VFRESTARTDAT * 0 = full run,1 = viewfactors only, 2 = resistors only $RUN_CONTROL:0 * $RESTART_FLAG:0 * $CONVERGE:-1.0 $ZERO:0.0 $APPROX_CURVE:0.1 *ContourDouble_areaWeighting $GAUSS_ORDER: 8 8 8 *minimum maximum $AXISYM_SURFACE: 5 13 $EOF:

Note:

Main Index

The above file corresponds to the defaults set when a Viewfactor job is spawned directly from Patran.

Chapter 5: Analysis Patran Viewfactor Analysis

5

Main Index

Analysis



Submitting a Viewfactor Job for Analysis



Output Created by a Viewfactor Execution



Reviewing the Viewfactor Output

103

98 101

98

Patran Viewfactor Analysis Submitting a Viewfactor Job for Analysis

5.1

Submitting a Viewfactor Job for Analysis Submitting a Viewfactor job for execution is very simple. Viewfactor analyses take a lot of computer time and real time. Therefore, it is recommended that the model and the VFCTL file be carefully reviewed to double check that the correct data files are present. This will protect you from submitting an analysis with the wrong data and from abnormal execution due to user error.

Review the Viewfactor Control/Parameters You should review the Viewfactor Solution Parameters in the Analysis form or the VFCTL file being used for this analysis asking the following questions: 1. Is the path name correct? 2. Do all of the input data files exist in the correct directory or will they be automatically created based on the submit options? 3. Are these input data files the correct files for the problem to be analyzed? The following parameters are described in Keywords in the VFCTL File, 88: 1. Does the run control parameter, $RUN_CONTROL:, specify the desired execution sequence? 2. Is the convergence criteria parameter, $CONVERGE:, correct? 3. Is the zero cutoff parameter, $ZERO:, correct? 4. Is the curve approximation parameter, $APPROX_CURVE:, correct? 5. Are the maximum integration order parameters, $GAUSS_ORDER:, correct for contour integration, double area integration, and finite element weighting? 6. Are the axisymmetric surface subdivision parameters, $AXISYM_SURFACE:, correct for the minimum number of subdivisions and maximum number of subdivisions?

Review Directory for Required Files Generally, a Viewfactor analysis will not get very far if the required files are not available. This is a problem if you submit a Viewfactor analysis, go to lunch, and expect to see results later. You may wish to double check that the input data files are actually the correct and desired files for the model you are analyzing. The specific files and filenames required depend on the value of the run control parameter and the filenames given in the VFCTL file. The generic file names for the various values of the run control parameter are given below: $RUN_CONTROL: 0 The following files are required: VFINDAT TEMPLATEDAT The following files are created: VFRESTARTSTAT VFRESTARTDAT VFRAWDAT VFRESDAT VFNODEDAT VFDIAG VFMSG

Main Index

Chapter 5: Analysis 99 Submitting a Viewfactor Job for Analysis

$RUN_CONTROL: 1 The following file is required: VFINDAT The following files are created: VFRESTARTSTAT VFRESTARTDAT VFRAWDAT VFDIAG VFMSG $RUN_CONTROL: 2 The following files are required: VFRAWDAT TEMPLATEDAT The following files are created: VFRESTARTSTAT VFRESTARTDAT VFRESDAT VFNODEDAT VFDIAG VFMSG Note: The above files will be automatically created if execution is spawned from within the Patran analysis menu.

The Viewfactor Command Line This section uses examples from a VAX specific installation. Your system may require different commands. Refer to the Introduction & Overview (Ch. 1) in the Patran Installation and Operations Guide or your system manager. A Viewfactor analysis is submitted by typing at your computer command prompt the command VSUBMIT followed optionally by the VFCTL filename. The filename must be separated from the command by a comma and/or one or more blank spaces. If no VFCTL file name is given, it will default to VFCTL in the default directory. VSUBMIT may be abbreviated as VSU.

Examples:

Let $ be the computer system level prompt. Commands are entered by pressing the RETURN key.

$VSUBMIT

Submits a Viewfactor analysis with the VFCTL file from the default directory.

$VSU VFCTL_TEST_RUN

Submits a Viewfactor analysis with the control file VFCTL_TEST_RUN from the default directory.

$VSU

Submits a Viewfactor analysis with the control file VFCONTROL.CON from the directory [PHILLIPS.VF.FILES] on the default disk.

[SMITH.VF.FILES]VFCONTROL.CON

The actual analysis will be run as a batch or background job and you will receive mail or other notification when the job has terminated. You may also run Viewfactor in interactive mode by using the command VF_RUN instead of VSUBMIT.

Main Index

100

Patran Viewfactor Analysis Submitting a Viewfactor Job for Analysis

Note:

Main Index

A Viewfactor Submit script is automatically created if execution is spawned from within Patran analysis menu.

Chapter 5: Analysis 101 Output Created by a Viewfactor Execution

5.2

Output Created by a Viewfactor Execution The output created by Viewfactor depends on the value of the run control parameter $RUN_CONTROL:. The files VFMSG and VFDIAG are always created (unless a severe error occurs before these files can be opened, which only happens when the file names or directories are invalid). The file VFRAWDAT is created whenever viewfactors are calculated (run control parameter equal 0 or 1). The files VFRESDAT and VFNODEDAT are created whenever Patran Thermal radiation resistors are made (run control parameter equal 0 or 2). The user’s computer system may also create a log file for the batch or background execution of Viewfactor. Viewfactor will also create VFRESTARTSTAT and VFRESTARTDAT files for its own internal use. These two files will automatically be deleted by Viewfactor before execution terminates. However, if execution is terminated abnormally, these files may be left on the disk and you must delete them. Examples are shown in Reviewing the Viewfactor Output, 103.

VFMSG This file contains messages from Viewfactor to the user. It begins with a header, followed by the title data from the VFCTL file. Then the control data that was actually used by Viewfactor is echoed to this file. This data may not look quite the same as the user’s VFCTL file. This is because the format may be different and if you allowed any values to default, the default values will be displayed. The rest of the file contains either informative messages on the progress of the Viewfactor analysis or error messages. Since Viewfactor has been designed to check for many errors and to terminate gracefully on an error condition, the VFMSG file is the only reliable place to find if an error has occurred. For the most part, Viewfactor will not abort at the computer operating system level. When it detects an error, it reports the error to the VFMSG file and terminates execution without triggering any system level errors. Since there is no system level traceback, the traceback information that Viewfactor outputs to the VFMSG file is very helpful if you need to discuss a problem with MSC’s technical support staff. VFMSG, the Viewfactor Message File, 103 contains a sample VFMSG file.

VFDIAG This file contains primarily numeric diagnostic data. The file begins with the title data from the VFCTL file, followed by any additional title data from the input data files. If viewfactors were calculated in this analysis, data is given about the sums of the viewfactors from each surface to every other surface in the enclosure. Similar data is also given for the nodal subareas on each surface. This data is grouped by enclosure. Each group of enclosure data begins with the keyword $ENCL: followed by the enclosure ID number, the number of surfaces in the enclosure, and the number of symmetric images created of this enclosure. Then for each surface in the enclosure there is a line of data containing the surface ID, the sum of the viewfactors from this surface to all other surfaces in the enclosure, 1.0 minus this sum, and the sum of all viewfactors from this surface to other surfaces which were set to zero because they were less than the zero cutoff value ($ZERO:). Following this line will be the sum of the viewfactors for each nodal subarea on the surface. Thus, there will be the same number of data items as there are nodes on the surface. These data will have four values per line until all of the nodes on this surface are used. Currently, none of the element faces supported by Viewfactor have more than four nodes and hence this data fits on one line. This pattern is repeated for each surface in the enclosure. After all of the surfaces in the enclosure have been accounted for in the above manner, some statistical data for the enclosure is given. This data is presented in five columns. The first line of statistical data is for the surfaces as wholes and the following lines are statistical data for the nodal subareas of the surfaces. The statistical data for the nodal subareas do not have a sound theoretical basis and should not be taken too seriously. It is, however, the best data available at this time.

Main Index

102

Patran Viewfactor Analysis Output Created by a Viewfactor Execution

The five columns contain the following data: Column 1:

The maximum deviation from unity for the sums of viewfactors from a surface or nodal subarea surface to all other surfaces or nodal subarea surfaces in the enclosure.

Column 2:

The average deviation from unity for the sums of viewfactors for this enclosure.

Column 3:

The standard deviation for the above average.

Column 4:

The average of the absolute values of the deviations from unity for the sums of viewfactors in this enclosure.

Column 5:

The standard deviation for the above average. The data for each enclosure is terminated with the $ENDENCL: keyword. If there is more than one enclosure, the above pattern is repeated for each enclosure. If Patran Thermal radiation resistors were created by this execution of Viewfactor, the VFMSG file will contain information on the radiosity nodes, emissivity resistors, and radiation resistors created for each enclosure. The file is terminated by the keyword $EOF:. Examples are given in VFDIAG, the Viewfactor Diagnostic Data File, 106.

VFRAWDAT This is an unformatted (binary) file containing data about the surfaces and raw viewfactor data. Since it is in binary form, the user may not readily examine its contents. The binary form was chosen because this data file tends to be very large and the binary form is considerably more compact than the text form.

VFRESDAT This is also a binary file. It contains data describing the Patran Thermal radiation resistors. This file is typically very large, hence its binary form. Patran Thermal can translate the binary form of this file to a text form which the user can read, but it can be a very large file. The procedure for doing this is described in Interface From Viewfactor to Patran Thermal, 111. This file must be referenced by the Patran Thermal QINDAT file in order for it to be included in the QTRAN thermal analysis. See Interface From Viewfactor to Patran Thermal, 111.

VFNODEDAT This file contains the information needed by Patran Thermal to define the additional radiosity nodes created by Viewfactor for the thermal analysis of the thermal radiation interchange. It contains comment lines which begin with a semicolon. If any radiosity nodes were created, the file will contain the line

DEFNOD

beginning_node_number

ending_node_number

1

for the nodes created by Viewfactor. If no radiosity nodes were created by Viewfactor (all surfaces were black) then the file will contain the comment ;NO ADDITIONAL RADIOSITY NODES WERE GENERATED. This file must be referenced by the Patran Thermal QINDAT file in order for it to be included in the QTRAN thermal analysis. See Interface From Viewfactor to Patran Thermal, 111.

Main Index

Chapter 5: Analysis 103 Reviewing the Viewfactor Output

5.3

Reviewing the Viewfactor Output It is strongly recommended that the VFMSG and VFDIAG files be reviewed after every Viewfactor analysis. Since Viewfactor has been designed to handle most of its own errors, there will be no indication at the computer system level that an error has occurred in Viewfactor. Actual errors are reported in the VFMSG file.

VFMSG, the Viewfactor Message File The VFMSG file should be examined to make sure the echoed control data is correct. Then the file should be examined for error messages. If no errors occurred in Viewfactor, the last line of the VFMSG file would be: Successful Execution Completed. The following is an example of a VFMSG file from a successful Viewfactor execution with the run control parameter set to 0. Sample Successful VFMSG File $TITLE: MSC Viewfactor VER. x.x 25-MAY-xx 14:55:10 Viewfactor is a product of MSC.Software Corporation Version: x.x Released: July 1, 20xx. Copyright 20xx, MSC.Software Corporation, Santa Ana, CA, USA. $TITLE: PARALLEL SEMIINFINITE PLATES OPEN TO LEFT AND RIGHT. Beginning of control data echo. $PATH: $MESSAGE_FILE: VF.MSG $STATUS_FILE: vf.stat $DIAGNOSTIC_FILE: VF.DIAG $TITLE: PARALLEL SEMIINFINITE PLATES OPEN TO LEFT AND RIGHT. $RESTART_FILE: vfrestart.dat $IN_DATA: VFIN.DAT $TEMPLATE_FILE: TEMPLATE.DAT $RAW_DATA: VFRAW.DAT $OUT_DATA: VFRES.DAT $RAD_NODE_FILE: VFNODE.DAT $RUN_CONTROL: 0 $RESTART_FLAG: 0 $CONVERGE: -1.000000000 $ZERO: 0.0000000000E+00 $APPROX_CURVE: 0.1000000015 $GAUSS_ORDER: 8 8 8 $AXISYM_SURFACE: 5 16 End of control data echo. VF has successfully completed initialization. Beginning to read/process the INPUT data. $TITLE: PARALLEL PLATES, SIMPLE, FINE, LINEAR HEAT AT NODES. $TITLE: 13-MAY-88 10:11:25 2.3 Beginning to read/process the NODE data. Completed reading/processing the NODE data. Beginning to read/process the ENCLOSURE 1 data. Completed reading/processing the ENCLOSURE 1 data.

Main Index

104

Patran Viewfactor Analysis Reviewing the Viewfactor Output

Beginning the obstruction and viewfactor Calculations completed for surface Calculations completed for surface Calculations completed for surface Calculations completed for surface Calculations completed for surface Calculations completed for surface Calculations completed for surface Calculations completed for surface Calculations completed for surface

calculations for ENCLOSURE 1. 1 of 18 in this enclosure. 3 of 18 in this enclosure. 5 of 18 in this enclosure. 7 of 18 in this enclosure. 9 of 18 in this enclosure. 11 of 18 in this enclosure. 13 of 18 in this enclosure. 15 of 18 in this enclosure. 17 of 18 in this enclosure.

Completed the obstruction and viewfactor calculations for ENCLOSURE 1. The viewfactor calculations are done and the data is in the raw data file. Beginning translation of raw viewfactor data into Patran Thermal resistors. Beginning to read/process the Template data. Completed reading/processing the Template data. Beginning to read/process the VIEWFACTOR data. Beginning to read/process the NODE data. Completed reading/processing the NODE data. Beginning to read/process the ENCLOSURE 1 data. Completed reading/processing the ENCLOSURE 1 data. Beginning to make radiation resistors for ENCLOSURE 1. 12 nodal subareas were black in this enclosure. 24 radiosity nodes were created for this enclosure. 24 resistors, subtype 5, were made from SURFACE to RADIOSITY nodes. 432 resistors, subtype 5, were made from RADIOSITY to RADIOSITY nodes. 36 resistors, subtype 5, were made from RADIOSITY to AMBIENT nodes. Completed making radiation resistors for ENCLOSURE 1. Completed translation of raw viewfactor data into Patran Thermal resistors. Successful Execution Completed.

Main Index

Chapter 5: Analysis 105 Reviewing the Viewfactor Output

The following is an example of a VFMSG file from an unsuccessful Viewfactor execution. The messages contained in the VFMSG file will be extremely useful for debugging purposes. If you should call one of MSC.Software Corporation support centers with Viewfactor, they will almost certainly ask you about the messages in your VFMSG file.

Sample Unsuccessful VFMSG File $TITLE:

MSC Viewfactor VER. x.x 26-MAY-xx 14:35:38 Viewfactor is a product of MSC.Software Corporation Version: x.x Released: July 1, 20xx. Copyright 20xx, MSC.Software Corporation, Santa Ana, CA, USA.

$TITLE: PARALLEL SEMIINFINITE PLATES OPEN TO LEFT AND RIGHT. Beginning of control data echo. $PATH: $MESSAGE_FILE: VF3.MSG $STATUS_FILE: VF3.STAT $DIAGNOSTIC_FILE: VF3.DIAG $TITLE: PARALLEL SEMIINFINITE PLATES OPEN TO LEFT AND RIGHT. $RESTART_FILE: VF3.STRT $IN_DATA: VFIN.DAT $TEMPLATE_FILE: TEMPLATE.JUNK $RAW_DATA: VF3.RAW $OUT_DATA: VF3.RES $RAD_NODE_FILE: VF3.NOD $RUN_CONTROL: 0 $RESTART_FLAG: 0 $CONVERGE: -1.000000000 $ZERO: 0.0000000000E+00 $APPROX_CURVE: 0.1000000015 $GAUSS_ORDER: 8 8 8 $AXISYM_SURFACE: 5 16 End of control data echo. VF has successfully completed initialization. Beginning to read/process the INPUT data. $TITLE: PARALLEL PLATES, SIMPLE, FINE, LINEAR HEAT AT NODES. $TITLE: 13-MAY-88 10:11:25 2.3 Beginning to read/process the NODE data. Completed reading/processing the NODE data. Beginning to read/process the ENCLOSURE 1 data. Completed reading/processing the ENCLOSURE 1 data

Main Index

106

Patran Viewfactor Analysis Reviewing the Viewfactor Output

Beginning the obstruction and viewfactor calculations Calculations completed for surface 1 of 18 in this Calculations completed for surface 3 of 18 in this Calculations completed for surface 5 of 18 in this Calculations completed for surface 7 of 18 in this Calculations completed for surface 9 of 18 in this Calculations completed for surface 11 of 18 in this Calculations completed for surface 13 of 18 in this Calculations completed for surface 15 of 18 in this Calculations completed for surface 17 of 18 in this

for ENCLOSURE 1. enclosure. enclosure. enclosure. enclosure. enclosure. enclosure. enclosure. enclosure. enclosure.

