May 2010 EMPro 2010 - Keysightedadownload.software.keysight.com/eedl/empro/2010/pdf/... · 2014. 7....

EMPro 2010 - Using EMPro

1

EMPro 2010May 2010

Using EMPro


2

© Agilent Technologies, Inc. 2000-20095301 Stevens Creek Blvd., Santa Clara, CA 95052 USANo part of this documentation may be reproduced in any form or by any means (includingelectronic storage and retrieval or translation into a foreign language) without prioragreement and written consent from Agilent Technologies, Inc. as governed by UnitedStates and international copyright laws.

AcknowledgmentsMentor Graphics is a trademark of Mentor Graphics Corporation in the U.S. and othercountries. Microsoft®, Windows®, MS Windows®, Windows NT®, and MS-DOS® are U.S.registered trademarks of Microsoft Corporation. Pentium® is a U.S. registered trademarkof Intel Corporation. PostScript® and Acrobat® are trademarks of Adobe SystemsIncorporated. UNIX® is a registered trademark of the Open Group. Java™ is a U.S.trademark of Sun Microsystems, Inc. SystemC® is a registered trademark of OpenSystemC Initiative, Inc. in the United States and other countries and is used withpermission. MATLAB® is a U.S. registered trademark of The Math Works, Inc.. HiSIM2source code, and all copyrights, trade secrets or other intellectual property rights in and tothe source code in its entirety, is owned by Hiroshima University and STARC.

The following third-party libraries are used by the NlogN Momentum solver:

"This program includes Metis 4.0, Copyright © 1998, Regents of the University ofMinnesota", http://www.cs.umn.edu/~metis , METIS was written by George Karypis([email protected]).

Intel@ Math Kernel Library, http://www.intel.com/software/products/mkl

SuperLU_MT version 2.0 - Copyright © 2003, The Regents of the University of California,through Lawrence Berkeley National Laboratory (subject to receipt of any requiredapprovals from U.S. Dept. of Energy). All rights reserved. SuperLU Disclaimer: THISSOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THEIMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSEARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BELIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, ORCONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OFSUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESSINTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER INCONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THEPOSSIBILITY OF SUCH DAMAGE.

AMD Version 2.2 - AMD Notice: The AMD code was modified. Used by permission. AMDcopyright: AMD Version 2.2, Copyright © 2007 by Timothy A. Davis, Patrick R. Amestoy,and Iain S. Duff. All Rights Reserved. AMD License: Your use or distribution of AMD or anymodified version of AMD implies that you agree to this License. This library is freesoftware; you can redistribute it and/or modify it under the terms of the GNU LesserGeneral Public License as published by the Free Software Foundation; either version 2.1 ofthe License, or (at your option) any later version. This library is distributed in the hopethat it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of

http://www.cs.umn.edu/~metis

http://www.cs.umn.edu/~metis

http://www.intel.com/software/products/mkl

http://www.intel.com/software/products/mkl


3

MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU LesserGeneral Public License for more details. You should have received a copy of the GNULesser General Public License along with this library; if not, write to the Free SoftwareFoundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA Permission ishereby granted to use or copy this program under the terms of the GNU LGPL, providedthat the Copyright, this License, and the Availability of the original version is retained onall copies.User documentation of any code that uses this code or any modified version ofthis code must cite the Copyright, this License, the Availability note, and "Used bypermission." Permission to modify the code and to distribute modified code is granted,provided the Copyright, this License, and the Availability note are retained, and a noticethat the code was modified is included. AMD Availability:http://www.cise.ufl.edu/research/sparse/amd

UMFPACK 5.0.2 - UMFPACK Notice: The UMFPACK code was modified. Used by permission.UMFPACK Copyright: UMFPACK Copyright © 1995-2006 by Timothy A. Davis. All RightsReserved. UMFPACK License: Your use or distribution of UMFPACK or any modified versionof UMFPACK implies that you agree to this License. This library is free software; you canredistribute it and/or modify it under the terms of the GNU Lesser General Public Licenseas published by the Free Software Foundation; either version 2.1 of the License, or (atyour option) any later version. This library is distributed in the hope that it will be useful,but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITYor FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License formore details. You should have received a copy of the GNU Lesser General Public Licensealong with this library; if not, write to the Free Software Foundation, Inc., 51 Franklin St,Fifth Floor, Boston, MA 02110-1301 USA Permission is hereby granted to use or copy thisprogram under the terms of the GNU LGPL, provided that the Copyright, this License, andthe Availability of the original version is retained on all copies. User documentation of anycode that uses this code or any modified version of this code must cite the Copyright, thisLicense, the Availability note, and "Used by permission." Permission to modify the codeand to distribute modified code is granted, provided the Copyright, this License, and theAvailability note are retained, and a notice that the code was modified is included.UMFPACK Availability: http://www.cise.ufl.edu/research/sparse/umfpack UMFPACK(including versions 2.2.1 and earlier, in FORTRAN) is available athttp://www.cise.ufl.edu/research/sparse . MA38 is available in the Harwell SubroutineLibrary. This version of UMFPACK includes a modified form of COLAMD Version 2.0,originally released on Jan. 31, 2000, also available athttp://www.cise.ufl.edu/research/sparse . COLAMD V2.0 is also incorporated as a built-infunction in MATLAB version 6.1, by The MathWorks, Inc. http://www.mathworks.com .COLAMD V1.0 appears as a column-preordering in SuperLU (SuperLU is available athttp://www.netlib.org ). UMFPACK v4.0 is a built-in routine in MATLAB 6.5. UMFPACK v4.3is a built-in routine in MATLAB 7.1.

Errata The ADS product may contain references to "HP" or "HPEESOF" such as in filenames and directory names. The business entity formerly known as "HP EEsof" is now partof Agilent Technologies and is known as "Agilent EEsof". To avoid broken functionality andto maintain backward compatibility for our customers, we did not change all the namesand labels that contain "HP" or "HPEESOF" references.

Warranty The material contained in this document is provided "as is", and is subject tobeing changed, without notice, in future editions. Further, to the maximum extentpermitted by applicable law, Agilent disclaims all warranties, either express or implied,

http://www.cise.ufl.edu/research/sparse/amd

http://www.cise.ufl.edu/research/sparse/amd

http://www.cise.ufl.edu/research/sparse/umfpack

http://www.cise.ufl.edu/research/sparse/umfpack

http://www.cise.ufl.edu/research/sparse




http://www.mathworks.com

http://www.mathworks.com

http://www.netlib.org

http://www.netlib.org


4

with regard to this documentation and any information contained herein, including but notlimited to the implied warranties of merchantability and fitness for a particular purpose.Agilent shall not be liable for errors or for incidental or consequential damages inconnection with the furnishing, use, or performance of this document or of anyinformation contained herein. Should Agilent and the user have a separate writtenagreement with warranty terms covering the material in this document that conflict withthese terms, the warranty terms in the separate agreement shall control.

Technology Licenses The hardware and/or software described in this document arefurnished under a license and may be used or copied only in accordance with the terms ofsuch license. Portions of this product include the SystemC software licensed under OpenSource terms, which are available for download at http://systemc.org/ . This software isredistributed by Agilent. The Contributors of the SystemC software provide this software"as is" and offer no warranty of any kind, express or implied, including without limitationwarranties or conditions or title and non-infringement, and implied warranties orconditions merchantability and fitness for a particular purpose. Contributors shall not beliable for any damages of any kind including without limitation direct, indirect, special,incidental and consequential damages, such as lost profits. Any provisions that differ fromthis disclaimer are offered by Agilent only.

Restricted Rights Legend U.S. Government Restricted Rights. Software and technicaldata rights granted to the federal government include only those rights customarilyprovided to end user customers. Agilent provides this customary commercial license inSoftware and technical data pursuant to FAR 12.211 (Technical Data) and 12.212(Computer Software) and, for the Department of Defense, DFARS 252.227-7015(Technical Data - Commercial Items) and DFARS 227.7202-3 (Rights in CommercialComputer Software or Computer Software Documentation).

http://systemc.org/

http://systemc.org/


5

EMPro Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Using the EMPro Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Using the FDTD and FEM Toggle Buttons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Features of the Application menu bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Project Tree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Workspace Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Troubleshooting Invalid Operations within the EMPro Interface . . . . . . . . . . . . . . . . . . . . . . . 30

Creating a New Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Modifying Existing Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Boolean Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Patterns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 3D Library Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Creating Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Adding a New Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Magnetic Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Appearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 Physical Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

Defining Circuit Components and Excitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Component Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Adding a New Component . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Adding a Component Using an Existing Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Circuit Component Definition Editor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Passive Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Feed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Diode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Nonlinear Capacitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 External Excitation Editor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Plane Wave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Gaussian Beam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 Waveform Editor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Broadband . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Gaussian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Gaussian Derivative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Modulated Gaussian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Ramped Sinusoid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 User Defined . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Notes on Choosing Waveform Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Defining the Grid and Creating a Mesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Grid Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 Gridding Properties Editor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 Meshing Properties Editor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Meshing Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Defining Outer Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Absorbing Boundaries vs Reflecting Boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Liao Boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 PML Boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 PEC Boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 PMC Boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Periodic Boundary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Saving Output Data with Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103


6

Sensor Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Port Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Near Field Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Far Zone Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 SAR Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

Running Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Simulations Workspace Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Starting the Calculation Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 Running EMPrFDTD From the Command Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 Running the Calculation Remotely . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Calculation Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Multi-Processing Modules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Acceleware Hardware Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Monitoring Calculation Progress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Calculation Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

Viewing Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Results Workspace Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 Viewing Numerical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 3-D Field Displays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 The Rotations Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 The Hearing Aid Compatibility Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

Additional Tools for Customizing and Organizing Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Parameters Workspace Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Defining Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 Scripting Workspace Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Libraries Workspace Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Appendix of Geometric Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 Editing Cross-Sections for 2-D and 3-D Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Grid Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Grid Concepts Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 Choosing an Appropriate Cell Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 Grid Regions vs. Fixed Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Debugging the Grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208


7

EMPro OverviewEMPro is a three-dimensional full wave electromagnetic solver based on the FDTD and FEMmethod. Since 1994, Agilent Technologies has provided powerful yet affordableelectromagnetic simulation software to academia, industry, and government users. WhileEMPro has its roots in the simulation of cellular telephones, current uses for the softwarehave reached markets as diverse as chemistry, optics, ground-penetrating radar, andbiomedical devices, in addition to wireless, microwave circuit, and radar scatteringapplications.

EMPro provides users a great deal of freedom and functionality while creating theirprojects. Consequently, there are multiple ways to set up a project and run simulations.This section provides a general overview of the process of creating a project. The followingfigure displays the EMPro 2010 interface.

Typically, an EMPro project begins with the creation of a physical geometry. Geometricmodeling takes place within the Geometry Tools dialog box of the Geometry workspacewindow. Here, objects can be created from scratch or imported from external files. TheSpecify Orientation tab within Modify dialog box provides exceptional functionality forsituating physical parts in the simulation space.

The next illustration shows the location of the Geometry Tools dialog box.

Material definitions are created as Definition objects in the Material Editor and stored inthe Project Tree so that they can be easily applied to geometry objects by drag-and-drop.

The figure below shows the location of EMPro project Definitions.


8

After geometry objects are created and given valid material definitions, discrete circuitcomponents may be added within the Component Tools dialog of the Geometry workspacewindow. Circuit component definitions are created within the Circuit Component DefinitionEditor and stored as Definition objects in the Project Tree. A definition can then be appliedeasily to the appropriate components by drag-and-drop.

In addition to discrete sources, external excitation sources can be added within theExternal Excitation Editor. Governing waveform definitions are created in the WaveformEditor as Definition objects within the Project Tree, and are applied by drag-and-drop toobjects that require waveform definitions.

The following illustration shows the location of EMPro project Circuit Components andExternal Excitations.

Next, the grid (which provides the "blueprint" for the meshing operation) is definedwithin the Grid Tools dialog. This dialog is accessed within the Geometry workspacewindow or by double-clicking on the Grid icon in the Project Tree. General characteristicssuch as limits and cell, bounding box and padding sizes are defined in this dialog.Additionally, customized regions may be defined by adding fixed points and grid regionsso that more important regions of the project can be meshed more finely, while less-important regions can be meshed with larger cells. Object-specific grid definitions may beapplied within the Gridding Properties Editor. This editor provides options for adding fixedpoints and grid regions to specific objects based on its characteristics, which in somecases may be more convenient than defining them in within the general grid definitionsprovided in the Grid Tools dialog.

The location of EMPro project Grid Tools and Gridding Properties Editor is shown in thefollowing figure:


9

Once the grid is defined, the project can be meshed. During the meshing operation,materials are applied to the appropriate cell edges. Meshing occurs automatically in EMProas soon as the project is viewed within Mesh View. Subsequent meshes are generated bymanually pressing the Remesh Now button or by selecting the Automatic Remeshingfeature, which will generate a mesh after every change to the geometry is made. Both ofthese options are found within View Tools along the right-side of the Geometry workspacewindow. Meshing considerations such as Meshing Property and Type (magnetic orelectrical) may be specified in the Meshing Properties Editor, or by right-clicking on aspecific Parts object in the Project Tree.

The location of the meshing options dialog is shown below.

Outer Boundary conditions, which regulate how the calculation engine treats theboundaries of the project, are specified within the Outer Boundary Editor. This editor isaccessed by double-clicking on the corresponding branch in the Project Tree.

The following shows the branch where you can access the Outer Boundary Editor.

Once the various components of the project are defined, Sensor objects can be added.Sensors are simply objects that request data. There are several different types of sensorsbased on what type of data they retrieve. They can be added from the Project Tree, orwithin the Sensor Tools dialog in the Geometry workspace window.

The two locations from which you can add Sensor objects is shown in the next illustration.


10

Calculation criteria are specified within the Simulations workspace window once sensorshave been added to retrieve all desired results. Here, specifications such as source type,parameter sweep definitions, S-parameter feeds, frequencies of interest, total/scatteredfield interfaces, and termination criteria are defined for each simulation that will be run bythe calculation engine, CalcFDTD.

After running the calculation, view the results from the Results workspace window.

The illustration below shows the location of the Simulations and Results workspacewindows.


11

Using the EMPro InterfaceThe EMPro interface has been designed with new features to integrate the FDTD and FEMsimulation engine. It utilizes a workspace design and Project Tree to control the variousaspects of the EMPro simulation. The EMPro workspace comprises several parts that areintended to make it as easy as possible to create projects and run calculations. Thefollowing figure highlights the various parts of the EMPro interface.

EMPro Interface

The EMPro interface includes the following parts:

FDTD and FEM toggle buttons: EMPro provides the FDTD and FEM toggle buttons,outlined in purple color. These buttons enable you to verify if there is any violation ina valid simulation setup.

Application Menu Bar: The Application menu bar, outlined in red color, is presentat the upper left of the figure. Using this menu, you can select various options forcreating and loading a project.

Project Tree: The project tree, outlined in blue, is used to display and organize thecontents of the project within categorized branches. The row of buttons above ittoggles the various branches of the tree on and off.

Workspace windows: The series of tabs along the right of the screen, outlined ingreen color, store the various workspace windows so that they can be easily accessedwithin the project workspace. Minimized workspace windows are stored within theactual workspace. Open workspace windows are displayed in this area as well.


12

In this section, you will learn about:

Using the FDTD and FEM toggle buttonsFeatures of the Application menu barBranches of the Project TreeMajor functions of the workspace windowsTroubleshooting Invalid Operations

Using the FDTD and FEM Toggle ButtonsEMPro integrates the FDTD simulation engine and FEM simulation engine. Using the FDTDand FEM toggle buttons at the bottom left of the EMPro window, you can switch the userinterface to check whether the setup is valid for:

Both FDTD and FEM simulation: Click the FDTD and FEM buttons to validate thesettings for both FDTD and FEM simulation engines. The following toggle selection willcheck whether all settings are valid for both FEM and FDTD:

FDTD simulation: Click the FDTD button to validate the settings for FDTD. Thefollowing toggle selection will check whether all settings are valid for FDTDsimulations. In this case, only setup violations that would prevent a valid FDTDsimulation will be flagged.

FEM simulation: Click the FEM button to validate the settings for FEM simulationengines. The following toggle selection will check whether all settings are valid forFEM simulations. In this case, only setup violations that would prevent a valid FEMsimulation will be flagged.

The FDTD and FEM toggle buttons are introduced because some of the features orsettings can be applied only to either of the simulators. Examples of this:

Only port feed components can be used with FEM, whereas with FDTD non-linearcircuit components can also be used.FEM supports only planar and far field sensors.

The toggle buttons will activate errors (in yellow) and warnings (in blue), moreinformation is displayed on the specific error or warning by moving the mouse over theerror or warning sign.


13

Besides triggering validity checks and violations for a specific simulation, the togglebuttons will also grey-out the portions of the UI that are not valid for the selection. If noneof the toggle buttons are pressed, the user interface will not be checking for any violationsfor a valid simulation setup.

Features of the Application menu bar

File Drop-Down Menu

New Project - Opens a new, blank project.

New Project from Template - Opens an project from a saved template.

Open Existing Projects - Opens an existing project.

Save Project - Saves a new project to a specified directory.

Save Project As - Saves the current project to another filename and opens thatproject.

Save Project Copy As - Saves the current project to another filename.

Archive Project - Archives the current project.


14

Unarchive Project - Unarchives an existing archived project.

Import - Imports an existing CAD file, voxel object file, AMDS 2007.x, AMDS meshfile, or any other supported file format into the current EMPro project.

Export Parts - Exports parts from EMPro to a file in a specified directory.

Managing Project Templates - Select this option to create a new template or toassign a default template to be loaded whenever a new project is opened.

To create a new template, simply select an existing saved project file. This file will besaved as a template. Click the Create New button and assign the template a name. Thedefault template is a blank project, but this can be changed to any other template thathas been saved in this dialog.

Templates are useful for beginning new projects that contain identical parts to existingprojects. For example, a new project may reuse the same geometry and discretecomponents that has already been created in an existing project file. In this case, theexisting project file can be used to create a template so that this common geometry canbe used over and over without having to rebuild the project from scratch.

Recent Projects - Lists the five most recently loaded EMPro projects.

Quit - Closes the current EMPro session.

Edit Drop-Down Menu

Undo/Redo - The Undo button will "undo" any action or series of actions. Similarly,the Redo button will repeat any actions that were mistakenly erased.

Project Properties - The Project Properties Editor contains the Display Units tab,which defines the default units used in EMPro and controls the units displayed in thegeometry window. This setting will also control the default unit setting assigned tovalues that are entered by the user without units. A Unit Set definition (SI Metric,Gaussian Metric (GCS) or US Customary) can be applied to all of the parameterslisted under this tab, or a Custom definition can be applied to any individualparameter.

The following illustration shows the Display Units tab.

Display Units tab in the Project Properties Editor


15

Under the Advanced tab, the user has the option of modifying the project timestep. Thetimestep that is used for the simulation is computed from the grid definition and thematerials that are used in the space. In some special circumstances, the user may wish tomake that timestep a little bit smaller to improve results or compensate for some otherspecial conditions. The Custom Timestep Multiplier is a factor by which the computedtimestep is multiplied to obtain the final timestep that is actually used. For moreinformation on how the timestep is calculated from the grid definition, refer to GridAppendix (using).

Under this tab, the user can also assign New Part Modeling Units, which defines the unitsassigned to a new part when it is created.

NoteThis unit defines how the part is modeled internally, and cannot be changed once the part is created. Note that this unit is not the same as Display Units.

The following figure shows the Advanced tab.

The Advanced tab in the Project Properties Editor


16

Application Preferences

The three main tabs under Application Preferences are General, Interface, and Modeling.

General tab under Application Preferences

The options in the Startup section control what happens at the start of EMPro.Show Tip Of The Day, when enabled, will display a pop-up with a helpful tipabout how to use EMPro.


17

Show License Window, when enabled, will cause EMPro to show the licensingwindow when the application starts where the user can specify where the licensefile/server is.

Under Templates, enter the default folder location where EMPro project templates arestored.Rendering Options controls several display options in the EMPro GUI:

The Transparency Algorithm determines the way EMPro renders the PartsOpacity of an object.