Completed the obstruction and viewfactor calculations for ENCLOSURE 1. The viewfactor calculations are done and the data is in the raw data file. Beginning translation of raw viewfactor data into Patran Thermal resistors. Beginning to read/process the Template data. Completed reading/processing the Template data. Beginning to read/process the VIEWFACTOR data. Beginning to read/process the NODE data. Completed reading/processing the NODE data. Beginning to read/process the ENCLOSURE 1 data. * * * * > > > > E R R O R < < < < * * * * An error has occurred in subroutine VFSRPR. The UID = 12 on ENCLOSURE 1 is not in THE VFAC TEMPLATES. Execution will be aborted. * * * * > > > > E R R O R < < < < * * * * Returned from subroutine VFSRPR with an error. * * * * > > > > E R R O R < < < < * * * * Returned from subroutine VFENRS with an error. * * * * > > > > E R R O R < < < < * * * * Returned from subroutine VFTORS with an error. * * * * > > > > E R R O R < < < < * * * * Returned from subroutine VFRES with an error. * * * * > > > > E R R O R < < < < * * * * Returned from subroutine VF1 with an error. Execution will be aborted.

VFDIAG, the Viewfactor Diagnostic Data File The numeric diagnostic data is more difficult to review since there are no simple rules for determining acceptable values. In general, for a closed enclosure, the sums of the viewfactors should not deviate much from one. How much is too much is left to the discretion of the user and will depend on how many surfaces are in the enclosure, how complicated the obstructions in the enclosure are, and on the VFCTL parameters for convergence and integration order $CONVERGE:, $GAUSS_ORDER:, and $AXISYM_SURFACE:. For an enclosure with symmetric images, if the object with its symmetric images is closed, then sums of viewfactors significantly greater than one indicate that too many symmetric images were created. If the sums are significantly less than one, then this may indicate that not enough symmetric images were created. Here a significant deviation from one is on the order of the inverse of the number of symmetric images.

Main Index

Chapter 5: Analysis 107 Reviewing the Viewfactor Output

If the enclosure is open, the sums for surfaces which have a view of the opening should be somewhat less than one. In some simple geometries, you may be able to calculate the view of the opening for some of the surfaces. This number may be compared with the value of one minus the sum for a very good test of the accuracy of the viewfactor analysis. For larger problems, the statistical information at the end of each enclosure in the VFDIAG file provides a convenient summary of the diagnostic data for that enclosure. The nature of this data was explained in VFDIAG, 101. Your own discretion must be used when evaluating this data. If you deem the accuracy of the analysis to be inadequate, then this may be remedied by increasing the maximum integration order, reducing the convergence criteria, and refining the finite element mesh in the model. Unfortunately, there are no guarantees here and no one particular method can be counted on to work in every situation. The following is an example of a VFDIAG file. The enclosure here is the interior of a hollow cube. The faces were modeled with a very coarse mesh and have both triangular and quadrilateral faces. There are 9 surfaces in the model. Since this is a closed enclosure, we expect the sums of the viewfactors should be very close to one. The diagnostic data indicates that for the surfaces the sums are very close to one. However, the sums for the nodal subareas tend to deviate from one by several percent. You might reduce this error by increasing the integration order, decreasing the convergence criteria, or refining the coarse mesh in the model.

Example VFDIAG File $TITLE: $TITLE: $TITLE: $ENCL:

MSC Viewfactor VER. x.x 26-MAY-xx 14:40:18 THIS IS A TEST. TEST DATA SET 001

1 9 1 1 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.1001562119E+01 0.1014224052E+01 0.9842141867E+00 2 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.1011039257E+01 0.1010891438E+01 0.9780694246E+00 3 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.1022326946E+01 0.9840516448E+00 0.9936221242E+00 4 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.1030953050E+01 0.9899941683E+00 0.9790534377E+00 5 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.9846858978E+00 0.1020451188E+01 0.9948636293E+00 6 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.1023659945E+01 0.9868957400E+00 0.9894446135E+00 7 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.9617350698E+00 0.1007973313E+01 0.1016313314E+01 0.1013978839E+01 8 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.1016294003E+01 0.9553130269E+00 0.1006751895E+01 0.1021641731E+01 9 0.1000000119E+01 -0.1192092896E-06 0.0000000000E+00 0.1015009522E+01 0.9638026357E+00 0.1008314252E+01 0.1012874246E+01 0.1192093E-06 -0.1192093E-06 0.0000000E+00 0.1192093E-06 0.0000000E+00 0.3826493E-01 -0.7473979E-02 0.2187306E-01 0.1938043E-01 0.1089598E-01 0.4468697E-01 0.7378088E-02 0.2276568E-01 0.1927586E-01 0.1270878E-01 0.2193058E-01 0.5483680E-02 0.1349805E-01 0.1245689E-01 0.6438631E-02 0.2164173E-01 -0.5388313E-02 0.8427733E-02 0.5388313E-02 0.8427733E-02 $ENDENCL: $EOF:

Main Index

108

Patran Viewfactor Analysis Reviewing the Viewfactor Output

Main Index

Chapter 6: Post-Analysis Patran Viewfactor Analysis

6

Main Index

Post-Analysis



Introduction



Interface From Viewfactor to Patran Thermal



noteplains on Resistor Values



THERMAL Analysis



THERMAL Results Postprocessing

110

116

117 118

111

110

Patran Viewfactor Analysis Introduction

6.1

Introduction Post-analysis, includes the activities that normally follow a Viewfactor analysis. These activities primarily involve interfacing to Patran Thermal for subsequent thermal network analysis. This chapter assumes that you have used Viewfactor to create Patran Thermal radiation resistors and now want to include them in a thermal analysis. Viewfactor may also be used to calculate viewfactors only and not make Patran Thermal radiation resistors, or be used to take already existing viewfactor data and combine it with Patran Thermal template file data to produce radiation resistors. These different modes of operation are determined by the $RUN_CONTROL parameter. For more information on these different modes of operation, see Description, 13, VFCTL, the Viewfactor Program Execution Control File, 85, Review Directory for Required Files, 98 and Changing the Surface Template Data After Viewfactors are Calculated (Ch. 7).

Main Index

Chapter 6: Post-Analysis 111 Interface From Viewfactor to Patran Thermal

6.2

Interface From Viewfactor to Patran Thermal There are two possible interfaces for the Viewfactor results in the form of the radiative resistors in the VFRESDAT file and the radiosity nodes in the VFNODEDAT file to the Patran Thermal analysis module. The first of these is the use of the data as input to the thermal network analysis code QTRAN and the second is through a translator in PATQ to change the binary form of the radiative resistor file to text form readable by the user. Refer to the MSC.Patran Thermal User’s Guide, Volume 1: Thermal/Hydraulic Analysis for more information on PATQ and QTRAN. Note:

The following steps are automatically handled if execution is spawned from within the Patran analysis form.

Viewfactor VFRESDAT and VFNODEDAT Files as Input to Patran Thermal’s QTRAN To include the thermal radiation network data from Viewfactor in the Patran Thermal analysis by QTRAN, the user must include the two files from Viewfactor (VFRESDAT and VFNODEDAT) in the QINDAT file for QTRAN. This is typically done by using QTRAN’s $INSERT command in the QINDAT file. If you are not using the standard filenames, then you need to substitute the filenames in use at the time. The VFNODEDAT file should be included in the QTRAN QINDAT. See VFNODEDAT, 102. This is done with the line

$INSERT VFNODEDAT This line must be in the QINDAT DEFNOD section, that is before the $ sign terminating the section and typically after the $INSERT NODEDAT line. The VFRESDAT radiation resistor data file should be included in the QTRAN QINDAT, VFMSG, the Viewfactor Message File, 103: RESISTOR DATA SETS of the QINDAT file. This is done with the line

$INSERT VFRESDAT,RAD A sample QINDAT file is shown with the parts relevant to the Viewfactor files VFRESDAT and VFNODEDAT shown in bold. This example was created on a VAX VMS computer platform. If you have a different platform, see your MSC.Patran Thermal User’s Guide, Volume 1: Thermal/Hydraulic Analysis for the system dependent file names. Note:

The QINDAT file is automatically created when execution is spawned from Patran and is described in the MSC.Patran Thermal User’s Guide, Volume 1: Thermal/Hydraulic Analysis.

Translating Binary Resistor File VFRESDAT to a Text File, VFRESTXT This capability is provided so that you may examine the contents of the radiation resistor file in text form. In general, this is not very practical due to the very large number of resistors created for the typical thermal analysis model. Procedure in Patran Thermal’s PATQ to Translate Binary Files to Text The MSC.Patran Thermal User’s Guide, Volume 1: Thermal/Hydraulic Analysis provides more complete information on the Patran Thermal program PATQ. Here you need only be concerned with the aspects of PATQ which relate to translating a VFRESDAT file in binary form to a VFRESTXT file in text form. The translation process is fairly simple. PATQ is an interactive menu-driven program with prompts for user input when required. Many of the prompts have default values that will be used if just a carriage return is entered. PATQ also generates diagnostic error messages if it detects any problems during the interactive session. You are cautioned that the resulting text file, VFRESTXT, will be much larger than the binary version of the data. You may wish to ascertain if this much disk space is available. If the files are of significant size, the translation may take a significant amount of time and this process is interactive. The procedure of translating a binary VFRESDAT file to a text file VFRESTXT using PATQ is fairly simple and is reproduced here as a sample.

Main Index

112

Patran Viewfactor Analysis Interface From Viewfactor to Patran Thermal

There are of course many different options in the translation process and they cannot all be presented here. The following dialogue is typical of a PATQ translation. If errors are detected by PATQ, error messages will be written to the terminal screen.

$PATQ

At the system level, run PATQ.

PATRAN <--> Q/TRAN <--> Viewfactor Preference Module Version x.x Release Date: 4/1/xx @ 14:00 Please Enter the Desired Option: 1 --> Quit 2 --> Read a PATRAN Neutral File and Generate Q/TRAN and Viewfactor Input Data File Segments 3 --> If necessary to create viewfactors, Submit VIEWFACTOR Code. 4 --> Generate a new Q/TRAN Main Program 5 --> Submit QTRAN for Compilation, Linkage, and Execution 6 --> Select additional PATQ Utility Options

Main Index

Chapter 6: Post-Analysis 113 Interface From Viewfactor to Patran Thermal

Main Index

114

Patran Viewfactor Analysis Interface From Viewfactor to Patran Thermal

PATQ Utilities Menu Please Enter the Desired Option: 1 --> Return to Main Menu 2 --> Access the Material Property Data Base Utilities 3 --> Convert a Q/TRAN Output File to PATRAN Nodal Results File(s) 4 --> Convert a Q/TRAN Output File to PATRAN Neutral Files 5 --> Generate Temperature vs. Time Plot Files 6 --> Convert CONDUC.DAT, VFRES.DAT, CAP.DAT or QPLOT.DAT Files from Binary to Text 7 --> Map Temperatures from one Neutral File to Another 8 --> Convert a Nodal Results File to a Neutral File 9 --> Lump QIN.DAT with the $INSERT Files to create a Single Bulk Data File. 10 --> Report the Times in Nodal Results Files. 11 --> Create LCI Material Properties from other Material Types

>6

Choose menu pick number 6 to translate a binary VFRESDAT file to a text VFRESTXT file. 1 2 3 4 5

--> --> --> --> -->

Convert Convert Convert Convert Return

>2

Binary Binary Binary Binary

Conductive Resistor File Radiative Resistor File Capacitor File Plot File

Enter the appropriate code for converting a binary radiative resistor file.

Please enter the name of the binary file to be converted.

>VFRES.DAT

Type the name of the desired binary file.

Please enter the name of the new text file to be created from the binary file.

>VFRES.T XT 1 2 3 4 5

--> --> --> --> -->

Convert Convert Convert Convert Return

Type the name of the desired text file. Binary Binary Binary Binary

Conductive Resistor File Radiative Resistor File Capacitor File Plot File

>5

Enter the appropriate code to return to the main menu.

>1

Press RETURN or ENTER to continue. Enter 1 RETURN to quit menu.

PATRAN <--> Q/TRAN <--> Viewfactor Preference Module Version 2.5 Release Date: 4/1/91 @ 14:00 Please Enter the Desired Option: 1 --> Quit 2 --> Read a PATRAN Neutral File and Generate Q/TRAN and Viewfactor Input Data File Segments 3 --> If necessary to create viewfactors, Submit VIEWFACTOR Code. 4 --> Generate a new Q/TRAN Main Program 5 --> Submit QTRAN for Compilation, Linkage, and Execution 6 --> Select additional PATQ Utility Options

>1

Main Index

The translation is now complete. Enter 1 (Quit) to exit PATQ.

Chapter 6: Post-Analysis 115 Interface From Viewfactor to Patran Thermal

The VFRESTXT Resistor Text File Refer to the MSC.Patran Thermal User’s Guide, Volume 1: Thermal/Hydraulic Analysis for help in interpreting the data in the VFRESTXT file.

Main Index

116

Patran Viewfactor Analysis noteplains on Resistor Values

6.3

noteplains on Resistor Values When Viewfactor makes resistors from a surface to the ambient environment node, it calculates the view to the ambient environment by summing the viewfactors from the surface to all other surfaces in the enclosure and subtracting from one. If the opening in the enclosure is small, it is possible to obtain a small negative number by this procedure due to numerical and round-off error. This will result in a resistor with a negative value. This is not a problem for the network analyzer QTRAN and you should not become unduly alarmed about these negative valued resistors. They are retained because they are the best approximation of the correct value.

Main Index

Chapter 6: Post-Analysis 117 THERMAL Analysis

6.4

THERMAL Analysis Thermal analysis is covered in detail in the MSC.Patran Thermal User’s Guide, Volume 1: Thermal/Hydraulic Analysis. You are referred there for a complete discussion of thermal analysis using the Patran Plus family of products. The basic procedure is to enter the PATQ program and choose menu pick number 4, Generate a new QTRANFOR. After exiting PATQ, you submit the thermal analysis for compilation, linking, and execution by the Patran Thermal command QTRAN. Status and progress of the thermal analysis may be monitored using the Patran Thermal command QS. Note:

Main Index

These commands are computer system dependent. They are the standard commands on a VAX/VMS system. If you have a different system or have customized your installation, then your commands may be different. Refer to the MSC.Patran Thermal User’s Guide, Volume 1: Thermal/Hydraulic Analysis or your system manager.

118

Patran Viewfactor Analysis THERMAL Results Postprocessing

6.5

THERMAL Results Postprocessing Postprocessing the thermal analysis results is described in detail in the Patran Thermal User’s Guide and postprocessing in general is described in the Patran Reference Manual. Basically, Patran Thermal will produce Patran nodal results or neutral files for use by Patran. You should note that the radiosity nodes, being nonphysical, are not part of the Patran geometric model and hence these nodes and the results at these nodes cannot easily be displayed in Patran Plus. The results at these nodes may be examined in the QOUTDAT file. Also, the QOUTDAT file has additional information on the radiation resistors and the heat flow through these resistors which you may wish to examine. The QOUTDAT file is documented in the Patran Thermal User’s Guide.

Main Index

Chapter 7: Changing the Surface Template Data After Viewfactors are Calculated Patran Viewfactor Analysis

7

Main Index

Changing the Surface Template Data After Viewfactors are Calculated 

Introduction



Compatible VFAC LBC and Template Data



New Resistors from Raw Viewfactor Data

120 121 122

120

Patran Viewfactor Analysis Introduction

7.1

Introduction Viewfactor has been carefully designed to segregate the viewfactor calculations (which depend only on the geometry of the model and are completely independent of the material properties) from that part of the program which creates the Patran Thermal radiation resistors (by combining the viewfactors and material property data). This gives you the luxury of calculating the viewfactors once for a given model. Then if you change materials or material properties, you can get the new Patran Thermal radiation resistors without having to recalculate the viewfactors. The viewfactor calculations typically take a lot of computer time. This capability provides a large saving in computer time for models in which the radiative material properties are modeled for various levels of sophistication or for models in which various materials or surface properties are being evaluated.

Main Index

Chapter 7: Changing the Surface Template Data After Viewfactors are 121 Calculated

7.2

Compatible VFAC LBC and Template Data There are certain compatibility requirements for the VFAC boundary condition and VFAC templates. These requirements are discussed in detail in Compatibility Requirements for Model and VFAC Templates, 65. If you plan to change the material properties and make new Patran Thermal radiation resistors without having to recalculate the viewfactors, it is especially important that these compatibility requirements be satisfied for all of the material properties to be used, and that the VFAC boundary condition data do not need to be changed to model the various materials.

Main Index

122

Patran Viewfactor Analysis New Resistors from Raw Viewfactor Data

7.3

New Resistors from Raw Viewfactor Data New and different Patran Thermal radiation resistors may be created by changing the Patran Thermal material property definitions in the VFAC templates and in the material property data files where the MPIDs are defined, typically in the Patran Thermal file MATDAT. If the VFAC templates are changed, then Viewfactor must be used to create new Patran Thermal radiation resistors and radiosity nodes. If the VFAC templates are not changed and only the Patran Thermal material property definitions are changed (but not the MPIDs), then you do not need to recreate the Patran Thermal radiation resistors and radiosity nodes. This is because Viewfactor only uses the VFAC templates and the raw viewfactor data to create the Patran Thermal radiation resistors and radiosity nodes. If the templates and viewfactors are not changed, the resistor will not change.

Changing TEMPLATEDAT VFAC Templates Changes in the VFAC templates are typically entered from a text editor. The VFAC templates are described more fully in Relationship of VFAC LBC Data to VFINDAT File Data, 51 and in the Patran Thermal User’s Guide.

If the VFAC templates are changed, then Viewfactor must be rerun to recreate the Patran Thermal radiation resistors and radiosity nodes. Note that when you change the VFAC templates you need not recalculate the viewfactors. You only need to recreate the resistor data. Instructions for running Viewfactor in this mode are given in Compatible VFAC LBC and Template Data, 121. Once again, remember the compatibility requirements for VFAC LBCs and VFAC templates referred to in the previous section and Patran Thermal TEMPLATEDAT Files for Surface Property Description, 52.

Changing Patran Thermal Material Definitions The material property definitions are part of Patran Thermal. Refer to the Patran Thermal User’s Guide for more information. As long as the material property ID, MPID, and VFAC template are not changed, the existing Patran Thermal radiation resistors and radiosity nodes will be valid. This allows for wide latitude in material properties, provided compatibility was insured when the model was created. For example, a material property which was originally modeled as a constant in its MPID definition may be changed to a temperature dependent function merely by changing it's MPID definition in the Patran Thermal MATDAT file. No changes need to be made to the radiative resistors or radiosity nodes.

Changing VFCTL If you wish to recreate Patran Thermal radiation resistors and radiosity nodes and not recalculate the viewfactors, then the run control parameter, $RUN_CONTROL:, in the VFCTL file must be set to the value 2.

Submitting the New Viewfactor Job All Viewfactor jobs are submitted in the same manner. Refer to Submitting a Viewfactor Job for Analysis, 98 for detailed instructions on submitting the Viewfactor job to recreate the Patran Thermal radiation resistors and radiosity nodes.

Main Index

Chapter 8: Theory and Computational Limitations Patran Viewfactor Analysis

8

Main Index

Theory and Computational Limitations



Introduction



Viewfactor



Mean Beam Length



Obstructions



Computational Limitations

124 125 126

128 129

124

Patran Viewfactor Analysis Introduction

8.1

Introduction The present Viewfactor program makes commercially available advances in the computer calculation of viewfactors and obstruction for finite element thermal models developed at MSC.Software Corporation. These advances are principally: 1. Improved performance as measured by the product of speed and accuracy in determining obstructed views and calculating viewfactors, and 2. A viewfactor analysis program designed specifically for finite element modeling.

Main Index

Chapter 8: Theory and Computational Limitations 125 Viewfactor

8.2

Viewfactor The finite element model for the viewfactor associated with the ith node on surface e to the jth node on surface f is represented by the following formula:

Fijef Z Ó

Nie N if ( r ef • nˆ e ) ( r ef • nˆ f ) dA e dA f ∫ ∫ e ∫ ∫ f -----------------------------------------------------------------------------------------------------------------ef 4 π r A A -------------------------------------------------------------------------------------------------------------------------------------------------------------e ∫ ∫ e Ni A

subject to the restrictions that ( r e f • nˆ e ) ( r ef • nˆ f )

ref

⎛ ⎜ ⎝



does not intersect any view obstructing surfaces and the product

is negative, and

e

=

Surface e ID

f

=

Surface f ID

i

=

Node ID on surface e

j

=

Node ID on surface f

Ae

=

Surface e area

Af

=

Surface f area

N ie

=

Primary variable finite element interpolation function associated with the ith node on surface e

r ef

=

nˆ e

= =

(8-1)

dA e⎞⎟

Vector from a point on surface e to a point on surface f Unit normal to surface e Magnitude of a vector

The right-hand side of this equation is evaluated using numerical methods. The numerical integration scheme is predominantly Gaussian quadrature. The quadrature order may be varied by the Viewfactor analysis program in an attempt to obtain the desired accuracy. The method for estimating the accuracy of the numerical integration is, of course, empirical, but seems to work very well. In general, if sufficient accuracy has not been obtained, then the quadrature order will be increased in an effort to improve the accuracy. The quadrature order is not increased globally throughout the integration domain, but only in those areas where the program determines the most benefit will result.

Main Index

126

Patran Viewfactor Analysis Mean Beam Length

8.3

Mean Beam Length For the enclosure with an isothermal participating medium (in equilibrium with itself), we may define a quantity analogous to the viewfactor above, called the exchange factor, by the equation:

I

o ijef

Z Ó

exp ( Ó κ r ef ) N ie Nif ( r ef • nˆ e ) ( r ef • nˆ f ) dA e dA f ∫ ∫ ∫ ∫ f ---------------------------------------------------------------------------------------------------------------------------------------------------------ef 4 e π r A A --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------e ∫ ∫ e Ni A

subject to the restrictions that ( r e f • nˆ e ) ( r ef • nˆ f )

ref

(8-2)

dA e

⎞ ⎟ ⎠

does not intersect any view obstructing surfaces and the product

is negative, and

e

=

Surface e ID

f

=

Surface f ID

i

=

Node ID on surface e

j

=

Node ID on surface f

Ae

=

Surface e area

Af

=

Surface f area

k

⎛ ⎜ ⎝

media extinction coefficient

N ie

=

r ef

=

nˆ e

= =

Primary variable finite element interpolation function associated with the ith node on surface e Vector from a point on surface e to a point on surface f Unit normal to surface e Magnitude of a vector

For the optically thin media, the exponential function in the above equation may be approximated by: ex p ⎛⎝ Ó κ

r ef ⎞⎠

Z 1 Ó κ r ef H hi gh er o rde r t erm s

(8-3)

Using this approximation and assuming that the extinction coefficient is constant (or close enough to constant that it may be taken outside the integration), define the mean beam length by the equation:

Rijef Z Ó

N ie N if ( r ef • nˆ e ) ( r ef • nˆ f ) dA e dA f ∫ ∫ e ∫ ∫ f -----------------------------------------------------------------------------------------------------------------ef 3 π r A A --------------------------------------------------------------------------------------------------------------------------------------------------------------

(8-4)

e ⎛ e⎞ ⎜ dA ⎟ ⎠ ∫ ∫ e Ni ⎝ A

Then in Patran Thermal the exchange factor is approximated by the equation:

I

o ijef

Z e xp ( Ó κ R iejf )F ijef

(8-5)

In this way the geometry of the model is isolated in the viewfactor and mean beam length quantities. The material properties, such as the extinction coefficient, may be time and temperature dependent functions in Patran Thermal without requiring recalculation of the viewfactor and mean beam length. This results in great saving of computer processing time.

Main Index

Chapter 8: Theory and Computational Limitations 127 Mean Beam Length

The assumptions made for the participating media model in Viewfactor and Patran Thermal are as follows: 1. The media is isothermal; 2. The media is in equilibrium with its internal energy; 3. The media extinction coefficient is not a function of position; 4. The optical thickness is small enough that the linear approximation of the exponential function is reasonable and the media is only weakly interacting with itself.

Main Index

128

Patran Viewfactor Analysis Obstructions

8.4

Obstructions It is understood that the finite element discretization of the model geometry represents the finest geometric detail available in the model. Therefore, surfaces are not subdivided beyond the nodal subareas for obstruction checking. This restriction on obstruction checking may be viewed as a bit too severe. Different schemes have been tried. The error introduced by this discretization of the obstruction checking process was compared with the error introduced in the viewfactors by the finite element discretization. Considering the cost in computer execution time for a finer obstruction discretization, it was better to limit the obstruction checking to nodal subareas. You have control over the accuracy of the obstruction test in the usual way that accuracy in a finite element model is controlled (i. e., by refining the finite element mesh). Thus, if more accuracy is needed in the obstruction checking, you should refine the finite element mesh in the regions where greater accuracy is desired.

Main Index

Chapter 8: Theory and Computational Limitations 129 Computational Limitations

8.5

Computational Limitations There are three areas in which we know the computational powers of the Viewfactor program are limited. These are not severe limitations and most users would probably never discover them. However, in the interest of completeness, they are explained in the following sections.

Grazing Incidence of the Intersurface Ray with the Surface The grazing incidence of an intersurface ray with a potentially obstructing surface may not be correctly determined. This is due to the finite precision of computer arithmetic. There is no hard and fast rule for when this problem will occur. Generally it is not a problem unless the incidence with the surface is somewhat less than 10-5 radian for computations on typical 32 bit computers.

Spatial Resolution The obstruction checking algorithm in Viewfactor is unable to detect potential obstructions which are very near one of the surfaces for which the view between is being checked for obstructions. Here “very near” means that the distance from the potential obstruction to one of the surfaces in the viewing pair is less than about one five-thousandth of the distance between the pair of viewing surfaces. This situation is shown schematically in Figure 8-1.

Obstruction Surface 1 Surface 2

View from Surface 1 to Surface 2

In this case, the obstruction may not be detected.

<1 Unit >5000 Units of Distance Distance

Figure 8-1

Spatial Resolution in VIEWFACTOR

Extreme Scales Models which have very large or very small numbers for their dimensions will cause an arithmetic overflow or underflow condition to occur in Viewfactor. The approximate limits at which this occurs depends on the dimensionality of the model. For the 2-D XY model, the limits are approximately the square root of the largest and smallest numbers representable in the default FORTRAN single precision variable. For the 2-D RZ axisymmetric model, these limits are approximately the cube root of the largest and smallest numbers representable in the default FORTRAN single precision variable. For the 3-D XYZ model, these limits are approximately the fourth root of the largest and smallest numbers

Main Index

130

Patran Viewfactor Analysis Computational Limitations

representable in the default FORTRAN single precision variable. If the user’s model has dimensions which exceed these upper or lower limits it will be necessary to scale the model by a suitable factor in order to avoid arithmetic overflow or underflow in Viewfactor. Viewfactor does not check for overflow and underflow conditions, since for most users this will not be an issue.

Main Index

Chapter 9: Data File Specifications Patran Viewfactor Analysis

9

Main Index

Data File Specifications



Introduction



VFINDAT (Input Data File)



VFRAWDAT (Raw Viewfactor Data)



VFRESDAT (Resistor Data)



VFDIAG (Diagnostic Data)



TEMPLATEDAT (Surface Pointer Data)



VFNODEDAT (Radiosity Node Lists)

132 133 140

144 145 149 151

132

Patran Viewfactor Analysis Introduction

9.1

Introduction A computer programmer may wish to write an interface to the Viewfactor data files. These data file specifications are intended to provide the information necessary. Unless you wish to examine the contents of these files, you don’t need to be concerned with this chapter.

Main Index

Chapter 9: Data File Specifications 133 VFINDAT (Input Data File)

9.2

VFINDAT (Input Data File) This file contains the data describing the model for which viewfactors are to be calculated. It is identified by the $IN_FILE: keyword in the Viewfactor command file. Its default name is VFINDAT. This file is currently a sequential, formatted (ascii) file. We may optionally provide this file in sequential, unformatted (binary) form sometime in the future to better accommodate large models. The file contains comments, keyword lines, and numeric data. Comments are blank lines or lines where the first nonblank character is *. Comments may not immediately precede or be interspersed with numeric data. Comments may immediately precede any keyword line. Keyword lines consist of a keyword which may be followed by data as described in more detail below. All keyword lines are required and must be in the order shown. The number of occurrences permitted for different keywords is described in more detail below. The valid keywords and order are: $TITLE: $SIZE: $SYM: $ENDSYM: $NODES: $ENDNODES: $ENCL: $ENDENCL: $EOF: Leading blanks may precede comments and keywords. Comments and keyword lines are read by a FORTRAN '(A)' format into a buffer 132 characters long. Thus, comments and keyword lines (including leading blanks) should not exceed 132 characters. Numeric data is associated with keywords and may occur on the keyword line or on line(s) immediately following the keyword line. Note that comments are not permitted immediately preceding or interspersed with numeric data. Numeric data is in a fixed format which will be described in detail (see Examples, 133).