Z-Sort is the most efficient rendering option, but may contain artifacts(small areas of inaccurate rendering).Depth Peeling is supported only on certain graphics cards. It is moreaccurate but runs slower than Z-SORT.Painters Algorithm is the most accurate rendering option, but runs theslowest.The Transparency Algorithm setting only affects the way an object isdisplayed in the EMPro interface, and will not change calculation results.

Text Color defines the color of text automatically displayed in geometry view,such as coordinates, lengths, etc.

The Undo/Redo History section controls the how many items can remain on the undostack. It gives the user the ability to enhance the application performance by limitingthe memory used by the GUI.

The Interface Tab

Interface tab under Application Preferences


18

Object Editing: Controls what happens when a new object is added to the ProjectTree (i.e., right-clicking on the tree to add a new material).

None, will simply add the new object to the tree.Edit Name, will add the object to the tree and provide a blinking cursor to addthe object's name in the tree.Edit Properties, adds the object and prompts the appropriate dialog window toimmediately pop-up in order to edit the new object's properties (i.e., will promptthe MATERIAL EDITOR to pop-up).

Layout: Enables you to set their preferences for saving or restoring the GUI layout.Tool Bar Icon Size: Adjusts the size of the icons in the toolbar (i.e., New Project,Loading Existing Projects, etc.).Project Tree: The Icon/Text Size scroll adjusts the size of the items in the ProjectTree.Workspace:

Show All Tabs shows all workspace windows in tabbed workspace regardless ofwhether they are active or not.Show Only Active Tabs stores only the active tabs that are stored in the projectworkspace.Don't Show Tabs removes the tabbed workspace. Windows can still be accessedin the View menu in the Application Toolbar.

Information: The Information pane allows you to set their preference for decimalprecision in tooltips and the dimensions of bounding boxes.Appearance: The Appearance pane allows you to change the appearance of theEMPro GUI and buttons within.


19

The Modeling Tab

In general, this tab provides several options for adjusting the color and appearanceof faces, edges, vertices, components, and the background.

Modeling tab under Application Preferences

The Graph Tab

The Graphs Tab contains check boxes (listed below) that enable the user to select whatinformation appears in the Plot Properties tab of a plot.

Project IDProject NameSimulation IDSimulation NumberSimulation NameRun IDRun NumberProject : Simulation : RunResult TypeSensorSensor TypeDomainField TypeMisc Info


20

Project TreeThe Project Tree provides a tree-structured representation of the active project. It isorganized into the following branches:

PartsCircuit ComponentsExternal ExcitationsStatic voltage PointsSensorsDefinitionsSimulation DomainScriptsGraphsGroups

The EMPro Project Tree is easy to manipulate by means of branch and object togglebuttons. This section describes in detail the branches and display of the tree as shown inthe following figure.

Project Tree


21

Branches of the Project Tree

Parts

The Parts branch organizes the physical parts of a project. It also lists material definitionsand modeling operations applied to any parts object in the tree. It is possible to organizesimilar parts objects in groups with an Assembly by right-clicking and selecting CreateNew: Assembly, as shown in the next illustration. Any parts object or assembly can beexported to a CAD file using this right-click menu as well. For information about addinggeometry parts to the project, refer to Using the Geometry Workspace (quickstart).

Adding a new Part or Assembly from the Project Tree


22

Circuit Components

The Circuit Components branch organizes discrete circuit components in a project. It alsolists any associated Definition objects applied to a component. Then next figure showshow to add a component by right-clicking in the Project Tree. For information aboutadding new components and associated tools, refer to Defining Circuit Components andExcitations (using).

Defining a new Circuit Component from the Project Tree

External Excitations

The External Excitations branch organizes the external excitations applied to a project. Italso lists any associated Definition object applied to an external excitation, such as awaveform. How to add an external excitation by right-clicking in the Project Tree is shownin the following illustration. For information about defining external excitations within theExternal Excitation Editor, refer to Defining Circuit Components and Excitation (using).

Creating a new External Excitation from the Project Tree


23

Static Voltage Points

The Static Voltage Points feature is used to create and place a voltage point on an objectmade of PEC material.These points are used by the LaPlace static solver to initialize the starting E field withinthe problem space to voltage values assigned by the user.With no static voltage points defined, the initial E field at the beginning of the FDTDcomputation is set to 0 V/m at each cell edge.To use this feature, right-click the Static Voltage Points branch of the Project Tree andselect New Static Voltage Point. A static voltage point will be added to the project.Double-click the new point to edit the points properties.Static Voltage Points are specifiedby graphically picking or manually entering the location of the staticvoltage point and by specifying the voltage. Click the Done button to apply your changes.

Only one entry per conducting object is necessary. Duplicate entries will overwrite thevoltage preset value. (PEC outer boundaries and any untagged metal objects will bepreset to 0V. If the boundary is PML, the fields at a PML boundary will be initializedappropriately to prevent non-physical reflections at the interface.)Additionally, static voltage points can be used by themselves to excite an objectcontaining PEC materials, or in conjunction with an external excitation or a discretesource.Following figure shows the available parameters when creating a static voltage point.

Sensors

The Sensors branch organizes the sensors defined in a project. Sensors are responsible forsaving all data collected during a calculation. They are added by right-clicking on theappropriate branch of the Project Tree and choosing the desired sensor, as seen below.For more information about sensors and related definitions, refer to Saving Output Data


24

with Sensors (using).

Adding a Near Field Sensor from the Project Tree

Definitions

The Definitions branch stores definitions that can be applied to or shared with otherobjects within the project. To add a new definition object, right-click on the branch andselect the desired object, as shown in the next illustration. This definition is applied toother objects in the Project Tree by clicking and dragging the definition object onto thedesired object. Several editors are accessible from this branch depending on what type ofdefinition object is selected.

NoteFor more information about the Material Editor, refer to Creating Materials (using). For more information about the Circuit Component Definition Editor, refer to Defining Circuit Componentsand Excitations (using). For more information about the Waveform Editor, refer to Defining Circuit Components and Excitations(using). For more information about several of the sensor editors, refer to Saving Output Data with Sensors(using).

Accessing a Definition Editor from the Project Tree


25

Simulation Domain

The Simulation Domain branch stores definitions associated with the outer boundaries ofthe project, as well as the grid and mesh, as seen below.

Double-clicking the Boundary Conditions icon will bring up the Boundary ConditionsEditor which is valid for both FDTD and FEMDouble-clicking the FDTD Grid icon will bring up the Grid Tools dialog, used tospecify the characteristics of the grid.Double-clicking the FDTD Mesh icon will enable Mesh View.Double-clicking the FEM Padding icon will bring up the FEM Padding Editor.

Viewing the Simulation Domain branch

Scripts

The Scripts branch stores user-defined scripts. Right-click on this branch to add a newscript or to import an existing macro or function script to the project. After adding thescript, right-click on the script object, as seen in the following illustration, to execute oredit the script in the Scripting workspace window. For more information on scripting, referto Additional Tools for Customizing and Organizing Projects (using).

Accessing scripting tools in the Project Tree

Graphs


26

The Graphs branch organizes the graphical output associated with data collected during acalculation. For more information about viewing calculation results, refer to ViewingOutput (using).

Groups

The Groups branch enables you to create fully customizable short-cut groups that mayinclude any grouping of objects (Parts objects, Sensor objects, Definition objects, etc.).The Groups branch is thus a tool for users to conveniently organize their projects. It alsocontains automatically generated groups created by EMPro to store deletions, additions, orother modifications to CAD files. To add a shortcut group, right-click on the branch asseen in the following figure. For more information on how to import CAD files, refer toCreating Geometry (quickstart).

Adding a Group within the Project Tree

Customizing the Display of the Project Tree

The row of buttons above the Project Tree toggle the branches of the tree on and off andsimplify the display. This row is shown below. Active branches are displayed as fullycolored buttons, and hidden branches are dulled.

Toggling the branches of the Project Tree

Expanding and Collapsing the Project Tree Display

In order to make the tree easier to navigate, EMPro enables the user to manually expand(+), collapse (-), and toggle the visibility of each branch. Demonstrates an expanded andcollapsed branch in red.

Expanding and collapsing individual branches in the Project Tree


27

Additionally, it is possible to expand or collapse all branches by right-clicking on anybranch in the tree and selecting the Tree icon, as seen in below.

Expanding and collapsing all branches in the Project Tree

Workspace WindowsThe EMPro workspace windows are a series of windows that each has its own designatedfunction. Each window is described below. This section also explains how to display activewindows and how to control the display of each workspace window button in the tabbedworkspace.

The following illustration shows the location of the EMPro workspace windows.

Workspace windows toolbar


28

Summary of Workspace Windows

Geometry

The Geometry workspace window comprises the main project viewing area. The windowcontains four main tools used to add and edit the fundamental elements of a project:

Geometry ToolsComponent ToolsFDTD Grid ToolsSensor Tools

This window also contains buttons to manipulate the project view. The capabilities of thiswindow are detailed in Creating Geometry (quickstart).

Simulations

The Simulations workspace window provides the main interface to define simulations tosend to the calculation engine. Each time the project is modified and saved, the user mustdefine a new simulation to register the change. This workspace window stores definitionssuch as source types, parameter sweeps, S-parameters, frequencies of interest,scattered/total field interfaces and termination criteria that are specific to a calculation.For a detailed discussion about setting up and running simulations within EMPro, refer toRunning Calculations (using).

Results

The Results workspace window stores all of the results available for a particular project. Itis also possible to load the results of a past project into this window without having to loadthe entire project itself. This makes it convenient to compare the results of severalprojects in a single interface. For more information about this workspace window andreviewing results Viewing Output (using).

Scripting

The Scripting workspace window enables you to create, edit, manage and execute user-


29

defined scripts. For more information on the capabilities of this window, refer to AdditionalTools for Customizing and Organizing Projects (using).

Libraries

The Libraries workspace window enables you to create and save collections of objectsgrouped by category. Because a library is saved to its own filename (and not to a specificproject), it is easy to import common objects and definitions to multiple projects. Forinformation about the capabilities of this window, refer to Additional Tools for Customizingand Organizing Projects (using).

Parameters

The Parameters workspace window enables you to create, edit, and delete parametersthat are referenced universally throughout the project. A parameter can be a simplenumeric value or a mathematical expression. They can be used to run "ParameterSweeps," which are specified in the Simulations workspace window. Details about thecapabilities of this window can be found at Additional Tools for Customizing andOrganizing Projects (using).