Examples Here are two examples of model input files:

Main Index

134

Patran Viewfactor Analysis VFINDAT (Input Data File)

Example 1 *Sample VFINDAT file. $TITLE: TEST DATA SET 003 * DIMCOD NNODE1 $SIZE: 2 6 * NSYMOB $SYM: 1 1.0 2.0 2.0 0.0 0.0 $ENDSYM: $NODES: 1 0.0 2 1.0 3 1.0 4 0.0 9 0.6 10 1.6 $ENDNODES: * ENCLID NSURF1 $ENCL: 1 4 1 7 1 0 0 1001 1.5 1 2 2 7 1 0 0 1001 1.5 2 3 3 7 1 0 0 1001 1.5 3 4 4 7 1 0 0 1001 1.5 4 1 $ENDENCL: $EOF:

Main Index

NENCL1 1

MXNODN 10

0.0 0.0 2.0 0.0 0.0

0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 1.0 0.6 0.6

0.0 0.0 0.0 0.0

2 0

2 0

1 0

0 1

2 0

2 0

1 0

0 1

2 0

2 0

1 0

0 1

2 0

2 0

1 0

0 1

2.5

2.5

2.5

2.5

Chapter 9: Data File Specifications 135 VFINDAT (Input Data File)

Example 2 *Sample VFINDAT file. $TITLE: TEST DATA SET 001 * DIMCOD NNODE1 NENCL1 $SIZE: 3 12 2 * NSYMOB $SYM: 0 $ENDSYM: $NODES: 1 0.0 2 1.0 3 1.0 4 0.0 5 0.0 6 1.0 7 1.0 8 0.0 9 0.1 10 1.1 11 1.1 12 0.1 $ENDNODES: * Beginning of first enclosure data. * $ENCL: 1 0 1 2 0 5 3 0

ENCLID 1 16 0 1.5 2 16 0 1.5 8 16 0 1.5 5

NSURF1 3 1 1001

ENCLID 17 16 0 1.5 2 16 0 1.5 4

NSURF1 2 1 1001

5 0 3 6 0 1 $ENDENCL: $EOF:_

Main Index

0.0 0.0 1.0 1.0 0.0 0.0 1.0 1.0 0.6 0.6 1.6 1.6

0.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 0.5 0.5 0.5 0.5

2 0

4 0

1 0

0 1

4 2 0

4 0

1 0

0 1

6 2 0

4 0

1 0

0 1

2 0

4 0

1 0

0 1

7 2 0

4 0

1 0

0 1

2.5 3 1 1001 2.5 7 1 1001

2.5 1 6 $ENDENCL: * Beginning of second enclosure data. * $ENCL:

MXNODN 12

2

2.5 6 1 1001 2.5 8

5

136

Patran Viewfactor Analysis VFINDAT (Input Data File)

Detailed Descriptions $TITLE The first noncomment line must be: $TITLE: title Multiple $TITLE: lines are allowed, but they must all occur before the $SIZE: line. Comments between multiple $TITLE: lines are allowed. $SIZE The $SIZE: card must be the next noncomment line. One and only one $SIZE: line is allowed. The format is: blanks, A10, 4I10, e. g., *bbbA10 I10 I10 $SIZE: DimCode

I10 #Node

I10 #Encl MaxNod#

Parameter

Description

DimCode

Code identifying geometric space of the model, -2= Axisymmetric RZ, 2= 2-D XY, 3= 3-D XYZ;

#Node

Number of nodes in this model (before symmetry operations);

#Encl

Number of enclosures in this model (before symmetry operations);

MaxNod#

Maximum node ID referenced in the model (before symmetry operations). $SYM and $ENDSYM The $SYM: card must be the next noncomment line. The format is:

blanks, A10, I10, e. g., *bbbbbbbbA10 I10 $SYM: #SymObj Parameter

Description

#SymObj

Number of symmetry objects defining symmetry operations for this model. The present maximum allowed value is 4. Immediately following and with no interspersed comments is the symmetry object numeric data, which is read by: READ( LU,'( 3 (E20.10) )',IOSTAT = IOS, END = 1, ERR = 1 ), $ ( ( ( SYMOBJ( K, J, I ), K = 1, 3 ), J = 1, 5 ), I = 1, #SymObj The contents of SYMOBJ( 1:3, 1:5, I ) are: Type X1 X2 X3 #Rot

Dummy Y1 Y2 Y3 Rot_Angle

Dummy Z1 Z2 Z3 Dummy

This is repeated #SymObj times. The symmetry operations are performed in order of increasing index I. Depending on the value of Type, the data for each symmetry object has different meanings.

Main Index

Chapter 9: Data File Specifications 137 VFINDAT (Input Data File)

For IFIX( Type ) = 1 this is a reflection in 2-D XY or RZ space about a line. The line is defined by two distinct points, (X1,Y1) and (X2,Y2), on the line All other subarray elements should be set to 0.0. For IFIX( Type ) = 2 this is a reflection in 3-D XYZ space about a plane defined by three distinct points, (X1,Y1,Z1), (X2,Y2,Z2), and (X3,Y3,Z3), all on the plane and not colinear. All other subarray elements should be set to 0.0. For IFIX( Type ) = 3, this is a rotation about an axis and is valid in 2-D XY and 3-D XYZ spaces. The positive axis of rotation by the right-hand rule is the vector from (X1,Y1,Z1) to (X2,Y2,Z2) and should have nonzero length. For 2-D XY space, this axis must be perpendicular to the XY-plane. The model will be replicated #Rot times by successively rotating #Rot times the model existing before any of these rotations (but including that created by previous symmetry operations) Rot_Angle degrees about the axis of rotation. All other subarray elements should be set to 0.0. All other values of IFIX(Type) result in a fatal error in P/VF. Only certain combinations of symmetry object types are valid. The valid combinations depend on the model space, as identified by DimCode. In general, symmetry object validity is not checked by P/VF. For 2-D RZ space (DimCode = -2), the only valid symmetry object is Type = 1 (line) and this line must be perpendicular to the Z-axis. Only one such symmetry object is permitted in 2-D RZ space. For 2-D XY space (DimCode = 2), only types 1 (line) and 3 (rotation) are valid. Not more than one type 3 object is permitted. Not more than two type 1 objects are permitted. Type 1 objects should be in the XY-plane. Type 3 objects should be perpendicular to the XY-plane. If there are two type 1 objects, they should be mutually perpendicular. For 3-D XYZ space (DimCode = 3), only symmetry object types 2 (plane) and 3 (rotation) are valid. Not more than one type 3 object is permitted. Not more than three type 2 objects are permitted. Two or more type 2 objects must be mutually perpendicular. Comments may follow the symmetry object data, but may not occur between the $SYM: keyword line and the end of the symmetry object numeric data. The next required keyword line is $ENDSYM:. This line contains no other data. If there are no symmetry objects, the $SymObj is zero and the required data is: $SYM: 0 $ENDSYM: $NODES and $ENDNODES Next comes the node data. The required keyword is $NODES: with no other data on this line. Immediately following, with no interspersed comments is #Node (from the $SIZE: data line) lines of node data in I10, 3(E20.10) format as: NodeID

x-coordinate

y-coordinate

z-coordinate

z-coordinate

dummy

for 3-D XYZ and 2-D XY, or as: NodeID

r-coordinate

for 2-D RZ. Unused coordinate fields should be set to 0.0. A fatal error will occur if #Node and the number of data lines do not match. Comments may follow the data. The next keyword line after the node numeric data is $ENDNODES:, with no other data on the line. One and only one occurrence of $NODES: and $ENDNODES: is allowed.

Main Index

138

Patran Viewfactor Analysis VFINDAT (Input Data File)

The next keyword line is: $ENCL:

Enc1ID

#Surf

with format blanks, A10, 2I10. Parameter

Description

EnclID

The ID number of this enclosure.

#Surf

Number of surfaces in this enclosure. Immediately following, and with no interspersed comments, is the surface data for the #Surf surfaces. A fatal error will occur if the #Surf and the number of surface data do not match. The format and meaning of the surface data is:

Parameter

( 7 ( I10 ) ), ( SurAtt( J ), J = 1, MaxAtt ) Description

MaxAtt

Maximum number of surface attributes supported = 14;

SurAtt

Array of surface attributes whose elements are: (1)=

Surface ID number,

(2)=

Surface configuration (6=RZ bar, 7=XY bar, 13=3-D tri, 16=3-D quad),

(3)=

Surface order (1=linear),

(4)=

Number of real data values associated with this surface,

(5)=

Number of nodes associated with this surface,

(6)=

ID number of the element for which this surface is a face,

(7)=

ID number of the element face this surface represents,

(8)=

Participating media node number for this surface,

(9)=

Ambient radiation node number for this surface,

( 10 ) = ( 11 ) =

User ID (UID in Patran Thermal) for material properties, Obstruction flag (0=this surface will be checked for obstructing the view between other surface pairs, not 0= this surface will not be checked for obstructing the view between other surface pairs. Convex Surface ID (surfaces with identical nonzero values here cannot see each other due to being on the same convex surface, Dynamic Flag, Not currently used, Patran Set ID, followed by ( 4 ( E20.10 ) ), ( SurDat( J ) , J = 1, SurAtt( 4 ) )

( 12 ) = ( 13 ) = ( 14 ) =

SurDat

Real data associated with the surface (presently none is required); followed by ( 8 ( I10 ) ), ( SurNod( J ), J = 1, SurAtt( 5 ) )

SurNod

List of node IDs associated with the surface and in the order specified for the corresponding Patran elements. This pattern is repeated for each surface in the enclosure. Comments may follow the last line of the last surface data. The next keyword line is $ENDENCL: with no data.

Main Index

Chapter 9: Data File Specifications 139 VFINDAT (Input Data File)

The pattern: $ENCL: Numeric data $ENDENCL:

EnclID

#Surf

is repeated for any additional enclosures. The number of enclosure data sets should equal the #Encl data from the $SIZE: line. This is not a fatal error here since discrepancies will not be detected until the enclosures have been processed (viewfactors calculated). A warning message will be issued and attempts to create resistors for the viewfactor data may have undesirable results. $EOF The last keyword line is $EOF: with no data. Comments may follow the $EOF: line.

Main Index

140

Patran Viewfactor Analysis VFRAWDAT (Raw Viewfactor Data)

9.3

VFRAWDAT (Raw Viewfactor Data) The VFRAWDAT file is a FORTRAN Unformatted Sequential Access file. Since this is an unformatted file, its specific content is machine dependent and so we will only give an informal specification of its form. Each item in curly braces corresponds to a record in the file.

VFRAWDAT==::[{(Nchar)(Comment_String)}]* {(Nchar)(Title_String)} [ [{(Nchar)(Comment_String)}]* {(Nchar)(Title_String)}] ]* [{(Nchar)(Comment_String)}]* {(Nchar)(Size_String)} [{(Nchar)(Comment_String)}]* {(Nchar)(Symmetry_String)} [{(Symmetry_Data)}]*Number_Symmetry_Objects [{(Nchar)(Comment_String)}]* {(Nchar)(End_Symmetry_String)} [{(Nchar)(Comment_String)}]* {(Nchar)(Begin_Node_Data_String)} [ [{Node_Data_Record}]*Symmetry_Multiplier *Number_Nodes [{(Nchar)(Comment_String)}]* {(Nchar)(End_Node_Data_String)} [ [{(Nchar)(Comment_String)}]* {(Nchar)(Begin_Enclosure_String)} {(Number_User_ID)} [{(User_ID)}]*Number_User_ID [ [{(Surface_Data)}]*Symmetry_Multiplier ]*Number_Surfaces [ [ [{(Surface_Pair_Record)} [{(View_Factor_Data)}]*If_Can_See ]*Symmetry_Multiplier ]*Upper_Number_Surfaces {(Sum_View_Factor_Data)} ]*Number_Surfaces [{(Nchar)(Comment_String)}]* {(Nchar)(End_Enclosure_String)} ]*Number_Enclosures [{(Nchar)(Comment_String)}]* {(Nchar)(End_File_String)} (EOF)

Main Index

Chapter 9: Data File Specifications 141 VFRAWDAT (Raw Viewfactor Data)

Parameter

Description

Nchar

Integer number of characters in the following character string, less than 132.

Comment_String

Character string beginning with the blank character or the * character.

Title_String

Character string beginning with the characters $TITLE: and followed by optional title characters. Maximum length is 80 characters.

Size_String

Character string beginning with the characters $SIZE: and containing the following data in the (A6,4X,4I10) format: Dimension_Code, Number_Nodes, Number_Enclosures, and Max_Node_ID. Dimension_Code Integer code for the dimensionality of the model. 2 ==> 2-D XY space -2 ==> 2-D RZ axisymmetric space 3 ==> 3-D XYZ space Number_Nodes - Integer number of nodes in one symmetric image of the model. Number_Enclosure - Integer number of enclosures in the model, before symmetry operations. Max_Node_ID - Integer largest node ID in the model.

Symmetry_String

Character string beginning with the characters $SYM: and followed by the data, Number_Symmetry_Objects, and formatted as (A5,5X,I10). Number_Symmetry_Objects - Integer number of symmetry operations for which data will follow.

Symmetry_Data

15 real constants, unformatted, representing the symmetry operation. See the specification for the VFINDAT file for a description of this data. The data is given in column order.

antisymmetric_ String

Character string containing the characters $ENDSYM:

Begin_Node_Data String

Character string beginning with the characters $NODES: and followed by the data, Number_Nodes and Symmetry_Multiplier, and formatted as (A7,3X,2I10). Symmetry_Multiplier - Integer giving the total number of images of the model after all symmetry operations have been completed, includes the original image.

Node_Data_Record

(Node_ID)(XYZ_Coordinates). Node_ID - Integer node ID number. XYZ_Coordinates - 3 real constants giving the X, Y, and Z coordinates of the node. For 2-D models, the third number should be zero. For 2D RZ axisymmetric models, the first two coordinates are the R and Z coordinates, respectively.

End_Node_Data_ String

Character string containing the characters $ENDNODES.

Begin_Enclosure_ String

Character string containing the characters $ENCL: and followed by the data, Enclosure_ID, Number_Surfaces, and Symmetry_Multiplier, and formatted as (A6,4X,3I10). Enclosure_ID - Integer ID of this enclosure. Number_Surfaces - Integer number of surfaces in this enclosure.

Number_User_ID

Integer number of User IDs, or UIDs, which are used in this enclosure and which will follow.

User_ID

Integer user ID which is used in this enclosure.

Surface_Data

(SURATT)(SURDAT)(SURNOD)(SURARA)(LFTHND). SURATT - 14 Integers, see the description of surface attributes in the VFINDAT file specification.

Main Index

142

Patran Viewfactor Analysis VFRAWDAT (Raw Viewfactor Data)

Parameter

Description SURDAT - SURATT number four reals, see the description of surface data in the

VFINDAT file specification.

SURDAT - SURATT number four reals, see the description of surface data in the VFINDAT file specification. SURNOD - SURATT number five integers giving the nodes on this surface and in the order specified for the corresponding Patran Plus elements. SURARA - SURATT number five plus one reals giving the surface area and the nodal subareas for this surface. LFTHND - Logical, true is surface has left handed orientation. Surface_Pair_ Record

(Surface_E_Index)(Surface_F_Index)(Symmetry_Index) (Subdivision_Index)(Number_Subdivision)(Can_See_Flag) Surface_E_Index - Integer index to the first surface in the pair. Surface_F_Index - Integer index to the second surface in the pair. Symmetry_Index - Integer index to the symmetric image of the second surface. Subdivision_Index - Integer index to the surface subdivision, not currently used. Number_Subdivision - Integer number of subdivisions of the surface, not currently used. Can_See_Flag - Logical, set true if View_Factor_Data is to follow.

If_Can_See

Not a formal item in the VFRAWDAT file. It is used to indicate the number of View_Factor_Data records that should follow. If_Can_See is set to one if Can_See_Flag is true and set to zero otherwise.

View_Factor_Data

[ (Nodal_Can_See)(Nodal_View_Factor). (Nodal_Mean_Beam_Length). ]*Number_Nodes_Surface_E. ]*Number_Nodes_Surface_F. Nodal_Can_See - Logical that node on surface E can see node on surface F. Nodal_View_Factor - Real value of the viewfactor from nodal subarea on surface E to nodal subarea on surface F. Nodal_Mean_Beam_Length - Real value of the mean beam length from nodal subarea on surface E to nodal subarea on surface F. Number_Nodes_Surface_E - SURATT number five for surface E. Number_Nodes_Surface_F - SURATT number five for surface F.

Upper_Number_ Surfaces

Number_Surfaces minus the current outer repeat structure index minus one. This provides the upper triangular portion of the surface pair matrix as the part for which the VFRAWDAT file contains data.

Sum_View_Factor_ Data

(SUMSUR)(SUMONE)(SUMZRO)(SUMNOD). SUMSUR - Real value containing the sum of the viewfactors from this surface to all other surfaces in this enclosure. SUMONE - Real value equal to one minus SUMSUR. SUMZRO - Real value containing the sum of all viewfactors from this surface to all other surfaces which were set to zero by virtue of being less than the zero cutoff value. SUMNOD 7- [Real value containing the sum of the viewfactors from this nodal subarea on surface E to all other nodal subareas on all other surfaces in this enclosure]*. Number_Nodes_Surface_E.

Main Index

Chapter 9: Data File Specifications 143 VFRAWDAT (Raw Viewfactor Data)

Parameter

Description

End_Enclosure_ String

Character string containing the characters $ENDENCL:

End_File_String

Character string containing the characters $EOF:

EOF

FORTRAN ENDFILE marker.

Character

FORTRAN CHARACTER*(*) Type with length less than 132 characters.

Logical

FORTRAN default LOGICAL Type.

Integer

FORTRAN default INTEGER Type.

Real

FORTRAN default REAL Type.

Main Index

144

Patran Viewfactor Analysis VFRESDAT (Resistor Data)

9.4

VFRESDAT (Resistor Data) The VFRESDAT file is a FORTRAN Unformatted Sequential Access file. The file may optionally contain some comment lines at the beginning. These are typically used to carry along such information as problem title, but are otherwise ignored. Since this is an unformatted file, its specific content is machine dependent and so this is only an informal specification of its form. Each item in curly brackets corresponds to a record in the file.

Parameter

VFRESDAT==::[{(Nchar)(Comment_String)}]* {(6)($BEGIN)} [[{(Resistor Record)}]* (EOF) Description

Nchar

Integer number of character in String

String

A string of Nchar characters, FORTRAN type Character

6

Integer constant 6.

$BEGIN

Character constant string '$BEGIN'

Resistor Record

((Restyp)(Subtyp)(Node1)(Node2)(Node3)(MPID (Data1)(Data2)(Data3)(lambda1)lambda2)) Restyp - Character, R or W Subtyp - Integer, 1 through 12 Node1- Integer, positive, Node 1 in Patran Thermal radiation resistors. Node2 - Integer, positive, Node 2 in Patran Thermal radiation resistors Node3 - Integer, positive, Node 3 in Patran Thermal radiation resistors MPID - Integer, nonzero, Patran Thermal material property ID Data1 - Real, First real data for the Patran Thermal radiative resistor Data2 - Real, Second real data for the Patran Thermal radiative resistor Data3 - Real, Third real data for the Patran Thermal radiative resistor lambda1- Real, Beginning wave length for waveband for Restyp W resistors lambda2 - Real, Ending wave length for waveband for Restyp W resistor

Character

FORTRAN CHARACTER*1 Type

Integer

FORTRAN default INTEGER Type.

Real

FORTRAN default REAL Type.

EOF

FORTRAN ENDFILE marker.

Main Index

Chapter 9: Data File Specifications 145 VFDIAG (Diagnostic Data)

9.5

VFDIAG (Diagnostic Data) Introduction This file contains the diagnostic data for the viewfactors which were just calculated. It is identified by the $DIAGNOSTIC_FILE: keyword in the Viewfactor command file. Its default name is VFDIAG. This file is currently a sequential, formatted (ascii) file. The file contains comments, keyword lines, and numeric data. Comments are blank lines or lines where the first nonblank character is *. Comments may not immediately precede or be interspersed with numeric data. Comments may immediately precede any keyword line. Keyword lines consist of a keyword which may be followed by data as described in more detail below. All keyword lines are required and must be in the order shown. The number of occurrences permitted for different keywords is described in more detail below. The valid keywords and order are: $TITLE: $ENCL: $ENDENCL: $EOF: Leading blanks may precede comments and keywords. Comments and keyword lines are read by a FORTRAN '( A )' format into a buffer 132 characters long. Thus, comments and keyword lines (including leading blanks) should not exceed 132 characters. Numeric data is associated with keywords and may occur on the keyword line or on line(s) immediately following the keyword line. Note that comments are not permitted immediately preceding or interspersed with numeric data. Numeric data is in fixed format which will be described in detail (see Examples, 146).

Main Index

146

Patran Viewfactor Analysis VFDIAG (Diagnostic Data)

Examples Here are two examples of model diagnostic files: Example 1 $TITLE: $TITLE: $TITLE: $TITLE:

PDA Viewfactor VER. 2.3 7-JUN-88 16:14:05 PARALLEL SEMIINFINITE PLATES OPEN TO LEFT AND RIGHT. PARALLEL PLATES, SIMPLE, FINE, LINEAR HEAT AT NODES. 28-MAR-88 08:30:09 2.2K-X

$ENCL:

1 18 1 1 0.8375167847E+00 0.1624832153E+00 0.8469786048E+00 0.8280547857E+00 2 0.7995740771E+00 0.2004259229E+00 0.8067734241E+00 0.7923747301E+00 3 0.7514496446E+00 0.2485503554E+00 0.7604730725E+00 0.7424262762E+00 4 0.6918253303E+00 0.3081746697E+00 0.7027459145E+00 0.6809046268E+00 5 0.6208980680E+00 0.3791019320E+00 0.6335698366E+00 0.6082264185E+00 6 0.5413827300E+00 0.4586172700E+00 0.5551127791E+00 0.5276526213E+00 7 0.8375174403E+00 0.1624825597E+00 0.8280556202E+00 0.8469793200E+00 8 0.7995739579E+00 0.2004260421E+00 0.7923744321E+00 0.8067733645E+00 9 0.7514500618E+00 0.2485499382E+00 0.7424264550E+00 0.7604734898E+00 10 0.6918258667E+00 0.3081741333E+00 0.6809051633E+00 0.7027463317E+00 11 0.6208983064E+00 0.3791016936E+00 0.6082265377E+00 0.6335700154E+00 12 0.5413829088E+00 0.4586170912E+00 0.5276528001E+00 0.5551129580E+00 13 0.6211041212E+00 0.3788958788E+00 0.6090478897E+00 0.6331601143E+00 14 0.5795320272E+00 0.4204679728E+00 0.5738914013E+00 0.5851726532E+00 15 0.5567277074E+00 0.4432722926E+00 0.5546888113E+00 0.5587666035E+00 16 0.5567269325E+00 0.4432730675E+00 0.5587658286E+00 0.5546880364E+00 17 0.5795311928E+00 0.4204688072E+00 0.5851718783E+00 0.5738904476E+00 18 0.6211042404E+00 0.3788957596E+00 0.6331602335E+00 0.6090482473E+00

0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00

0.4586173E+00 0.3333322E+00 0.1052845E+00 0.3333322E+00 0.1052845E+00 0.4723472E+00 0.3333322E+00 0.1057708E+00 0.3333322E+00 0.1057708E+00 0.4723474E+00 0.3333322E+00 0.1057709E+00 0.3333322E+00 0.1057709E+00 $ENDENCL: $EOF:

Main Index

Chapter 9: Data File Specifications 147 VFDIAG (Diagnostic Data)

Example 2 $TITLE: $TITLE: $TITLE:

PDA Viewfactor VER. 2.3 THIS IS A TEST. TEST DATA SET 001

7-JUN-88 16:14:05

$ENCL:

1 9 1 1 0.1000062227E+01 -0.6222724915E-04 0.0000000000E+00 0.9988046288E+00 0.1027215719E+01 0.9741664529E+00 2 0.1000062227E+01 -0.6222724915E-04 0.0000000000E+00 0.1011101127E+01 0.1010952711E+01 0.9781325459E+00 3 0.1000062227E+01 -0.6222724915E-04 0.0000000000E+00 0.1046023250E+01 0.9557061195E+00 0.9984572530E+00 4 0.1000062227E+01 -0.6222724915E-04 0.0000000000E+00 0.1070914149E+01 0.9665873647E+00 0.9626849890E+00 5 0.1000062227E+01 -0.6222724915E-04 0.0000000000E+00 0.9566107392E+00 0.1043834686E+01 0.9997408986E+00 6 0.1000062227E+01 -0.6222724915E-04 0.0000000000E+00 0.1043245673E+01 0.9702598453E+00 0.9866808653E+00 7 0.1000062346E+01 -0.6234645844E-04 0.0000000000E+00 0.9536437988E+00 0.9969801307E+00 0.1027588487E+01 0.1022036433E+01 8 0.1000062346E+01 -0.6234645844E-04 0.0000000000E+00 0.1031251073E+01 0.9417309761E+00 0.9850846529E+00 0.1042182207E+01 9 0.1000062346E+01 -0.6234645844E-04 0.0000000000E+00 0.9940649271E+00 0.9654292464E+00 0.1021910191E+01 0.1018844128E+01 0.6234646E-04 -0.6226698E-04 0.5960464E-07 0.6226698E-04 0.5960464E-07 0.7091415E-01 -0.1173993E-01 0.4022554E-01 0.3326791E-01 0.2297935E-01 0.5826902E-01 0.1347813E-01 0.3483813E-01 0.3170105E-01 0.1695543E-01 0.3731501E-01 0.7283741E-02 0.2151740E-01 0.1828345E-01 0.1210838E-01 0.4218221E-01 -0.9229196E-02 0.1522104E-01 0.9229196E-02 0.1522104E-01 $ENDENCL: $EOF:

Detailed Descriptions $TITLE The first noncomment line must be: $TITLE: title Multiple $TITLE: lines are allowed, but they must all occur before the $ENCL: line. Comments between multiple $TITLE: lines are allowed. $ENCL and $ENDENCL The next keyword line is: $ENCL:

Main Index

EnclID

#Surf

SymMul

148

Patran Viewfactor Analysis VFDIAG (Diagnostic Data)

with format blanks, A10, 3I10. Parameter

Description

EnclID

The ID number of this enclosure

#Surf

Number of surfaces in this enclosure,

SymMul

Number of symmetric images of each surface. Immediately following, and with no interspersed comments, is the surface data for the #Surf surfaces. A fatal error will occur if the #Surf and the number of surface data do not match. The format and meaning of the surface data is: ( 1X, I10, 3E19.10 ) SURID, SUMSUR, SUMONE, SUMZRO ( 1X, 4E19.10 ) ( SUMNOD( I ), J = 1, NNODE ). Description

Parameter SURID

Surface ID;

SUMSUR

Sum of the viewfactors from this surface to all other surfaces in this enclosure;

SUMONE

1.0 - SUMSUR;

SUMZRO

Sum of the viewfactors from this surface to all other surfaces in this enclosure which were set to zero by virtue of being less than the zero cutoff value;

SUMNOD

Sum of the viewfactors from each nodal subarea on this surface to all other nodal subarea on all other surfaces in this enclosure;

J

Index to the nodal subareas on this surface;

NNODE

Number of nodes on this surface. This pattern is repeated for each surface in the enclosure. After this data has been given for all surfaces in the enclosure, some additional statistical data for the enclosure is given. The format and meaning of the statistical data is: ( 1X, 5E15.7 ) ( MX( J ), AV( J ), SD( J ), AB( J ), ASD( J ) ), J = 0, MXND )

Parameter

Description

MX

Maximum absolute deviation from unity for the sums of viewfactors;

AV

Average deviation from unity for the sums of viewfactors;

SD

Standard deviation of the data used to calculate AV;

AB

Average absolute deviation from unity for the sums of viewfactors;

ASD

Standard deviation of the data used to calculate AB;

J

Index, 0 = entire surface, 1 through MXND = nodal subareas;

MXND

Maximum number of nodes on any surface in this enclosure. The next keyword line is $ENDENCL: with no data. The pattern:

$ENCL: Numeric data $ENDENCL:

EnclID

#Surf

SymMul

is repeated for any additional enclosures. $EOF The last keyword line is $EOF: with no data. Comments may follow the $EOF: line.

Main Index

Chapter 9: Data File Specifications 149 TEMPLATEDAT (Surface Pointer Data)

9.6

TEMPLATEDAT (Surface Pointer Data) The TEMPLATEDAT file is fully specified in the Patran Thermal User’s Guide. Look there for additional information. This section contains the specification for a VFAC Template in this file. Comment lines may be interspersed at any place in the file, except in the midst of a data line. Comment lines are those lines which begin with an asterisk or a semicolon. VFAC Template==::[{(Comment)}]* {(VFAC Template Header)} [[{(Comment)}]* {(VFAC Template Data)}]*MAX(1,nbands) Comment==::((;)|(*))[(anything)]

VFAC Template Header ==::[(del)](VFAC)(del) (TID)[((del)(nbands) [((del)|(;))[(anything)]])|([(del)][(;)[(anything)]](CRLF) and the length of the VFAC Template Header must be less than or equal to 80 characters. VFAC Template Data ==::[(del)](epsilon)[(del)[(tau)[(del)[(empid)[(del)[(tmpid) [(del)[(lambda1)(del)(lambda2)[(del)[(kflag)[(del)[(collapse) [((del)|(;))[(anything) ]]]]]]]]]]]]]] [[(del)](;)[(anything)]] (CRLF)

and the length of the VFAC Template Data must be less than or equal to 80 characters. Parameter

Description

del

(,|blank)[(blank)*]

VFAC

VFAC|'VFAC

V

V|v

F

F|f

A

A|a

C

C|c

TID

positive integer

nbands

non-negative integer, default = 0

anything

any character string representable on the machine

CRLF

carriage return line feed

epsilon

real

tau

real, default = 1.0

empid

integer, default = 0

tmpid

integer, default = 0

lambda1

real, default = 0.0

lambda2

real, default = 0.0

kflag

integer, default = 0

collapse

non-negative integer, default = 0

Main Index

150

Patran Viewfactor Analysis TEMPLATEDAT (Surface Pointer Data)

real

Real number representable on the machine in default real type variable and having 20 or less digits, including +, -, ., E, etc. Default format is G20.10

integer

Integer number representable on the machine in default integer type variable and having 20 or less digits, including +, -, etc. In addition to the above form requirements, only certain combinations of data values are valid. These are determined by the following tests: 1. If nbands equals 0 and either lambda1 not equal to 0.0 or lambda2 not equal to 0.0, then this is an error. 2. If nbands greater than 0 and either lambda1 greater than or equal to lambda2 or lambda 1 less than 0.0, then this is an error. 3. If kflag is not equal to 0 or 1, then this is an error. 4. If empid equals 0 and either epsilon is less than or equal to 0.0 or epsilon is greater than 1.0, then this is an error. 5. If empid is not equal to 0 and epsilon is not equal to 0.0, then this is an error. 6. If tmpid equals 0 and kflag equals 0 and either tau is less than or equal to 0.0 or tau is greater than 1.0, then this is an error. 7. If tmpid equals 0 and kflag equals 1 and tau is less than 0.0, then this is an error. 8. If tmpid is not equal to 0 and tau is not equal to 0.0, then this is an error. Additionally, an error will occur in Viewfactor if tmpid is not equal to 0, and there is not a MEDNOD available with the referenced surface or if kflag equals 1, and an AMBNOD is associated with the reverenced surface.

Main Index

Chapter 9: Data File Specifications 151 VFNODEDAT (Radiosity Node Lists)

9.7

VFNODEDAT (Radiosity Node Lists) The VFNODEDAT file corresponds to a Patran Thermal DEFNOD file. You may refer to the Patran Thermal User’s Guide for the description of that file. The VFNODEDAT file is a FORTRAN Formatted Sequential Access file. The file may contain comment lines which will be ignored by Patran Thermal. Comment lines begin with the character ; or *. Viewfactor creates at most one line of data in VFNODEDAT. If no new radiosity nodes are created by Viewfactor, then no data is output to VFNODEDAT, but a commented message that no new nodes were created is output. Data lines in VFNODEDAT follow the FORMAT (1X, A10, 3I10). The first character field contains the string 'DEFNOD'. The three integer fields contain the beginning node number, the ending node number, and the node number increment in that order for the nodes being defined. For Viewfactor, the node number increment is always 1 and the ending node number is always greater than the beginning node number.

Sample VFNODEDAT File which Defines Nodes ;BEGINNING OF VFNODEDAT FILE EXAMPLE 1 ; ;$TITLE: THIS IS A TEST ; DEFNOD 1001 1500 1 ;1vt ;END OF EXAMPLE 1

Sample VFNODEDAT File which does not Define Nodes ;BEGINNING OF VFNODEDAT FILE EXAMPLE 2 ; ;$TITLE: THIS IS A TEST ; ; NO ADDITIONAL RADIOSITY NODES WERE GENERATED. ; ;END OF EXAMPLE 2

Main Index

152

Patran Viewfactor Analysis VFNODEDAT (Radiosity Node Lists)

Main Index

Chapter 10: Rules for Radiation Resistors Patran Viewfactor Analysis

10

Main Index

Rules for Radiation Resistors



Introduction



General Rules for Radiation Resistors



Rules for Emissivity Resistors



Rules for Radiosity Resistors

154

156 157

155

Chapter 10: Rules for Radiation Resistors 154 Introduction

10.1

Introduction Viewfactor uses rules to determine which Patran Thermal radiation resistors to make. Which resistors are made depends on the data in the VFAC Templates and VFRAWDAT file. This chapter describes these rules. Restrictions on the valid ranges of data in the VFAC Templates is described in the previous chapter in the section on the VFAC Template data specification, VFRAWDAT (Raw Viewfactor Data), 140. These will not be repeated here.

Main Index

Chapter 10: Rules for Radiation Resistors 155 General Rules for Radiation Resistors

10.2

General Rules for Radiation Resistors If the value of nbands is zero, either by default or by specification, then the resistor type will be the Patran Thermal type R. If the value of nbands is greater than zero, then the resistor type will be the Patran Thermal type W.