Organizing Project Workspace

There are three ways to display and organize active windows in the project workspace:Maximized, Cascade and Tile.

In Maximized mode, windows are maximized to take up the entire workspace. Switchingtabs in tabbed workspace will bring the corresponding window to the front.

In Cascade mode, the windows appear stacked on top of each other, each with its owntitle bar showing. Users can switch windows by clicking on the title bar, or by using thetabs at the top of the workspace. To assume this mode, select the Cascade mode icon inthe lower right-hand corner, as seen in below.

Project workspace view modes

In Tile Mode, the workspace automatically sizes any non-minimized windows to fill up thespace without overlapping. This is useful in order to quickly view all windows as large aspossible. Any minimized windows will remain minimized. To assume this mode, select theTile mode icon in the lower right-hand corner, as seen in the previous illustration.

Tabbed Workspace Window Display

All of the workspace windows are easily accessible from the workspace to the right of the


30

screen. Unless the project preferences are defined to show all windows, only activeworkspace windows are located in this workspace. Any additional workspace windows canbe added to the active workspace by right-clicking and selecting the desired window, or byselecting View in the Application Menu Bar. If a window is already open and present in theworkspace, right-clicking in the tabbed workspace or using the View menu to select awindow will instead close the window. Similarly, the windows can be closed by clicking thebutton in the upper-right corner of the appropriate tab.

The tabbed workspace, displaying only Geometry and Scripting workspace tabs

Troubleshooting Invalid Operations within the EMProInterfaceThe EMPro interface will display an error or warning icon, to indicate when any invaliddefinition is created within a project. This error icon will appear in the dialog or editorwindow as the invalid definition is created, as well as in the Project Tree beside the invalidobject or definition. Additionally, an error icon will appear next to the project's name atthe top-level of the Project Tree to indicate globally that the project contains an invaliddefinition. A few of the most common invalid operations are discussed below, but in mostcases the source of an error are easily pinpointed by holding the mouse over the erroricon. A brief message will appear, summarizing the source of the error and the stepsnecessary to validate the project.

The error icon often appears when objects are created without applying the appropriatedefinitions. For example, the error icon will appear if a Parts object is created withoutapplying a material definition, or if an External Excitations object is added without anapplied waveform definition. In these cases, the error will be resolved by dragging-and-dropping the appropriate Definitions object (if one exists) onto the invalid object in thetree.

Other common errors occur due to invalid or conflicting FDTD grid definitions. If griddefinitions are assigned that cause the project to exceed the maximum number of cells ormemory size (defined in the Limits tab of the Grid Tools dialog), an the error icon willappear. In this case, the limit may need to be increased or removed, or else the numberof cells in the project will need to be reduced by increasing their minimum cell size. EMProwill also generate an error if the Maximum Cell Step Factor defined in the Limits tab is too


31

low to create valid transition regions, or if a Minimum cell size definition is specifiedanywhere in the interface that falls below the universal Minimum Cell Size definition in thistab. For more information about grid definitions and associated invalid operations, refer toDefining the Grid and Creating a Mesh (using).


32

Creating a New GeometryAfter selecting Geometry Tools in the drop down list of the Geometry workspace window,click Create to prompt a drop-down menu to appear. This menu includes the followingmodeling operations:

ExtrudeRevolveSheet BodySheet Body From FacesWire BodyBondwire

The following figure shows the Create button that displays when Geometry Tools isselected.

Additionally, these tools are accessible from the Project Tree by right-clicking the PARTSbranch, as seen below.

Accessing the Create menu from the Project Tree

Selecting any of these operations will prompt a similar series of cross-sectional editingtools.

Specify Orientation Tab

The Specify Orientation tab provides tools for orienting geometric parts in the simulationspace. For a detailed discussion of this tab, refer to Orienting Objects in the Simulation


33

Space (quickstart).

Edit Cross Section Tab

In the Edit Cross Section tab, four toggle buttons including Shapes, Constraints, Tools,and Snapping. Each have a corresponding series of buttons in its own toolbar below.Clicking these labeled buttons will toggle the corresponding toolbars on and off. All ofthese buttons, in addition to the View Tools detailed earlier, are also available in the drop-down menus in the upper-left part of the screen.

Additionally, a button labeled Construction Grid is available to edit the spacing of thevisible grid lines in the 2-D sketcher. (This has no impact on the FDTD grid definition.).

The Edit Cross Section tab of the Geometry workspace window

The current object is named in the box labeled Name in the upper right section of the EditCross Section tab. If a name is not defined in this box, the object is assigned a defaultname in the Project Tree when it is added to the project. The object can be renamed atany time in the Project Tree by right-clicking the object and selecting Rename.

To the right of the Name dialog box are two buttons, Undo and Redo. Clicking the Undobutton will undo all actions carried out in the Edit Cross Section tab. Similarly, the Redobutton will repeat any actions mistakenly erased during an undo operation.

ShapesThe Edit Cross Section tab contains a number of Shapes sketching tools that are useful for


34

creating simple 2-D geometries for wire bodies and sheet bodies. They also serve as acommon starting point to define 2-D cross sections for 3-D bodies such as extrusions,revolutions, and more complicated solid modeling operations. The Shapes tools areselected by clicking their respective icon.

Pressing |Esc| or |Backspace| will back-up one step when using a multi-stepcreation tool.Pressing |Esc| a second time will deactivate the edge creation tool and activate thedefault Select tool.Pressing |Tab| will bring up a dialog to specify the position.

The following is a list of the 2-D shapes available in EMPro.

Straight EdgePolyline EdgePerpendicular EdgeTangent LineRectanglePolygonN-Sided Polygon3-Point ArcArc Center, 2 Points2-Point ArcCircle Center, Radius3-Point Circle2-Point CircleEllipse

NoteFor a detailed description of each shape tool, refer to Shapes (using) in the "Appendix of GeometricModeling".

ToolsThe Tools buttons provide useful functionality to users while sketching in the 2-D sketcher.

Select/ManipulateTrim CurvesInsert VertexFillet Vertex

NoteFor a detailed description of each 2-D sketcher tool, refer to Tools (using) in the "Appendix of GeometricModeling".

ConstraintsConstraints are restrictions placed on geometric parts that must be satisfied in order toconsider the model valid. They ensure that the user's intent is sustained throughout acalculation when parameters may change. Some objects are created with constraintsalready embedded. For instance, a rectangle is composed of four straight edges that areconstrained perpendicularly as seen in preceding illustration. Other constraints are user-defined by means of Constraint tools.


35

Applying a constraint to an object will often affect other characteristics of the object. Forinstance, applying a horizontal constraint to one side of an irregular quadrilateral will mostlikely change the length of one or more sides and the angles that form with thoseconnecting sides. Thus, it is important to lock any points that are intended to stay static.There are two main ways to do this:

By selecting the Lock Constraint tool and clicking the the appropriate vertex or side.By selecting the Select/Manipulate tool, right-clicking the appropriate vertex or side,and selecting Lock Position, as shown below.

Locking or editing a vertex's position with the Select/Manipulate tool

NoteFor more about the Select/Manipulate tool's functionality, refer to Select/Manipulate (using) in the"Appendix of Geometric Modeling".

Each type of the following types of Constraint tools has its own green symbol or letter thatis visible when the mouse is held over the constrained segment.

HorizontalVerticalCollinearParallelPerpendicularTangentConcentricAngleDistanceEqual LengthEqual DistanceRadiusEqual Radius


36

NoteFor a detailed description of each constraint, refer to Constraints (using) in the "Appendix of GeometricModeling".

SnappingSnapping tools are available to facilitate the exact placement of vertices on the sketchingplane. When snapping is enabled, the mouse will be snapped to the closest of one ormore snapping landmarks if one comes within range. For example, if Snap To Grid Lines isselected, the mouse is moved or snapped to points on the closest grid line as it is movedaround in the sketching plane. This makes it much easier to place a vertex in the desiredposition without having to zoom in to a discrete position. Blue dots and blue linesrepresent the snapped location of the mouse when snapping is enabled.

In the case that the mouse is not within sufficient range of a selected landmark listedbelow, a vertex will be placed at its exact location on the sketching plane as if snappingwere not turned on. (For example, if the mouse is dragged to the middle of a cell and theSnap To Grid Lines option is selected, the vertex will be placed in the center of the cellbecause it is not close enough to a surrounding grid line.)

Several snapping options can be selected at a time, in which case, the vertex will besnapped to the closest landmark that is within range of the mouse.

Snap To Grid LinesSnap To Grid/Edge IntersectionsSnap To VerticesSnap To EdgesSnap To Edge/Edge Intersections

NoteFor a detailed description and image of each, refer to Snapping (using) in the "Appendix of GeometricModeling".

Customizing the Construction GridThe Construction Grid drop-down controls the appearance of the grid without impacting itsactual cell size.

Automatically Adjust Line Spacing causes the construction grid to adjust its linespacing with the current zoom level. As you zoom in, the lines are moved to be closerto each other. As you zoom out, they decimate and become further apart.Line Spacing is available when automatic isn't checked. This is the spacing betweenadjacent lines of the construction grid.Highlight Interval controls the interval which lines are highlighted. Every "Nth" linewill be made bold.Mouse Spacing controls the minimum resolvable distance by the mouse. As you movethe mouse, you will be unable to move between two points closer than this specieddistance.

Editing the Construction Grid


37

3-D Operation Tabs

If subsequent tabs are available to the right of the Edit Cross Section tab, continue on tocomplete a 3-D operation. (These tabs are not available for 2-D objects.)

The figure below shows the Advanced drop-down menu inside of the Extrude tab,available when an Extrude operation is selected. This menu contains operations that canbe applied to the 3-D object. For more information on these operations, refer to Advanced3-D Solid Modeling Operations (using) in the "Appendix of Geometric Modeling".

The Extrude tab of the Geometry workspace window


38

Modifying Existing GeometryThe Modify button in the Geometry workspace window may be selected to modify thegeometry of existing objects in the project.

Specify OrientationChamfer EdgesBlend EdgesShell FacesLoft FacesRemove FacesOffset Faces

NoteFor images of each of these operations, refer to Modifying Existing Geometry (using) in the "Appendix ofGeometric Modeling".

Boolean OperationsThe following boolean operations are available in EMPro:

Two PartsExtrudeRevolve

The Two Parts tool provides several boolean operations to subtract, intersect, or unite twoobjects. For these operations, one object must be selected to be the Blank, and the otherthe Tool which acts on the blank.

Holes may also be extruded or revolved through any part with its respective tool in thismenu. An object is selected in the Pick Blank tab and the cross section of the hole issketched and oriented in the Edit Profile and Feature Orientation tabs, as described in theEdit_Cross_Section_Tab and Specify_Orientation_Tab, sections respectively. Then theshape of the removed section is specified in the Extrude Boolean tab, or Revolve tabdepending on which operation is selected. The Preview tab shows a preview of the objectbefore the changes are formally applied to the project. For more information on definingextrusions or revolutions, refer to 3-D Solid Modeling Options (using) in the Appendix ofGeometric Modeling. An image of each boolean operation is available in BooleanOperations (using) in the "Appendix of Geometric Modeling".

PatternsPatterns are created in EMPro by replicating a single selected object multiple times in oneof the organized arrangements listed below.

Linear/RectangularCircular/Elliptical


39

NoteFor the definitions and images associated with these patterns, refer to 3-D Patterns (using) in the"Appendix of Geometric Modeling".

3D Library ComponentsEMPro consists of a built-in library of parameterized 3D objects that includes the followingtypes of shapes:

BoxCylinderPiramidRingSphereConeHelix

To insert the shape, click on the symbol at the bottom of the EMPro window. You can alsoedit the parameters prior to entering the object or modify them after insertion. The objectcan be positioned after insertion by using the Specify Orientation menu.

If a new parameter name is used that does was not defined in the Parameters menu inEMPro, the new parameter name will automatically be added.


40

Creating MaterialsIn this section, you will learn how to:

Add materials to your EMPro projectDefine materials and apply them to geometric objects

Once objects are created and situated correctly in the simulation space, materialdefinitions must be added or else the project will not be considered valid. The electricaland magnetic materials available within EMPro are detailed in this section.

The Material Editor window is the main interface used to define materials to be applied toobjects in a simulation. The series of tabbed windows within the editor are used to definea material based on its constitutive parameters. After adding materials to the project,simply drag-and-drop the material in the Project Tree onto the desired geometry to applyit to that object.

The following section describes the options under each tab within the Material Editor.

Adding a New MaterialTo add a new material, right-click Definitions: Materials branch of the Project Tree andselect New Material Definition, as seen in the following illustration. A Material object willbe added to this branch. Depending on the project preferences, the Material Editor windowwill appear automatically. If not, simply double-click on this object to bring up the editor.Similarly, double-click on any existing Material icon to edit an existing material within theMaterial Editor.

For more information about project preference definitions, refer to ApplicationPreferences (using).

Adding a New Material Definition to the project

Once the Material Editor window is open, type in the name of the new material in theName dialog box. Define the Type of material as Physical or Freespace. Freespace is themost basic material definition. Every other type of material is included within the Physicaldefinition, in which case the Electric and Magnetic types should be assigned in theirrespective drop-down lists below.

There are five electrical and magnetic material types available in EMPro.

FreespacePerfect Conductor


41

IsotropicDiagonally AnisotropicAnisotropic (Electric only)

Although frequency-independent materials require the least memory during FDTDsimulations, there are some cases in which frequency-independent materials are notappropriate. Frequency-dependent or dispersive materials should be used in theseinstances. Some common examples of frequency-dependent materials are high watercontent materials such as human tissues, and metals excited at optical frequencies. EMProhas the capability of simulating electric and magnetic Debye and Drude materials such asplasmas, Lorentz materials, and anisotropic magnetic ferrites, as well as frequency-independent anisotropic dielectrics and nonlinear diagonally anisotropic dielectrics. Theseadditional sub-types are specified within the Isotropic, Diagonally Anisotropic andAnisotropic definitions.

The following sections will detail the various types of materials.

Freespace

Freespace is the most basic material. By default, the EMPro problem domain is initializedto free space. This material sets relative permittivities and permeabilities to one, andconductivities to zero.

The following figure shows the Material Editor when the Freespace material is defined.Notice that no Electric or Magnetic tab is available, since both are defined as Freespacematerial.

Defining a Freespace material

Perfect Conductors

A Perfect Conductor has infinite conductivity and all fields found within it are zero. It hasthe same settings as the Freespace material, as seen in the figure above. It shouldtypically be used as an approximation when a good conductor is needed in anelectromagnetic calculation and losses aren't important. Attempting to include the effectsof a good conductor (rather than perfect conductor) may be difficult since the wavelengthinside the good conductor will become very small, requiring extremely small FDTD cells toprovide adequate sampling of the field values inside the material. This can, however, beovercome by checking the Surface Conductivity box in the Edit Material dialog.


42

NoteYou can read more about the Surface Conductivity box in the Complex Permittivity, Loss Tangent andSurface Conductivity Correction Overview section.

Electric Materials

Isotropic Materials

EMPro includes several Isotropic materials:

NondispersiveDebye/DrudeLorentzSampled (FDTD only)Nonlinear (FDTD only)

The next figure shows the Material Editor when an Isotropic material is defined. Note thatonly the Electric tab is available since Magnetic is defined as Freespace. If Magnetic wasdefined as another type, a Magnetic tab would be available as well.

Defining an Electric material

Nondispersive

Nondispersive material properties do not very with frequency. The continuous-timeexpressions of Maxwell's equations for linear, isotropic, nondispersive materials which willbe discretized in EMPro are:

and

where:


43

represents the electric permittivity,

represents the electric conductivity,

represents the magnetic permeability, and

represents the magnetic conductivity.

Defining a Nondispersive material

Debye/Drude

For a Debye/Drude material, the electrical Conductivity ( ) in , Infinite Frequency

Relative Permittivity ( ), Number Of Poles, Static Relative Permittivity ( ), andRelaxation Time ( ) in seconds must be specified. For a Debye material, must equalzero. A non-zero conductivity value results in a Drude material.

NoteThis is discussed in detail in Chapter 8 of the Kunz and Leubbers text [1(using)].

Defining a Debye/Drude material


44

These parameters cannot be set arbitrarily or instability can occur in FDTD simulations.One constraint is that the FDTD timestep must be small enough to accurately calculate thetransient behavior of the material. If the timestep is 3% of the relaxation time or smaller,the time variation of the material parameters should be sufficiently resolved. Typically, thetimestep is a very small fraction of the relaxation time. In order to be clear about thesigns in the following discussion, note that we are using the engineering time variation of:

and we are defining the complex permittivity as:

For the FDTD calculation to be stable, the imaginary (loss) part ( ) of the complexpermittivity, including the effect of the conductivity term, must be positive for allfrequencies from zero frequency to infinite frequency. This condition results in a passivematerial. If is negative, then the material has gain and FDTD simulations will becomeunstable as the field amplitudes grow.

NoteSee equation 8.29 of the Kunz and Leubbers text [1(using)].

For a Debye material ( ), stability is assured by setting to a larger value than .

In order to have realistic behavior at high frequencies, should be no less than one andshould not be much larger than one. Thus the condition for strictly Debye material to bestable for FDTD simulations is:


45

> = 0

If the conductivity is not zero, then the material has Drude behavior. There are differentconditions that can be satisfied for the imaginary part of the complex permittivity to bepositive so that FDTD simulations produce stable results. If the static permittivity isgreater than the infinite frequency permittivity then the conductivity can have any positivevalue. This results in the simplest set of conditions for a stable Drude Material:

>

These conditions are, however, too restrictive to specify general Drude materials. The

more general Drude conditions are:

If ( ),

then:

otherwise: 0

where is the Freespace Permittivity of 8.854e-12 !img227.png!.

NoteMore general conditions for Drude materials can be determined from the discussion in Chapter 8, Section3 the Kunz and Leubbers text [1 (using)].

Lorentz

Stability in Lorentz materials for FDTD simulations should be obtained as long as

Conductivity and the FDTD timestep is 3% of the relaxation time or less. The limitson the material parameters are:

> 0

> 0


46

Defining a Lorentz material

Sampled

This material enables you to enter multiple relative permittivities and conductivities at onetime. It will behave like a nondispersive material when the calculation engine is called andthe Wideband Eval Frequency dictates what parameters to use.

NoteThis is not a dispersive material and will not automatically be converted to one.

Defining a Sampled material

Nonlinear


47

The relative permittivity of a nonlinear isotropic dielectric material satisfies:

Where:

is relative permittivity

E is instantaneous cell edge E-field

is static (low ) relative permittivity

is infinite relative permittivity

is the E magnitude above which the material becomes non-linear

is a scaling term

, and are coefficients

NoteNonlinear materials are not supported in FEM simulations.

Defining a Nonlinear material


48

Diagonally Anisotropic

The definitions for a Diagonally Anisotropic are equivalent to those correspondingdefinitions detailed for Isotropic materials, except the definitions in each of the principledirections are independently specified.

Anisotropic

Frequency-independent Anisotropic materials are defined in EMPro by the relativepermittivity, , and Conductivity, , tensors.

Defining an Anisotropic material


49

The parameters below Conductivity represent the terms of and the parameters belowPermittivity (Infinite Frequency) represent the terms of as follows:

The conductivity and permittivity for frequency-independent anisotropic dielectricmaterials are represented by and , unlike the equations for linear, non-dispersive,frequency-dependent, isotropic materials. These are used in the time-domain FDTDupdate equations in place of and :

Complex Permittivity, Loss Tangent and Surface ConductivityCorrection Overview

Complex Permittivity

The value of complex permittivity may need to be calculated for some materials. The realpart of the complex permittivity may be used for the relative permittivity. The conductivitycan be calculated from the imaginary part of the complex permittivity by multiplying by a


50

desired output frequency value (in radian frequency), as shown by:

Loss Tangent

The loss tangent can be entered directly into EMPro when it is known, typically for gooddielectrics. The FDTD engine can then calculate the conductivity as a function of frequencyusing:

Surface Conductivity Correction

In the case of a frequency-dependent conductor, the Surface Conductivity Correction_ boxcan be checked to correct the conductivity of a material for a single frequency of asinusoidal excitation. This is necessary in cases where the penetration and loss in goodconductors needs to be included in the calculation. Enabling this option allows for thereduction of wavelength in these materials without reducing the cell size to maintain the10 !img249.png! limit.