Main Index

Chapter 10: Rules for Radiation Resistors 156 Rules for Emissivity Resistors

10.3

Rules for Emissivity Resistors For the emissivity resistors: • If empid is zero and epsilon is one, then no emissivity resistor is made and the surface node will serve as the

radiosity node for any radiosity resistors to be made. • If empid is zero and epsilon is greater than zero but less than one, then a radiosity node will be created and

joined to the surface node with a Patran Thermal subtype 5 resistor. • If empid is not zero and epsilon is zero, then a radiosity node will be created and joined to the surface node with

a Patran Thermal subtype 1 resistor. •

Main Index

If Viewfactor detects a zero valued resistor, then that resistor will not be made and the surface node will serve as the radiosity node.

Chapter 10: Rules for Radiation Resistors 157 Rules for Radiosity Resistors

10.4

Rules for Radiosity Resistors For the radiosity resistors: • If tmpid is zero and tau is one and kflag is zero, then a Patran Thermal subtype 5 resistor will be made between

the radiosity nodes. • If tmpid is zero and tau is greater than zero but less than one and kflag is zero, then three Patran Thermal

resistors, all of subtype 5 will be made. One will join the radiosity nodes and the other two will join the two radiosity nodes to the participating media node. • If tmpid is zero and tau is zero and kflag is one, then a Patran Thermal subtype 5 resistor will be made between

the radiosity nodes. • If tmpid is zero and tau is greater than zero and kflag is one, then three Patran Thermal resistors, all of subtype 5

will be made. One will join the radiosity nodes and the other two will join the two radiosity nodes to the participating media node. • If tmpid is not zero and tau is zero and kflag is zero, then three Patran Thermal resistors will be made. One will

be a subtype 9 resistor between the radiosity nodes and the other two will be subtype 10 resistors between the two radiosity nodes and the participating media node. • If tmpid is not zero and tau is zero and kflag is one, then three Patran Thermal resistors will be made. One will be

a subtype 11 resistor between the radiosity nodes and the other two will be subtype 12 resistors between the two radiosity nodes and the participating media node. • If Viewfactor detects a zero valued resistor, then that resistor will not be made.

Main Index

Chapter 10: Rules for Radiation Resistors 158 Rules for Radiosity Resistors

Main Index

Chapter A: Typical Errors and Probable Causes for Viewfactor Errors Patran Viewfactor Analysis

A

Typical Errors and Probable Causes for Viewfactor Errors 

Main Index

Purpose

2

2

Patran Viewfactor Analysis Purpose

A.1

Purpose Viewfactor performs extensive error checking and reporting. It utilizes an error message generator which creates error messages as needed from phrases stored in memory. In general, these error messages provide more information than we could supply with a reasonable number of error codes. Error messages, if any, will be found in the VFMSG file, along with a traceback of the subroutine calling sequence leading to the error condition. This traceback is provided because Viewfactor was designed to terminate normally, even under error conditions. It is the only way for us to know the program status associated with ^an error. Some typical errors and their probable causes are: File Errors

These occur when an expected file is not present or available to the program, or a file to be created is already present. The solution is to make the expected file available or to remove or rename the file already present. The files referenced may also be changed by altering the names given in the VFCTL file.

Format Errors

This means the data in a file is not in an acceptable format. This usually occurs with the TEMPLATEDAT file since this is usually the only data file which the user must build using the system editor.

UID/TID Errors

Reference UIDs are not available in the VFAC TIDs. This error is corrected by adding the appropriate VFAC TID records to the TEMPLATEDAT file.

Surface Incompatibility Errors

The properties assigned by the VFAC DFEG data and VFAC Template data to a pair of surfaces which can see each other in an enclosure are not compatible. For example, the surfaces might have a different number of wavebands assigned to each. For additional information, you can refer to Compatibility Requirements for Model and VFAC Templates, 65 and Introduction, 120 which address surface compatibility.

Note:

Main Index

Please refer back to Submitting a Viewfactor Job for Analysis, 98 and recheck each item called out for review prior to submitting a Viewfactor job.

Chapter B: Quick Reference Guide to Viewfactor Patran Viewfactor Analysis

B

Quick Reference Guide to Viewfactor



Main Index

Purpose

4

4

Patran Viewfactor Analysis Purpose

B.1

Purpose The general procedure for performing a thermal radiation analysis is:

Patran

Build model and assign boundary conditions.

Analysis Menu

Select Viewfactor Solution under Solution Type.

Hit Apply

This executes a script that will check for obstructions, calculate viewfactors and make radiation resistors for Patran Thermal.

Analysis Menu

If a thermal analysis was requested under solution type, following the Viewfactor analysis, Patran Thermal will create a QTRAN source file for the problem and perform the thermal network analysis.

Patran

Display the results. The Patran Plus VFAC boundary condition is: Parameter

Description

UID

User template ID.

MEDNOD

Participating media node if any Flag for top or bottom of shell.

AMBNOD

Ambient or space node.

CNVSID

Convex surface ID.

NONOBSTRUCTING FLAG Flag for nonobstructing surface. TOP/BOTTOM FLAG

Flag for top or bottom of shell.

ENCLOSURE ID

Enclosure ID.

The surfaces on which the VFAC boundary conditions are applied are defined under Application Region. The VFAC Template in the Patran Thermal TEMPLATEDAT file has the form: VFAC, TID, nbands epsilon, tau, empid, tmpid, lambda1, lambda2, kflag, collapse

Main Index

Parameter

Description

TID

Integer, Template ID.

nbands

Integer, Optional, Default = 0, Number of wavebands in this template.

epsilon

Real, surface emissivity.

tau

Real, Optional, Default = 1.0, Participating media transmissivity (kflag = 0) or extinction coefficient (kflag = 1).

empid

Integer, Optional, Default = 0, Emissivity material property ID.

tmpid

Integer, Optional, Default = 0, Transmissivity (kflag = 0) or extinction coefficient (kflag = 1) material property ID.

lambda1

Real, Optional, Default = 0.0, Waveband beginning wavelength, microns.

lambda2

Real, Optional, Default = 0.0, Waveband ending wavelength, microns.

kflag

Integer, Optional, Default = 0, Flag that tau or tmpid refer to transmissivity (kflag = 1) or to extinction coefficient (kflag = 1).

Chapter B: Quick Reference Guide to Viewfactor 5 Purpose

Parameter

Description

collapse

Integer, Optional, Default = 0, ID to control the collapsing of radiosity nodes associated with a surface node.

The line, epsilon, tau, empid, tmpid, lambda1, lambda2, kflag, collapse must be repeated for each of the nbands in the template and once for nbands = 0. The Viewfactor command line is: Note:

Main Index

VSUBMIT VFCTL This command is automatically issued when execution is spawned from within Patran.

6

Patran Viewfactor Analysis Purpose

The VFCTL file contains information to control the execution of Viewfactor. If no filename is given, a default VFCTL file on your system will be used. This default file may have been altered on your system. The default VFCTL file supplied with Viewfactor is: Default VFCTL file * * Sample VFCONTROL file. * * Pathname $PATH: * Message file name $MESSAGE_FILE: vfmsg * Diagnostic data file name $DIAGNOSTIC_FILE: vfdiag * Title $TITLE: 'THIS IS A TEST' * Input data file name $IN_DATA: vfindat * Template file name $TEMPLATE_FILE: templatedat * Raw viewfactor data file name $RAW_DATA: vfrawdat * Radiation resistor file name $OUT_DATA: vfresdat * Radiosity node file name $RAD_NODE_FILE: vfnodedat * $STATUS_FILE: vfrestartstat * $RESTART_FILE: vfrestartdat * 0 = full run, 1 = viewfactors only, 2 = resistors only $RUN_CONTROL: 0 * $RESTART_FLAG: 0 * $CONVERGE: -1.0 $ZERO: 0.0 $APPROX_CURVE: 0.1 * Contour Double_area Weighting $GAUSS_ORDER: 8 8 8 * minimum maximum $AXISYM_SURFACE: 5 13 $EOF:

Note:

Main Index

The above parameters have been defaulted in the Analysis form, Viewfactor Solution Parameters.

Chapter C: Memory Requirements for Viewfactor Execution Patran Viewfactor Analysis

C

Memory Requirements for Viewfactor Execution 

Main Index

Purpose

8

8

Patran Viewfactor Analysis Purpose

C.1

Purpose Viewfactor has fixed memory requirements for about 820 K bytes on the VAX 8600. In addition to this memory, Viewfactor calculates the amount of memory needed based on the number of nodes in the model, number of surfaces in the current enclosure, dimensionality of the model, number of symmetric images and number and size of the VFAC templates. The memory requirement for calculating viewfactors also depends on the number of potentially obstructing surfaces in the enclosure and on a few other parameters in a complicated manner. Therefore, we will only give an upper bound on the memory requirement. The memory required, in bytes, to calculate viewfactors will be less than approximately the value of the expression: [( 19•DIM2 + 59•DIM + 85 )•SYMMUL•NSURF + 4•SYMMUL•NNODE]•4 + 930000 Description

Parameter

+

19•NSURF

DIM SYMMUL

Dimensionality of the model, 2 for 2-D XY and RZ axisymmetric models, 3 for 3-D XYZ models.

NSURF NNODE

Number of surfaces in the enclosure for which viewfactors are currently being calculated.

Number of images of the model after all symmetry operators have been applied to the model, includes the original image of the model. Number of nodes in the model. The memory requirement for converting viewfactors into Patran Thermal radiation resistors depends on the number of resistors to be created in an enclosure. The virtual memory system will request enough memory to hold all of the resistors created in an enclosure before any of the resistors have been merged. Since the number of resistors created depends on the viewfactors calculated, we do not know in advance how much memory will be required. Viewfactor was designed to access as much memory as needed through the computer’s virtual memory system. Viewfactor will handle as large of problem as there is virtual memory available for it to use. Also note that the memory requirements are linear with respect to the number of surfaces in each enclosure and also linear with respect to the number of nodes in the model. Thus, in general, computer memory will not be the limiting resource for viewfactor analysis with Viewfactor. However, a large amount of memory may be needed for sorting and merging the resistors made after the viewfactor calculations.

NVFAC TOTBND

Number of VFAC templates found in the TEMPLATEDAT file.

NUID

Number of distinct UIDs (User IDs) in the current enclosure.

total number of bands, or VFAC template data records (not header record found in the TEMPLATEDAT file.

Viewfactor was designed to access as much memory as needed through the computer‘s virtual memory; it will handle as large of problem as there is virtual memory available for it to use. Also note that the memory requirements are linear with respect to the number of surfaces in each enclosure and with respect to the number of nodes in the model. Thus, in general, computer memory will not be the limiting resource for viewfactor analysis with Viewfactor.

Main Index

Chapter D: Machine-Specific File Names for Viewfactor Patran Viewfactor Analysis

D

Machine-Specific File Names for Viewfactor 

Main Index

Purpose

10

10

Patran Viewfactor Analysis Purpose

D.1

Purpose This document uses generic filenames. The generic names may be translated to machine specific names by use of the following table. Generic Name

DEC VAX Name

UNIX Name

TEMPLATEDAT

template.dat

template.dat

VFCTL

vf.ctl

vf.ctl

VFDIAG

vf.diag

vf.diag

VFINDAT

vfin.dat

vfin.dat

VFMSG

vf.msg

vf.msg

VFNODEDAT

vfnode.dat

vfnode.dat

VFRAWDAT

vfraw.dat

vfraw.dat

VFRESDAT

vfres.dat

vfres.dat

VFRESTARTDAT

vfrestart.dat

vfrestart.dat

VFRESTARTSTAT

vfrestart.stat

vfrestart.stat

Main Index

Chapter E: Example Thermal Radiation Problems Patran Viewfactor Analysis

E

Main Index

Example Thermal Radiation Problems 

Purpose



Problem 1 - Steady-State Radiative Boundary Conditions



Problem 2 - Parallel Semi-Infinite Plates



Problem 3 - Heated Reaction Chamber

12

32 35

13

12

Patran Viewfactor Analysis Purpose

E.1

Purpose These examples are simple problems designed to demonstrate the thermal radiation analysis process. The first example, steady-state radiative boundary conditions, is a two-enclosure radiative model. The second, parallel semi-infinite plates, is an idealized situation for which the correct thermal solution is known. The third example, more representative of a real world problem, models the interior of a reaction chamber. In the first example, a step-by-step model definition is given, including figures showing the Patran forms. For the subsequent examples, to avoid repeating the forms, only the significant model data is given.

Main Index

Chapter E: Example Thermal Radiation Problems 13 Problem 1 - Steady-State Radiative Boundary Conditions

E.2

Problem 1 - Steady-State Radiative Boundary Conditions Objectives In this lesson you will perform the following tasks: • Construct a 2D model that incorporates two enclosures. • Define separate radiative boundary conditions for gray body and wavelength-dependent radiation within the

enclosures. • Perform the Steady-State thermal analysis and postprocess the analysis results with Patran’s Result tools.

Model Description In this lesson you will construct a model with two separate radiation enclosures: one for gray body radiation and the other for wavelength-dependent radiation. No material (e.g., air) will be defined in the enclosure; therefore, only Radiation heat transfer can transfer heat energy across the enclosures. In the enclosure where it is assumed that the surfaces are gray, the emissivity will be constant regardless of the surface temperatures. The other enclosure will incorporate wavelength-dependent radiation which is a significant extension of the gray body theory. Normal radiosity is divided into discrete frequency bands with emissivity and transmissivity assumed to be gray within these frequency bands. o

1500 C (fixed)

0.3

0.5

0.4

0.5

0.3

Iron E-1

1.6

E-2

0.6

0.5

2.0 o

Node 1000 o

0 C (fixed)

T=200 C (fixed)

Enclosure Emissivity Information: Enclosure 1

Gray

Enclosure 2

For:

ε = 0.9 0.0 ≤ λ ≤ 5.0 5.0 < λ ≤ ∞

Main Index

ε(λ)=0.9 τ=0.4 ε(λ)=0.2 τ=0.4

14

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

Exercise Procedure 1. Start Patran and create a New Database named, exercise_13.db. 2. Set the Tolerance to Default and the Analysis Code to THERMAL Vol. 2-Viewfactor Analysis. 3. To create the geometry of the enclosure model clicking on the Geometry toggle in the main form. Set the Action, Object, and Method respectively to Construct, Patch, and XYZ. Change the Vector and Origin Coordinate Lists to <0.3, 0.5, 0.0> and [0, 0, 0] respectively. Click on the Apply button to create the patch that represents the bottom left region of the model. Before creating the next piece of the model set the Display Lines to zero. To create the next region of the model change the Vector Coordinate List to <0.5, 0.5, 0> click in the Origin Coordinate List, and select Point 4 in the viewport. Using the above construction technique complete the remaining portion of the model’s geometry. Your completed patch geometry should look similar to that shown below.

4. Next, you will mesh the 13 patches using the ISO mesher.

Main Index

Chapter E: Example Thermal Radiation Problems 15 Problem 1 - Steady-State Radiative Boundary Conditions

Click on the Finite Elements Toggle in the main form. Change the Object to Mesh. Specify a Global Edge Length of 0.16666 for the QUAD 4 elements. The completed Finite Elements form and meshed model are shown below for your reference. Finite Elements Action:

Create

Object:

Mesh

Type:

Surface

Output Ids Node Id List 1 Element Id List 1 Global Edge Length 0.16666 Element Topology Quad4 Quad5 Quad8 Mesher IsoMesh

Paver

IsoMesh Parameters... Node Coordinate Frames... Surface List Surface 1:13

-Apply-

Main Index

16

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

To simplify the visual image of your model, turn off all entity labels. Your model should now look like the one shown below.

5. To equivalence your model, change the Finite Element form’s Action, Object, and Type respectively to Equivalence, All, and Tolerance Cube. Click on the Apply button to equivalence the finite elements. 6. The model’s Enclosure 2 (hole on the right-hand side) will contain a medium that will participate in the radiation heat transfer occurring throughout that enclosure.

Main Index

Chapter E: Example Thermal Radiation Problems 17 Problem 1 - Steady-State Radiative Boundary Conditions

To thermally represent the participating medium, create a Node within the enclosure. Use 1000 as its ID and create the node so that it is not associated with the model’s geometry. Use the Select a Screen Position option in the select menu to select a point inside Enclosure 2. Your completed Finite Elements form should look similar to the one shown below. Finite Elements Action:

Create

Object:

Node

Method: Edit Node Id List 1000 Analysis Coordinate Frame Coord 0 Refer. Coordinate Frame Coord 0 Associate with Geometry Auto Execute Node Location List [1.364971 0.836337 0.00

-Apply-

Main Index

18

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

To better visualize the Node locations, set their radius to 6 pixels. Your model should now look like the one shown below.

7. You will now create the required boundary conditions for your model. Temperature Boundary Conditions: Click on the Loads/BCs toggle in the main form. In the Loads/Boundary Conditions form enter, Temp_Participating_medium, as the New Set Name and then click on the Input Data… button. In the Input Data form, assign a fixed temperature of 200 (remember TID=-1). Click on the OK button to close the Input Data form. In the Loads/Boundary Conditions form click on the Select Application Region… button. In the Select Application Region form, change the Geometry Filter to FEM and then click in the Select Nodes databox. Select Node 1000 from the viewport. This node will represent the Participating Medium temperature. The completed forms are shown below for your reference.

Main Index

Chapter E: Example Thermal Radiation Problems 19 Problem 1 - Steady-State Radiative Boundary Conditions

Load/Boundary Conditions Create

Action: Object:

Temp (PThermal)

Type:

Nodal

Analysis Type: Thermal Current Load Case: Default... Type:

Input Data

Select Application Region

Load/BC Set Scale Factor 1

Temperature 200 Template ID -1

Geometry Filter Geometry FEM

Application Region Select Nodes

Static

Existing Sets

Add

Remove

Application Region Spatial Fields

Node 1000

New Set Name Temp_Participating_medi

Input Data... Select Application Region... OK

Reset

OK

-Apply-

Next, assign fixed temperatures of 1500°C and 0°C respectively to the top and bottom geometry edges of the model. Use T_top and T_bottom for their respective New Set Names. The completed forms are shown below for your reference.

Main Index

20

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

Input Data

Load/Boundary Conditions Action:

Create

Object:

Temp (PThermal)

Type:

Nodal

Analysis Type: Thermal Current Load Case: Default... Type:

Select Application Region

Load/BC Set Scale Factor 1

Temperature 1500 Template ID -1

Geometry Filter Geometry FEM

Application Region Select Geometry Entities

Static Add

Existing Sets Temp_Participating_medium

Remove

Application Region Surface 9:13.2

Spatial Fields New Set Name T_top

Input Data... Select Application Region... -Apply-

Main Index

OK

Reset

OK

Chapter E: Example Thermal Radiation Problems 21 Problem 1 - Steady-State Radiative Boundary Conditions

Input Data

Load/Boundary Conditions Action:

Create

Object:

Temp (PThermal)

Type:

Nodal

Analysis Type: Thermal

Temperature 0 Template ID -1

Current Load Case: Default... Type:

Select Application Region

Load/BC Set Scale Factor 1

Geometry Filter Geometry FEM

Application Region Select Geometry Entities

Static Add

Existing Sets Temp_Participating_medium T_top

Remove

Application Region Surface 1:5.4

Spatial Fields New Set Name T_bottom

Input Data... Select Application Region...

OK

Reset

OK

-Apply-

Viewfactor Boundary Condition: To create the viewfactor boundary conditions for the two enclosures you will first supply geometric information in the Patran Loads/BCs form and then enter data concerning the Emissivity and Transmissivity values in the template.dat file. In the Loads/Boundary Conditions form, change the Action, Object, and Type option menus respectively to Create, Viewfactor (P/THERMAL), and Element Uniform. Change the Target Element Type to 2D. Use the diagram and Table E-1 and Table E-2 to determine the required geometric information for the eight viewfactors which define the two enclosures.

203

103 102

104 101

Main Index

202

204 201

22

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

Table E-1

Table E-2

Enclosure 1 New Set Name

Enc1_101

Enc1_102

Enc1_103

Enc1_104

Vfac Template ID

100

100

100

100

Partic. Media Node ID

-----

-----

-----

-----

Ambient Node ID

-----

-----

-----

-----

Convex Surface ID

101

102

103

104

Clos. Flag

-----

-----

-----

-----

Top/Bot Flag

-----

-----

-----

-----

Enclosure ID

1

1

1

1

New Set Name

Enc1_101

Enc1_102

Enc1_103

Enc1_104

Vfac Template ID

200

200

200

200

Partic. Media Node ID

1000

1000

1000

1000

Ambient Node ID

-----

-----

-----

-----

Convex Surface ID

201

202

203

204

Clos. Flag

-----

-----

-----

-----

Top/Bot Flag

-----

-----

-----

-----

Enclosure ID

2

2

2

2

Enclosure 2

You will now complete the Viewfactor definitions by entering the Emissivity and Transmissivity information into the template.dat file. Create a separate x-window shell and make a subdirectory in the directory you are running Patran. Use your current Patran database name for this subdirectory name. Change to that subdirectory, open and edit a file named template.dat. Next, enter the required VFAC commands to define the Emissivity and Transmissivity for Enclosures 1 and 2. The syntax of the command is: VFAC e

Main Index

TID t

NBANDS εid

τid

λ1

λ2

K-flag

Collapse

Chapter E: Example Thermal Radiation Problems 23 Problem 1 - Steady-State Radiative Boundary Conditions

Each term of the command is defined in Chapter 3 of the MSC Patran Thermal User’s Guide. Shown below is a table that lists the required information for the two VFAC commands and the template.dat file created with this information for your reference. TID

NBANDS

e

t

εid

τid

λ1

λ2

K flag

Collapse

100

0

0.9

1

0

0

0

0

0

1

200

2

0.9

0.4

0

0

0

5

0

1

0.2

0.4

0

0

5

1E6

0

1

Your model with its applied boundary conditions should now look like the one shown below.

8. Before you set up and run the thermal analysis you must first define the Element Properties for the models Iron material. To do this, click on the Element Props toggle in the main form. When the form appears set its Action, Dimension, and Type option menus respectively to Create, 2D, and Thermal 2D. Enter Iron, for the New Set Name and then click on the Input Properties… button. Enter 18 in the Material Name databox and then click on the OK button to close the

Main Index

24

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

form. Next, click in the Select Members box and select all the models Patches in the viewport. Finally click on the Apply button in the Element Properties form. The completed forms are shown below for your reference. Element Properties Action:

Create

Dimension: 2D Type:

Thermal 2D

Input Properties

2D Property Name

Value

Value Type

Existing Property Sets Material Name

Property Set Name Iron

18

Mat Prop Name

[Material orient. -X]

Real Scalar

[Material orient. -Y]

Real Scalar

[Material orient. -Z]

Real Scalar

Input Properties ... Application Region

Material Property Sets

Select Members

Add

Remove

Application Region OK

-Apply-

9. You will now set up the thermal analysis run. Click on the Analysis toggle in the main form. When the Analysis form appears check that the Action, Object, and Method option menus are respectively set to Analysis, Full Model, and Full Run. Click on the Solution Type… button. When the P⁄THERMAL Solution Type form appears specify a Steady State Thermal Analysis. To cause Viewfactor to calculate the viewfactor for both enclosures click on the Perform Viewfactor Analysis

Main Index

Chapter E: Example Thermal Radiation Problems 25 Problem 1 - Steady-State Radiative Boundary Conditions

button. In the Select Viewfactor Solution frame, select the 0, Viewfactors --> Resistors option. Click on the OK button to close the form. P/Thermal Solution Type Perform Thermal Analysis Perform Hydraulic Analysis Coupled Thermal/Hydraulic Analysis Select Thermal Solution 0, Data Check Run Only 1, Transient Run 2, SS --> Transient Run 3, Steady State Run 4, Transient --> SS Run 5, SS --> Transient --> SS Run

Perform Viewfactor Analysis Select Viewfactor Solution 0, Viewfactors --> Resistors 1, Viewfactors Only 2, Resistors Only

OK

Main Index

Defaults

Cancel

26

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

In the Analysis form, click on the Solution Parameters… button. In the P⁄Thermal Solution Parameters form, set the Calculation Temperature Scale to Celsius and select 0, Standard Solution for the Solver Option. Click on the OK button to close the P/Thermal Solution Parameters form. P/Thermal Solution Parameters - Thermal - Viewfactor Calculation Temperature Scale Celsius

Kelvin

Fahrenheit

Rankine

Thermal Solution Parameters Solver Option:

0, Standard Solution

Iterations between complete update =

4

Run Control Parameters... Convergence Parameters... Iteration Parameters... Relaxation Parameters...

Viewfactor Solution Parameters Run Control Parameters... Viewfactor File Names...

OK

Defaults

Cancel

In the Analysis form, click on the Output Requests… button. In the P⁄Thermal Output Request form, set the Units Scale for Output Temperatures to Celsius. Click on the Print Intervals Controls… button. When the P⁄Thermal Print Intervals

Main Index

Chapter E: Example Thermal Radiation Problems 27 Problem 1 - Steady-State Radiative Boundary Conditions

Control form appears, set the Initial Print Interval to 5. This will cause P/Thermal to print out solution information every 5 intervals. Click on the OK button in both forms. P/Thermal Output Request Echo Input in Output File Units Scale for Output Temperatures Celsius

Kelvin

Fahrenheit

Units Definition for Time Label =

Rankine

Seconds

Print Interval Controls... Nodal Results File Format... Print Block Definition... Plot Block Definition... Diagnostic Output... SINDA Input Deck Format... OK

Defaults

Cancel

Submit the analysis run by clicking on Apply in the Analysis form. 10. When the Analysis Run is finished, change the Action option menu on the top of the Analysis form to Read Results. Click on the Select Results File… button and select the results file, nro.nrf.1, located in the exercise_13 subdirectory. Remember to select the P⁄Thermal results template before you click on the Apply in the Analysis form to cause Patran to read the results into the database.

Main Index

28

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

11. Click on the Results toggle in the Top Menu Bar. When the Results form appears display the temperature distribution across the model. Your model should now look like the one shown below.

As expected the temperature distribution is not horizontally symmetrical due to the different radiation boundary conditions in each enclosure. 12. Temperature contours can be created and the temperature at each node can be determined by creating Contour and Cursor tools.

Main Index

Chapter E: Example Thermal Radiation Problems 29 Problem 1 - Steady-State Radiative Boundary Conditions

To do this, set the Action and Object, to Create and Contour, respectively. Click on the Results Selection… button and select 1.1 Temperature, (nodal) from the Contour Results List Box. Click on the OK button to close the from. Click on the Apply to create the Temperature Contours. Your model should now look similar the one shown below.

To create Cursor tool change the Object to Cursor and then click on the Results Selection… button. Again select 1.1Temperature, (nodal) in the Cursor Results list and click on OK to close the form. Click on the Apply button to create the Cursor Tool. When the Cursor Tool form appears click on the Apply button. Next, click somewhere on the model.

Main Index

30

Patran Viewfactor Analysis Problem 1 - Steady-State Radiative Boundary Conditions

You should see the temperature of the Node nearest to the mouse cursor printed on the model and in the Cursor Results form. Your model should now look similar to the one shown below.

Main Index

Chapter E: Example Thermal Radiation Problems 31 Problem 1 - Steady-State Radiative Boundary Conditions

An example Cursor Results form and its corresponding temperature locations are shown below for your reference.

Finish this exercise by quitting Patran.

Main Index

32

Patran Viewfactor Analysis Problem 2 - Parallel Semi-Infinite Plates

E.3

Problem 2 - Parallel Semi-Infinite Plates This example is of two parallel plates of finite width and modeled as infinitely long. The boundary conditions are independent of the location along the infinite length of the plates. This allows us to model the problem in 2-D XY space. In this example, the plates have equal width and are directly opposed to each other. You may wish to consider other arrangements such as directly opposed plates of unequal width or equal width plates offset from direct opposition. You may also vary the separation between the plates and the width of the plates. The problem is shown schematically in Figure E-1, along with a typical finite element model of the problem. Patch 2

q

Patch 3

Ambient Surface

Ambient Node

q

Patch 1 Figure E-1

Example Problem, Parallel Semi-infinite Plates with Opening to Space

By appropriate specification of boundary conditions, we will create symmetry in the heat flux from the plates to the ambient environment from the openings to the left and right. Only the facing surfaces of the plates will be able to radiate thermal energy. All other surfaces will be perfectly insulated. Also, the plates will be thin and good insulators themselves. Thus, other modes of heat transfer, such as conduction and convection, may be neglected. A heat flux varying along the width of each plate will be imposed on the radiating face of each plate. The variation in heat flux from left to right on the bottom plate’s face will be the same as the variation from right to left on the top plate’s face. In this example, we will use a linear ramp variation. The plates will radiate to each other and to the ambient environment at the left and right. By symmetry arguments, the total energy radiated to the ambient environment on the left and on the right should be equal. By conservation of energy their sum must equal the heat flux applied to the plates’ surfaces. As in the previous example, we will model one of the openings to the ambient environment as an ambient surface. Note that this ambient surface is not connected to the plates. The other opening in the enclosure to the ambient environment will be modeled with an ambient node. Note that each surface, the top plate, the bottom plate, and the ambient surface, is convex. The material with MPID 693 used for this problem is a mica brick. The model files for this example were delivered with Patran and should be available on your computer system by typing ‘get_view’ and selecting the directory ‘pplate’. For assistance in locating these files,

Main Index

Chapter E: Example Thermal Radiation Problems 33 Problem 2 - Parallel Semi-Infinite Plates

please contact your system administrator. You may wish to experiment with other geometric dimensions, heat fluxes, and ambient temperatures. A Patran Thermal TEMPLATEDAT file is needed. You might use the one from the previous example, editing the comments. A TEMPLATEDAT file is shown below.

TEMPLATEDAT File MID * SURFACE VFAC 0.8 * SURFACE VFAC 1.0

693 69301 69301 69301 69304 69305 PROPERTIES OF THE PARALLEL PLATES, PATCHES 1 AND 2 11 0

0

PROPERTIES OF THE AMBIENT SURFACE, PATCH 3 12 0

Likewise a Patran Thermal MATDAT file is needed. Use the MATDAT file from the previous example. Copy that file to the present directory for this problem. Otherwise you will need to create a new MATDAT file. The resulting Patran model is translated into Patran Thermal data files and Viewfactor input data files by clicking Apply on the Analysis menu. Be sure to choose the X-Y pick for the dimensionality of the problem under Translation Parameters. Remember to check the VFMSG file for error messages when the Viewfactor analysis is completed. The VFDIAG file from this analysis is shown below. Since this is not a closed enclosure, the viewfactors to each surface do not sum to one. The ambitious user may wish to calculate the views from various surfaces to the opening between the plates and compare these calculated values to the one minus sum values in the VFDIAG file.

VFDIAG File $TITLE: $TITLE: $TITLE:

PDA Viewfactor VER. 2.5 4-APR-91 17:18:55 PARALLEL SEMI-INFINITE PLATES OPEN TO THE LEFT AND RIGHT. PARALLEL PLATES, SIMPLE, FINE, LINEAR HEAT AT NODES.