Magnetic MaterialsEMPro also includes several types of magnetic materials. Many of these materials aresimply the magnetic counterpart to the dielectrics described in Electric Materials. Allrestrictions noted in the Electric Materials section apply to their magnetic counterparts.

The figure below shows the Material Editor when a Magnetic Isotropic material is defined.Note that only the Magnetic tab is available since Electric is defined as Freespace. IfElectric was defined as another type, a Electric tab would be available as well.

Defining a Magnetic material


51

Isotropic

Nondispersive.

See Nondispersive in the Electric Materials section.

Debye/Drude.

See Debye/Drude in the Electric Materials section above.

Magnetized Ferrites.

The first parameter related to magnetized ferrites is the Applied Field, ( ). Enter its

value in units of . This number will be used to calculate the Larmor precession

frequency ( ),

where is the gyromagnetic ratio ( ).

Next, enter the Internal Magnetization ( ) in units of . This number is used to

calculate the saturation frequency ( ), .

Then, use the Damping Coefficient to account for damping in the ferrite or of anyabsorption of power due to the ferrite. Finally, enter the direction of the biasing field usingthe spherical direction fields THETA and PHI.


52

NoteThere are several informative references that discuss the form of the permeability tensor used for theferrites [5 (using),6 (using),7 (using),8 (using)]. (The first two references do not discuss the dampingcoefficient.) See the Kung text for parameters for some commercially available ferrites [8 (using)].

NoteFerrite materials are not supported in FEM simulations.

Defining an Magnetized Ferrite material

Sampled.

See Sampled in the Electric Materials section above.

Nonlinear.

See Nonlinear in the Electric Materials section above.

Diagonally Anisotropic

See Diagonally Anisotropic in the Electric Materials section above.

AppearanceUse the Appearance tab to assign the aesthetic properties of each defined material. Colorsand other properties can be assigned to the faces, edges, and vertices of objects thatcontain the material so that they can be easily distinguished from other materials in theproject.

Physical ParametersThe Physical Parameters tab governs the definitions most commonly associated with


53

biological tissue. These definitions are thus necessary when performing biologicalcalculations. These values are computed automatically for tissues in Agilent Technologieshigh fidelity meshes.


54

Defining Circuit Components andExcitationsIn this section, you will learn about:

Adding passive and active circuit elements to your EMPro projectOther excitations you can include, such as plane waves and Gaussian beams

This section focuses on the steps associated with creating and defining valid componentsand excitations in an EMPro project.

Circuit Components are discrete components which may be used almost anywhere in theproblem space. Examples include resistors, capacitors, inductors, voltage sources, currentsources, switches and diodes. The definitions associated with these components aredefined within the Circuit Component Definition Editor.

There are three primary input excitation forms in EMPro. The first type of excitation can beapplied at one or more discrete locations with a voltage or current source. The second andthird excitations are applied externally in the form of incident plane waves, for scatteringcalculations, or Gaussian beams, for optical frequency calculations. The External ExcitationEditor governs both plane wave and Gaussian beam excitations.

The Waveform Editor governs the time variation of all three types of excitations. Awaveform may be defined as a pulse for broadband calculations, a sinusoidal source, or auser-defined waveform. A waveform definition must be applied to a defined excitation forit to be valid.

The following four sections outline the various aspects of creating and definingcomponents and excitations.

Component ToolsThis section will document the capabilities of Component Tools, found in the first drop-down menu in the Geometry workspace window. The Component Tools dialog enablesusers to add discrete components to the EMPro project. These discrete componentsinclude voltage sources, current sources, feeds, lumped resistors (R), capacitors (C),inductors (L), diodes, non-linear capacitors, and switches. Discrete sources (such asvoltage and current sources) are locations at which the electric field is modified by theaddition of some type of input waveform.

Each circuit component has its own specific set of definitions (i.e., spacial orientation,polarity, alignment), which are included in the Circuit Components dialog. Definitions suchas resistance, inductance, capacitance, etc. are specified in the Circuit ComponentDefinition Editor so that they can be reused for multiple components if necessary. Toaccess this editor, navigate to the Definitions:Circuit Component Definitions branch of theProject Tree and double-click on the component's Circuit Component Definition object.

The following illustration shows the various components available within the CircuitComponents dialog box.


55

Adding a new component with Component Tools

Adding a New ComponentOnce Circuit Components is visible in the first-drop down button of the Geometryworkspace window, clicking Create With New will prompt a drop-down menu to appear.This menu includes:

PassiveloadFeedSwitchNon-Linear CapacitorDiode

Selecting any of these components will prompt a similar series of tools to place thedesired component into the simulation space. The following sections will detail thedefinitions associated with these tools.

For more information about any one of these discrete sources, refer to CircuitComponent Definition Editor.

Circuit Components can also be added by right-clicking in the Project Tree as shownbelow.

Accessing the Component Tools dialog from the Project Tree

Connections Tab


56

When choosing one of the new New Circuit Components... menus, two tabs appear. In theConnections tab, you can easily specify the location of a component by using existingobject vertex points. One defines circuit components in solid space (connected to e.g. twoobjects) and not in the mesh space. The actual component, however, can only span onemesh cell and must adapt to changing mesh densities local to the component definition.This mesh cell is always in the middle of the feed as defined in solid space. The other cellsbetween the two endpoints defined in this tab are automatically filled with a PEC wire. Formany applications, as when feeding a monopole at its base, one needs to make sure thatthe feed starts exactly at the first mesh cell from the wire endpoint. This can be achievedby making the solid feed shorter or equal to one grid cell size and having one endpoint beattached to the wire vertex.

One can use the placement tool (white arrow icon) to easily pick off vertex points fromdifferent objects. Make sure to look at the bubble up help instructions. For example, if awire is attached to the top of a box object, you can click this icon, then hover over thewire until its edge and vertex points are highlighted, and then click "l" (lower-case L) totemporarily unlock it so that the bottom vertex can be snapped to.

Properties Tab

In the Properties tab, the name of the component, alignment, and polarity is defined. Thecomponent(s) can be aligned with either the X-, Y-, or Z-axes, by selecting X, Y, or Z,respectively. Otherwise, the default Auto selection will automatically align the componentbased on Endpoint 1 and Endpoint 2 defined in the Connections tab.

The check box, This Component Is A Port, can be selected to assign the component asa port. When this box is selected, EMPro will automatically add a Port sensor at thislocation.

The check box, Evenly Spaced In Orthogonal Directions, can be selected to enforceproper grid creation in the region around components. This is checked by default andshould be left on except in special cases.

NoteKeep in mind that a port that contains only passive components cannot be the active port. Lumpedreactive elements should not be used in the active port specification.

For more information about port sensors and the data that they collect, refer to SensorTools Port Sensors (using).

The figure below displays the Properties tab for editing a feed.

Editing component properties


57

If a component is added before it is defined in the Circuit Component Definition Editor, adefault definition will be created so that the component is valid. Simply double-click onthis default definition in the Definitions branch of the Project Tree to edit its properties.Likewise, if the component requires a waveform definition, a default definition will beadded to this branch.

Adding a Component Using an Existing DefinitionThe Create With drop-down menu functions similarly to the Create With New drop-downmenu described above, except that a pre-existing component definition is applied to thecomponent that is to be added. For this reason, the Create With menu is not active until acomponent definition has already been created within the Create With New dialog. It isthus a list all of the pre-existing component definitions that hav already been added to theproject. This menu makes it easy to add identical components to a project.

Circuit Component Definition EditorThe Circuit Component Definition Editor is used to define the parameters associated withdiscrete components. Components such as voltage sources, current sources, feeds,lumped resistors (R), capacitors (C), inductors (L), diodes, non-linear capacitors, andswitches are defined in this window.

It is important to recognize the purpose of the Circuit Components interface versus that ofthe Circuit Component Definition Editor. The former places the physical component intothe project (and creates an object in the Project Tree that represents the actualcomponent), while the latter creates a Circuit Component Definition object for theparameters of that component (or components) that can be used over and over again bydropping it onto multiple components.

The Circuit Component Definition Editor is accessed by double-clicking on any object foundin Definitions: Circuit Component Definitions branch of the Project Tree. The interfaceshown in the figure below is a sample setup for a feed. Note that the Type of componentdetermines the diagram in the editor.

The Circuit Component Definition Editor


58

Passive LoadWhen a component with no source is desired, the user may create it by selecting PassiveLoad. Passive components include lumped resistors (R), inductors (L), and capacitors (C).A passive component is one that does not add any energy to the problem space.

Passive loads may be combined at the same location with a single active component, suchas a voltage or a current source. Since passive lumped loads do not radiate energy, theymay be added to the calculation when either Planewave or Gaussianbeam excitations areselected.

For more information about these excitations, refer to the External Excitation Editor.

The RLC elements lumped at component locations can be combined with each other eitherin series or parallel. For passive components used with a voltage or current source, theRLC components will be in series with the voltage source or in parallel with the currentsource. The Series and Parallel choices refer only to how the RLC components arecombined with respect to each other. For the series load combination, all of the lumpedcircuit elements are in series and are located on one FDTD mesh edge. For a parallel loadcombination, all of the lumped circuit elements are in parallel and are located on oneFDTD mesh edge. The selected configuration is shown schematically in this window.

Below a sample setup for a Passive Load within the Circuit Component Definition Editor isshown.

Editing a load within the Circuit Component Definition Editor


59

Displacement of Current and Lumped Elements

There are physical limitations to the lumped element approximation in FDTD. Each FDTDmesh cell includes a volume of free space. That free space volume has capacitance, and adisplacement current flows through it. f a lumped RLC element is specified that results in ahigh impedance, the free space displacement current may be significant relative to thecurrent through the lumped elements in EMPro. In this situation, if the displacementcurrent is neglected, the result is non-physical and Maxwell's equations cannot besatisfied. So for lumped element calculations in EMPro, the displacement current isincluded even though the corresponding mesh cell capacitance is not indicated on the portcircuit schematic.

To determine the relative significance of the displacement current, one can calculate thecapacitance of the FDTD mesh cell. For a -directed component, the capacitance of the

cell is given by ,where,

, , and are the mesh cell dimensions and the permittivity of the material at

that cell location, usually that of free space, 8.854e-12 .

This is discussed on page 192 of the Kunz and Luebbers FDTD book.

As long as the impedance of this mesh cell capacitor is large compared with theimpedance of the RLC lumped element circuit, its inclusion in the EMPro calculation has anegligible effect. If the RLC lumped element circuit has an impedance comparable orlarger than the mesh cell capacitance, then the inclusion of this capacitance keeps theresult stable and physically correct.

For calculating impedances, a frequency corresponding to the sine wave frequency or thehighest frequency of interest within the spectrum of the excitation waveform should beused.


60

Similarly, a lumped inductor should not be less than the inductance of the FDTD cell, given

by , where is the permeability of that point in space and is the mesh cell lengthin the direction of the inductor.

FeedAn active source, or Feed, usually refers to an active component together with any passivecomponents that are at the same cell edge. An active component is a cell edge on whichthe electric field is modified by the addition of some type of input waveform. Voltage andcurrent sources are active components.

If a voltage or current source is included, the amplitude of the input waveform may bespecified as well as the polarity. The phase of the source may be specified if the inputwaveform is a sinusoid, otherwise this option will be unavailable.

NoteFor all voltage and current source specifications, the amplitudes specified are peak values, notRMS.

A time delay may also be specified for voltage and current sources when a Gaussian,Gaussian Derivative, or Modulated Gaussian waveform is used. The time delay is specifiedin timesteps and is applied to the beginning of the source waveform. For example, if atime delay of 200 timesteps is specified for a given source port, the input waveform atthat port will begin 200 timesteps later than the defined waveform. This functionalityallows for any number of Gaussian excitations to be applied at different times throughouta simulation.

The voltage across the FDTD mesh edge (electric field times edge length) and the FDTDmesh edge current include the effects of both the RLC components at that mesh edge and

the voltage/current source. A voltage source, , in series with a source resistance, isillustrated below. The voltage across the FDTD mesh edge is determined by the voltagesource in combination with the source resistance, so that the mesh edge voltage differsfrom the source voltage by the voltage drop across the source resistance.

Feed schematic, including FDTD mesh edge voltage, V, and current, I


61

Multiple Voltage and/or Current Sources

For calculations with multiple voltage and/or current sources, such as antenna arrays ormulti-port S-parameter calculations, multiple feeds may be specified. They are specified inthe Simulations workspace window before the calculation is run.

NoteFor more information about setting up an S-parameter simulation, refer to S-parameters SimulationSetup.

For antenna calculations, all feeds are normally excited. However, for S-parametercalculations only one feed can be excited for a particular EMPro calculation. For broadbandfeeds, each source function must have the same pulse width but may have differentamplitudes. Alternatively, they may instead all use the same user-supplied file of voltageversus time. The polarity can be adjusted by clicking the desired button. This may beuseful in controlling the sign of the phase terms in S-parameter calculations. For eachfeed, independent source resistances may be specified. For sinusoidal excitations, eachfeed can be specified with a different magnitude and phase.

Specifying the Source Resistance

For most EMPro calculations, active sources will consist of a voltage source with a seriessource resistance. This is the default configuration for feeds. The default value for thesource resistance is , since that is the most common reference. If an S-parametercalculation is made, the S-parameters will be in reference to the port resistance.

NoteS-parameters can be calculated for any reference impedance by changing the value of the resistance ateach port.

If the value of the source resistance is not determined by the desired S-parameterreference, then for most calculations the source resistance should be chosen to match thestructure being driven. This will strongly excite the structure and also dissipate resonancesmost efficiently. For example, for a microstrip with characteristic impedance, asource resistance of would typically be a good choice.

For antenna calculations, determining a good approximation to an actual antenna feed isnot always simple. Many antennas are fed with coaxial cable. For most EMPro calculationsthe coaxial cable itself need not be meshed, since it is only used to feed the antenna andthe fields inside of it are not of primary interest. The simplest approach to simulating thisis to locate a port in line with the center conductor of the coaxial cable where the cable isconnected to the antenna. The impedance calculated by EMPro will then be at this point inthe antenna.

NoteThe port resistance typically should be set equal to the characteristic impedance of the coaxial cable usedto feed the physical antenna. This will automatically refer S-parameters to this resistance value.

For broadband calculations, to reduce the number of timesteps needed for the transientsto dissipate, include a source resistance equal to the characteristic impedance of thecoaxial cable being approximated. This is similar to driving an actual circuit or antennausing a matched source.


62

For some situations, it is desirable to match a voltage or current source to a reactive load.In this situation the RLC capabilities of EMPro components can be utilized.

If the coaxial cable or other feed geometry is important to the calculation, EMPro may beused to mesh the cable itself. In this situation it is important to determine thecharacteristic impedance of the coaxial cable as meshed, and to match the port resistanceto the characteristic impedance.

Another important advantage to including a source resistance is reducing the number ofFDTD timesteps necessary for convergence of the electromagnetic calculation. This isespecially important for resonant devices, such as many antenna and microstrip circuits.With a "hard" source consisting of a voltage source without series resistance, a resonantmicrostrip antenna may require 64,000 timesteps for the transients to dissipate. Theaddition of a source resistance might reduce this to 4,000 timesteps. Similar time savingsmay be encountered for microwave circuits.

DiodeA Diode can be defined by specifying the following:

Saturation Current (!img283.png! )

Junction Potential (!img284.png! )

Depletion Capacitance at zero bias (!img285.png! )

the sum of Transit Time (!img286.png! ) for holes and electrons

Emission Coefficient (!img287.png! )

junction Grading Coefficient (!img288.png! )

the Forward Coefficient (!img289.png! ), which determines when the junction isheavily forward biased

For stability consideration, voltages applied to the diode should not exceed 15 volts.

NoteFurther information regarding the FDTD diode formulation may be found in Bibliography.

The figure below shows editing dialog for a Diode within the Circuit Component DefinitionEditor.

Editing a diode within the Circuit Component Definition Editor


63

Nonlinear CapacitorThe Non-Linear Capacitor contains parameters which correspond to the followingequation:

where:

is instantaneous cell edge capacitance

is instantaneous cell edge voltage

is static (low ) capacitance

is infinite capacitance

is voltage magnitude

is a scaling voltage

, and are coefficients

A non-linear capacitor may be combined with a parallel resistor.


64

The next figure shows editing dialog for a Non-Linear Capacitor within the CircuitComponent Definition Editor.

Editing a nonlinear capacitor within the Circuit Component Definition Editor

SwitchA special feature of EMPro is its ability to include timed and programmable Switchcomponents. This allows a change in the configuration of the geometry during acalculation. Any number of switches may be introduced into the geometry, with eachswitch being specified as a separate component. Switch states may be changed only onceduring a calculation.

The initial state of the switch, whether open or closed, may be specified as well as thetimestep where the switching action begins. The switching action is spread over a variablenumber of timesteps to reduce switching transients. The switch transition should on theorder of 60 timesteps.

To define the properties of the switch, select Open or Closed to define the switch's initialstate in the Circuit Component Definition Editor. Click |+| to add a transition and definethe Start Time, Duration, and Transition type of the switch by double-clicking on thedefault values provided in the chart in the editor.

The following figure shows editing dialog for a multiple transition Switch within the CircuitComponent Definition Editor.

Editing a switch within the Circuit Component Definition Editor


65

A switch may be programmed to allow multiple open/close transitions during a simulationby adding subsequent entries to this initial definition. A switch transition consists of thetimestep at which the switch will activate and the duration (in timesteps) that theGaussian switching function will be applied. Simply click the button found above thedefinitions chart to add one or more switch transitions and define the Start Time, andDuration that the transition will occur. The transition type will be automatically generatedbased on the start time of each transition. Furthermore, the Sort button will automaticallysort the transition from the earliest to latest occurring state.

During the switch transition, the electric field at the switch is changed from the openswitch value to zero when closed (or vice versa) following a Gaussian function. Thismethod provides stability for the EMPro calculation. In order to apply the Gaussianswitching function, the value of the electric field at the switch location is first calculated asif the switch were not present. This value is then multiplied by the Gaussian function withthe appropriate argument based on the time since the switch state was changed. For aclosed switch, the multiplier is zero, for an open switch it is one, and for intermediatetimes during the transition the value is that for the normalized Gaussian function. For thisreason, the function of the switch depends on the material that exists at the meshlocation. Usually this is free space, but it might also be dielectric.

The use of timed switches is for situations where only the actual transient results directlycalculated by EMPro are of interest. The introduction of the switching action violates theassumptions of linear system theory, so that applying Fourier transformations to thetransient results produced in an EMPro calculation with one or more switches whichchange state will not result in valid impedances, S-parameters, steady-state far-zonefields, or other results involving Fourier transformation, even if the transient results finallydecay to zero. Transient far-zone fields will be valid, however.

External Excitation EditorThe External Excitation Editor window is used to define an External Excitation to beapplied to an FDTD simulation. There are two types of external excitations, which aredetailed in the following sections: Planewave, and _Gaussianbeam.


66

The following figure shows the External Excitation Editor. Note that a Waveform must beapplied for the external excitation to be valid. In this interface, a plane wave excitation isdefined.

The External Excitation Editor

To access the editor, double-click on an existing object in the External Excitations branchof the Project Tree. If no external excitations are defined, right-click on this branch andselect New External Excitation.

Plane WaveThe first external excitation definition form available within the External Excitation Editoris the incident plane wave. The plane wave parameters are enabled when the source typeis set to Planewave in the Simulations workspace window.

NoteFor more on how to set up a simulation, refer to Creating a New Simulation.

A Plane Wave source is assumed to be infinitely far away so that the constant fieldsurfaces are planar and normal to the direction of propagation. Calculations of radar crosssection or scattering may be performed using this source. All calculations with the incidentplane wave input are performed in scattered-field, but total-field values may also becomputed.

The incident direction defined by Incident Phi and Incident Theta must be specified byangle. Incident Phi is measured from the X-axis to the Y-axis while Incident Theta ismeasured from the Z-axis to the XY plane. The incident waveform may be phi- or theta-polarized. The electric field values in the X , Y , and Z directions are displayed under theIncident Amplitudes heading in this window and are updated each time the polarization orincident direction is modified.