$TITLE: $ENCL:

22-MAR-91

16:51:24 2.5 1 18 1 1 0.8375167251E+00 0.1624832749E+00 0.8469785452E+00 0.8280549049E+00 2 0.7995741367E+00 0.2004258633E+00 0.8067734241E+00 0.7923747897E+00 3 0.7514496446E+00 0.2485503554E+00 0.7604730725E+00 0.7424262762E+00 4 0.6918255091E+00 0.3081744909E+00 0.7027461529E+00 0.6809048057E+00 5 0.6208978891E+00 0.3791021109E+00 0.6335697174E+00 0.6082262397E+00 6 0.5413827300E+00 0.4586172700E+00 0.5551127791E+00 0.5276526213E+00 7 0.8375174403E+00 0.1624825597E+00 0.8280556202E+00 0.8469793200E+00 8 0.7995739579E+00 0.2004260421E+00 0.7923744321E+00 0.8067733645E+00 9 0.7514500618E+00 0.2485499382E+00 0.7424265146E+00 0.7604734898E+00 10 0.6918258667E+00 0.3081741333E+00 0.6809051633E+00 0.7027463317E+00 11 0.6208983064E+00 0.3791016936E+00 0.6082265377E+00 0.6335700154E+00 12 0.5413829088E+00 0.4586170912E+00 0.5276528001E+00 0.5551129580E+00 13 0.6211039424E+00 0.3788960576E+00 0.6090477705E+00 0.6331601143E+00 14 0.5795322657E+00 0.4204677343E+00 0.5738915801E+00 0.5851728916E+00

Main Index

0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00 0.0000000000E+00

34

Patran Viewfactor Analysis Problem 2 - Parallel Semi-Infinite Plates

15 0.5567277074E+00 0.4432722926E+00 0.0000000000E+00 0.5546888113E+00 0.5587666035E+00 16 0.5567269325E+00 0.4432730675E+00 0.0000000000E+00 0.5587658286E+00 0.5546880364E+00 17 0.5795311928E+00 0.4204688072E+00 0.0000000000E+00 0.5851718783E+00 0.5738904476E+00 18 0.6211042404E+00 0.3788957596E+00 0.0000000000E+00 0.6331602335E+00 0.6090482473E+00 0.4586173E+00 0.3333322E+00 0.1052847E+00 0.3333322E+00 0.1052847E+00 0.4723472E+00 0.3333322E+00 0.1057708E+00 0.3333322E+00 0.1057708E+00 0.4723474E+00 0.3333322E+00 0.1057709E+00 0.3333322E+00 0.1057709E+00 $ENDENCL: $EOF: $QTRAN Finally, you will want to look at the Patran Thermal output data in the QOUTDAT file. Use the system editor to find the first occurrence of the string '1TIME'. Note the system heat balance. This is approximately the imposed heat flux to the plates’ surfaces. Page down through the data to node 1000, the ambient node. Note that the heat flux at this node is approximately half of the imposed heat flux as expected. Sum up the heat fluxes to the ambient surface and note that this sum is approximately half of the imposed total heat flux. The nodes numbered above 1000 are the radiosity nodes created by Viewfactor. This is a simple problem and you may wish to try numerous variations on it, such as refining the mesh, changing the convergence criteria, changing the double area parameter of the $GAUSS_ORDER, changing the imposed heat flux, changing the ambient environment temperature, specifying the plate surface temperatures, and entering more complicated descriptions of the radiative surface properties.

Main Index

Chapter E: Example Thermal Radiation Problems 35 Problem 3 - Heated Reaction Chamber

E.4

Problem 3 - Heated Reaction Chamber This example represents a simple version of a reaction chamber. The main chamber is a short vertical cylinder with hemispherical caps on each end. The main chamber also has a smaller chamber teed off from its side. The smaller chamber is a short cylinder, with a smaller diameter than the large chamber, and capped with a hemispherical cap. The chamber walls are constructed of a high-strength steel which corresponds to material ID 365 in the material property database provided with the Patran Thermal module. The model was created using the MKS system of units for the physical and material properties. The bottom cap of the main chamber is heated with a flux of 5000 watts per square meter applied to the exterior surface. The rest of the exterior surface is convectively coupled to the ambient environment at 300 K. The convection coefficients are constant with the value for the main chamber being 1 watt per square meter per degree K and the value for the smaller chamber being 20. The interior of the chamber has only thermal radiation boundary conditions. A participating media node is included in the model in case we wish to use it in the future. However, it is not used in the analysis presented here. The Patran model uses a very coarse mesh, since we do not want to use up a large amount of computer time on an example problem. The model files for this example were delivered with Patran Thermal and should be available on your computer system by typing ‘get_view’ and selecting the directory ‘chamber.’ For assistance in locating these files, please contact your system administrator. A Patran Thermal TEMPLATEDAT file is needed. A TEMPLATEDAT file is shown below.

TEMPLATEDAT File MID * VFAC 0.8

365 365 1.0

0 0

36501 0

0

0

0

36501

36501

36504

36505

36506

1

The material template ID 365 is for the high-strength steel used for the vessel walls and the VFAC template 365 is for the interior surfaces of the vessel. The template gives the emissivity as a constant value of 0.8. No other radiation property data is given in this case since this is a simple model.

Main Index

36

Patran Viewfactor Analysis Problem 3 - Heated Reaction Chamber

Likewise, a Patran Thermal MATDAT file is needed. We have created a file by using our system editor and extracting the data from the material property data file for MKS units supplied with the Patran Thermal module in the THERMAL$DIR:[LIBRARY] directory as MPID.MKS. Our MATDAT file is shown below.

MATDAT File MPID 36501 CONSTANT KELVIN 1.0 STEEL, ULTRA HIGH STRENGTH TYPE 300-M References: 1 Data Quality: EXCELLENT MDATA 5.77806E+01 / MPID 36504 CONSTANT KELVIN 1.0 STEEL, ULTRA HIGH STRENGTH TYPE 300-M References: 1 Data Quality: EXCELLENT MDATA 7.84000E+03 / MPID 36505 CONSTANT KELVIN 1.0 STEEL, ULTRA HIGH STRENGTH TYPE 300-M References: 1 Data Quality: EXCELLENT MDATA 4.47324E+02 / MPID 36506 PHASE KELVIN 1.0 STEEL, ULTRA HIGH STRENGTH TYPE 300-M References: 1 Data Quality: EXCELLENT MDATA 1.77315E+03 1.51190E+05 /

--> Thermal Conductivity (W/(m*Sec*K))

--> Density (Kg/m**3)

--> Specific Heat (J/(Kg*K))

--> Latent Heat (J/Kg)

The resulting Patran model is now translated into thermal input data files and Viewfactor input data files by clicking on Apply from the Patran Analysis menu. This viewfactor analysis takes about 900 CPU seconds on a VAX 8600, so be forewarned that this job will require a significant amount of computer time and you may not wish to spend your computer resources running this example problem. Output for this analysis has been included with the Viewfactor delivery and is available on your system. Remember to check the VFMSG file for error messages when the Viewfactor analysis is done. The last 40 lines of the VFDIAG file from this analysis is shown below. Since the interior of the chamber is a closed radiation enclosure, we expect the sums of viewfactors from any surface to all other surfaces to be one, or at least very close to one (after taking into account computer and numerical approximations and discretization errors during obstructed view checking). From the diagnostic data file, VFDIAG, we observe that the maximum deviation from one for these sums is about 0.03 and the average deviation is about 0.01. Both of these values are reasonable for a 108 surface enclosure.

Main Index

Chapter E: Example Thermal Radiation Problems 37 Problem 3 - Heated Reaction Chamber

VFDIAG File $TITLE: $TITLE: $TITLE: $TITLE:

PDA VIEWFACTOR VER. 2.5 4-APR-91 17:58:55 HEATED REACTION CHAMBER WITH SIDE CHAMBER. EXAMPLE PROBLEM, REACTION CHAMBER, 3D, ABOUT 110 ELEMENTS. 4-APR-91

17:44:28

2.5

$ENCL:

1 108 1 1 0.9969453812E+00 0.3054618835E-02 0.0000000000E+00 0.1003143072E+01 0.1001876235E+01 0.9858158231E+00 2 0.9985128045E+00 0.1487195492E-02 0.0000000000E+00 0.1004488468E+01 0.9986609221E+00 0.9923893809E+00 3 0.9980445504E+00 0.1955449581E-02 0.0000000000E+00 0.9956341982E+00 0.1000977635E+01 0.1001693606E+01 0.9938086867E+00 4 0.9978431463E+00 0.2156853676E-02 0.0000000000E+00 0.9944566488E+00 0.1001591563E+01 0.1000043273E+01 0.9952377677E+00 5 0.9988200068E+00 0.1179993153E-02 0.0000000000E+00 0.1005183816E+01 0.9964703321E+00 0.9948055148E+00 6 0.9989739060E+00 0.1026093960E-02 0.0000000000E+00 0.1002386332E+01 0.9964384437E+00 0.9980962276E+00 7 0.9983404279E+00 0.1659572124E-02 0.0000000000E+00 0.9980260730E+00 0.1000427961E+01 0.9992659688E+00

**some lines missing** 99 0.9866157770E+00 0.1338422298E-01 0.0000000000E+00 0.9843998551E+00 0.9941601753E+00 0.9812870622E+00 100 0.9874985814E+00 0.1250141859E-01 0.0000000000E+00 0.9909937382E+00 0.9941576719E+00 0.9773445725E+00 101 0.9940232038E+00 0.5976796150E-02 0.0000000000E+00 0.9886876345E+00 0.9990730286E+00 0.9977318645E+00 0.9905698895E+00 102 0.9940228462E+00 0.5977153778E-02 0.0000000000E+00 0.9993922710E+00 0.9883680344E+00 0.9906400442E+00 0.9976602793E+00 103 0.9908416271E+00 0.9158372879E-02 0.0000000000E+00 0.9801433682E+00 0.9940931797E+00 0.9982880354E+00 104 0.9921196103E+00 0.7880389690E-02 0.0000000000E+00 0.1005351782E+01 0.9940947294E+00 0.9769117832E+00 105 0.9946594238E+00 0.5340576172E-02 0.0000000000E+00 0.9900299907E+00 0.9957625866E+00 0.9983595610E+00 0.9941036105E+00 106 0.9946811795E+00 0.5318820477E-02 0.0000000000E+00 0.9960818887E+00 0.9897401333E+00 0.9942038655E+00 0.9983152747E+00 107 0.9866156578E+00 0.1338434219E-01 0.0000000000E+00 0.9844000340E+00 0.9941600561E+00 0.9812868237E+00[5;9H[21;H 108 0.9874987006E+00 0.1250129938E-01 0.0000000000E+00 0.9909937978E+00 0.9941574931E+00 0.9773442149E+00 0.3032351E-01 0.1014543E-01 0.9183009E-02 0.1014543E-01 0.9183009E-02 0.6392515E-01 0.1271649E-01 0.1550230E-01 0.1388588E-01 0.1445413E-01 0.6393278E-01 0.1234503E-01 0.1532391E-01 0.1322156E-01 0.1456719E-01 0.2590126E-01 0.8442800E-02 0.8157597E-02 0.9063553E-02 0.7455045E-02 0.2589691E-01 0.6228297E-02 0.7944188E-02 0.6822405E-02 0.7435330E-02 $ENDENCL: $EOF:

Under Solution, type select Viewfactor Analysis and make sure the option is for steady state, option 3. A few minor modifications need to be made to the Patran Analysis form. Under Viewfactor Solution Parameters, change the title to: Viewfactor EXAMPLE PROBLEM CHAMBER. Under Solution Parameters, set EPSISS to1.0000000000d-03.

Main Index

38

Patran Viewfactor Analysis Problem 3 - Heated Reaction Chamber

Under Output Requests, Diagnostic Output, set all of the toggles off, especially the radiation resistors to avoid receiving printout for tens of thousands of radiation resistors. Under Solution Parameters, run control, set the initial temperature to 300 K. The thermal analysis is spawned when Apply is selected on the Analysis menu. The analysis will take much longer than for a similar model without any radiative interchange. When the radiative interchange is modeled in this example, nearly every nodal subarea on the interior surface of the vessel is connected to nearly every other nodal subarea on the interior surface by means of radiation resistors. Thus the resistor network which QTRAN must solve has many times more resistors than a similar model without radiation coupling. Also, the heat transfer across the radiative resistors is highly nonlinear. This further increases the time required for QTRAN to solve the network equations. The QTRAN thermal network analysis will require approximately 6000 CPU seconds on a VAX 8600. You may not wish to spend your computer resources running this example problem. Output for this analysis has been included with the Viewfactor delivery and should be available on your system. When the analysis is done, the following results may be read into the Patran database under the Analysis menu with the Action set to Read Results. The results can be visualized with any of the visualization tools under Results. Finally, you will want to look at the Patran Thermal output data in the QOUTDAT file. Use the system editor to find the first occurrence of the string '1TIME'. Note the system heat balance. This is approximately the imposed heat flux to the chamber’s bottom cap. Also note that although the temperatures are converged to high accuracy, the total system heat balance is not nearly so accurate. This illustrates the significance of the fourth power temperature dependence for radiant energy exchange. If accurate heat flows are required for a thermal analysis of a high temperature radiation environment, then very accurate temperatures must in general be calculated. The nodes numbered above 2000 are the radiosity nodes created by Viewfactor.

Main Index

jp`Kc~íáÖìÉ=nìáÅâ=pí~êí=dìáÇÉ

Index Viewfactor Analysis

Symbols V a A s

i c n i

e w f t o r a l y s

$APPROX_CURVE, 93 $AXISYM_SURFACE, 95 $CONVERGE, 92 $DIAGNOSTIC_FILE, 89 $ENDNODES, 137 $ENDSYM, 136 $EOF, 95 $GAUSS_ORDER, 94 $IN_DATA, 89 $MESSAGE_FILE, 88 $NODES, 137 $OUT_DATA, 90 $PATH, 88 $RAD_NODE_FILE, 90 $RAW_DATA, 90 $RESTART_FILE, 89 $RESTART_FLAG, 92 $RUN_CONTROL, 91 $SIZE, 136 $STATUS_FILE, 89 $SYM, 136 $TEMPLATE_FILE, 90 $TITLE, 89, 136 $ZERO, 92

errors, 1 file, 2 format, 2 surface incompatibility, 2 UID/TID, 2 execution modes, 18 mode 1, 18 mode 2, 19 mode 3, 19

F file errors, 2 filenames generic, 10 machine specific, 10 finite element analysis, 3 format errors, 2

H heated reaction chamber, 35

I input data file, 133

M A ambient radiation node, 44 analysis, 98 analysis cycle, 8, 20 axisymmetric Model, 48

C compatibility, 65 computational limitations, 129 convex surface, 47, 48

D diagnostic data, 145 diffuse surfaces, 3

E emissivity resistors rules, 156 enclosure, 25, 26 enclosure ID, 25

Main Index

material property definition, 122 material property ID, 56 memory requirements, 8 model diagnostic file example, 146, 147 model input file example, 133 MPID, 56 MSC.Patran THERMAL, 2, 3, 13, 21, 52

N nomenclature, 9

O obstruction, 128

P parallel semi-infinite plates, 32 PATQ, 80, 111 post-analysis, 110 postprocessing, 118

40 Viewfactor Analysis

Q QINDAT, 21 QOUTDAT, 15 QTRAN, 111 quick reference guide, 3

R radiation enclosure, 24, 25 radiation resistor, 54 rules, 155 radiation resistor file, 21 radiative resistor, 111 radiosity node, 111 radiosity node file, 21 radiosity node lists, 151 radiosity resistors rules, 157 raw viewfactor data, 140 resistor data, 144 run control parameter, 98

S steady-state radiative boundary conditions, 13 surface incompatibility errors, 2 surface orientation, 24, 31, 34, 36 surface pointer data, 149 symmetry, 67, 74 in 2-D XY Space, 68 in 3-D XY Space, 70 purpose, 67 symmetry operations, 67

T TEMPLATEDAT, 14, 149 examples, 61 thermal analysis, 117 thermal radiation, 3, 20 example problem, 13, 32, 35 thermal radiation analysis, 4 thermal radiation modeling, 67, 74 torus, 48

U UID/TID errors, 2

V VFAC, 20 VFAC LBC, 38 VFAC template, 58 VFCONTROL, 18 VFCTL, 17, 21, 85, 122, 6 default, 6 sample file, 96 VFDIAG, 15, 101, 107, 145 example file, 107

Main Index

VFINDAT, 14, 51, 84, 133 VFMSG, 15, 101, 103 example file, 103, 105 VFNODEDAT, 14, 21, 101, 102, 111, 151 VFRAWDAT, 14, 101, 102, 140 VFRESDAT, 14, 21, 101, 102, 111, 144 VFRESTARTDAT, 101 VFRESTARTSTAT, 101 VFRESTXT, 15, 111 viewfactor analysis, 21 overview, 8 viewfactor form, 38 viewfactor solution parameters, 98 VSUBMIT, 99

W waveband, 25

Related Documents


More Documents from "Kevin"