Selecting Scattered-Field Versus Total-Field Plane Wave


67

Total-field values may be saved and displayed since they can be determined by adding thespecified incident field with the computed scattered field. Scattered fields cannot besampled inside the total-field region and vice versa. Calculations of radar cross section orscattering are based on fields inside the scattered-field region.

The interface between the total-field and scattered-field regions must be free space. Fornon-periodic boundaries, the six sides of the total-field region interface are located eightcells into the FDTD mesh. When periodic boundaries are specified, certain sides for theinterface may be turned off and the total-field region may extend to the boundary. Thesedefinitions are specified in the Simulation workspace window.

NoteFor a discussion of these definitions, refer to Specifying Field Formulation.

In most cases, total-field plane wave is preferable to scattered-field plane wave. Below,two examples of an electric field propagating (from right to left) through a shelledgeometry of Perfect Electric Conductor (PEC) material. The image on the left represents ascattered plane wave source, while the image on the right shows a total-field plane wavesource. Since the box is PEC material, the electric field within the bounds of the boxshould be zero, and therefore, the total-field plane wave source is more valid.

The E-field of a scattered-field (left) vs. total-field (right) plane wave source

In some cases, however, the total-field plane wave source will also create inconsistenciesin the calculations.

The next figure shows the resultant electric field when an object crosses the total-field/scattered-field interface (intersection represented by the white arrows). This willproduce incorrect results. To fix this problem, the interface may be moved so that it issufficient distance from the geometry. However, to move the interface, the bounding boxof the project must be increased which will also increase the memory requirements of theproject. Alternatively, a scattered-field plane wave can be used with the understandingthat the region within the shelled geometry is incorrect.

Incorrect results due to an object crossing the interface


68

The following figure demonstrates a similar problem where the interface is too close to thebounds of the object. In this case the fields fall within the "shadow" region of the objectand are not calculated correctly. The image on the left shows the effect of the shadowregion early in the field sequence. The image on the right shows the incorrect field valueslater in the sequence at the interface between the total-field/scattered-field interface. Theinterface, like in the previous example, must be adjusted if total-field plane wave is to beused.

Incorrect results when a field propagates within the Shadow Region of the geometry


69

Gaussian BeamThe second external excitation definition available within the External Excitation Editor isthe Gaussian Beam source. The focused Gaussian beam is characterized by an incidentelectric field that has a two-dimensional, radially-symmetric Gaussian distribution inplanes normal to the incident direction. It converges to maximum intensity at the focuspoint. As with the plane wave source, all calculations with a Gaussian beam source areperformed in scattered-field, though total-field values may also be saved and displayedalso. Unlike the plane wave and discrete sources, the Gaussian beam source requires thatthe source waveform be sinusoidal.

NoteExamples where this type of source is useful include structures used at optical frequencies and situationswhere it is desired to illuminate only a portion of the geometry.

The Gaussian beam is specified by Incident Direction, Polarization, Waveform andwaveform Amplitude, as described for the plane wave source above, and by focal point (

Focus X, Focus Y, Focus Z) and Focus Radius. The Focus Radius is the beam radius, ,at the focal point, in the plane normal to the direction of travel and containing the focus,

at which the field strength drops to -8.686 dB of its maximum value.

In the next illustration we see the dialog for defining a Gaussian beam in the ExternalExcitation Editor. To use this excitation as the source for a calculation, it must be specifiedwithin the Simulation workspace window.

NoteFor more information about the Simulations workspace window, refer to Creating a New Simulation.

Defining a Gaussian Beam source in the External Excitation Editor


70

In other planes normal to the direction of travel, is the radius at which the field

strength drops to of its maximum value in that plane and is given by:

where:

is the free space wavelength and, for simplicity, the direction of travel is assumed to beparallel to the Z-axis and the focus in the plane.

Since passive lumped loads do not radiate energy, they may be added to the calculationwhen either Planewave or Gaussian beam excitations are selected.

NoteAll active ports will be set to passive lumped loads when either of these excitations areselected.

Waveform EditorThe Waveform Editor is used to edit waveforms that can be used in conjunction with anExternal Excitation or a Feed to inject energy into the space for an FDTD simulation. Whenused in a simulation, a waveform's field value at each timestep is applied to the field inthe space as part of the field update calculation.

To create a Waveform, right-click on the Definitions:Waveforms branch of the Project Treeand select New Waveform Definition. Double-clicking on a Waveform under the same


71

branch will open that waveform for editing.

There are six types of waveforms available:

BroadbandGaussianGaussian DerivativeModulated GaussianRamped SinusoidUser-Defined

The choice of waveform should be based on the desired output, as some waveforms aremore appropriate than others for specific problems. For broadband calculations, theBroadband or one of the Gaussian-type waveforms should be used, since they will injectenergy into the space in a wide band of frequencies. The Ramped Sinusoid may be chosenfor cases where steady-state results are desired at a single frequency. The User Definedwaveform can be used when none of the other choices meet the requirements of thesimulation.

BroadbandWhen broadband results are desired, it is almost always the case that the Broadbandwaveform type should be used. This waveform provides a Gaussian pulse with the largestfrequency content possible for the specific FDTD space when Excite All PossibleFrequencies is selected (see the following figure). Alternatively, the user can clamp theupper end of the frequency range by selecting Excite Up To A Maximum Frequency andchoosing a frequency response magnitude and corresponding frequency. In this case, notethat the waveform frequency response is truncated in the simulation to the maximumallowed, even if a wider frequency response is specified in this editor. The figure belowshows a Broadband waveform in the Waveform Editor.

NoteFor more information on waveform frequencies, refer to the Waveform Editor.

Defining a Broadband waveform in the Waveform Editor


72

GaussianThe Gaussian waveform provides broadband input and is also suitable for use whenbroadband results are desired. The width of the pulse is user-specified.The following figure shows a Gaussian pulse in the Waveform Editor.

NoteFor more information on specifying the width of the pulse, refer to Notes on Choosing WaveformParameters.

Defining a Gaussian pulse in the Waveform Editor


73

Since the Gaussian pulse has a non-zero average value, it should not be used for a Feedwhen there is a closed path (loop) of perfect conductor connected to the Feed unless thefeed also contains resistance. This is because the Gaussian has a DC (zero-frequency)component which starts a steady current flowing in the loop that will never decay throughloss or radiation. The symptom of this will be a source current that has an average valuenot equal to zero. If this occurs, the Gaussian Derivative or Modulated Gaussian pulsescan be used or a non-zero source resistance specified for the Feed.

Gaussian DerivativeThe Gaussian Derivative is nearly identical to the Gaussian except that it has a zeroaverage value and thus the DC component is removed. For the same simulation, the pulsewidth of a Gaussian Derivative should be set somewhat greater than what would be set fora Gaussian since the Gaussian Derivative will have a wider frequency spectrum for thesame pulse width. The following figure shows a Gaussian Derivative in the WaveformEditor.

NoteFor more information on pulse width, refer to Notes on Choosing Waveform Parameters.

Defining a Gaussian Derivative waveform


74

Modulated GaussianThe Modulated Gaussian should be used only when a specific frequency range is desired.This is useful in structures where low frequencies could excite non-radiating modes whichcould resonate and invalidate results. This waveform is a sinusoid with a Gaussianenvelope, with the sinusoid centered in the envelope so that the average of the pulse iszero.

In the figure below we see a Modulated Gaussian in the Waveform Editor.

Defining a Modulated Gaussian waveform


75

The Pulse Width for the Modulated Gaussian may be adjusted to enclose a specificfrequency range. This is useful, for example, in waveguide simulations so that onlyfrequencies in the band of single-mode operation are excited. It is also useful when band-limited devices are being simulated. For example, a broadband antenna, such as a spiral,may be designed for a specific frequency range. Exciting the antenna at frequenciesoutside this range may greatly increase the simulation time needed for convergence sincethe out-of-band energy cannot readily radiate or be otherwise dissipated by the antennastructure.

NoteFor more information on pulse width, refer to Notes on Choosing Waveform Parameters.

Ramped SinusoidThe Ramped Sinusoid is useful when only one frequency is of interest. When thiswaveform is used, a single frequency simulation is performed that provides additionalresult types over broadband simulations. The "ramped" part of the waveform is used andautomatically configured to avoid introducing energy at any frequency other than that ofthe sinusoid into the simulation.

Since the Ramped Sinusoid is different than the other waveforms, it requires some specialconsiderations. It is recommended that the length of a simulation be at least five cycles ofthe sinusoid when not using automatic convergence. In some cases, such as extremelylow frequencies, running five cycles may not be practical, in which case even just afraction of a cycle may be used.


76

The figure below shows a Ramped Sinusoid in the Waveform Editor.

Defining a Ramped Sinusoid waveform

User DefinedAn arbitrary source voltage vs. time may be specified using the User Defined waveformtype. A time record of the waveform is imported from a user-defined text file and useddirectly as the discretized waveform for the simulation. The file format is as follows:

The first line must be an integer, which is the number of timesteps in the waveform.The first data point corresponds to the first timestep, and zero will be used for thewaveform for timesteps beyond the provided data (the user-defined waveform iszero-padded).The remainder of the file contains two floating-point numbers on each line, one linefor each successive timestep. The values of these numbers depend upon whether thewaveform is being used for a voltage source or incident plane wave (it is incumbentupon the user to fulfill this requirement; EMPro does not validate the data).

For a voltage source, the two numbers are the normalized voltage and its timederivative at the source at that timestep.For an incident plane wave, the two numbers are the normalized electric fieldmagnitude (in Volts/meter) and the time derivative of the electric field.

The frequency response of the User Defined waveform is computed from the input time


77

data. The next figure shows a User Defined waveform in the Waveform Editor.

Defining a User Defined waveform

Notes on Choosing Waveform ParametersFor obtaining broadband results, the usual choice for a waveform should be the Broadbandwaveform. However, for certain situations, a specific pulse type may be desired. For thesecases, when choosing the pulse width and/or frequency for a waveform, the constraints ofthe FDTD method must be kept in mind. The time rate of change of a pulse or the sinewave frequency must be low enough so that the waveform is accurately sampled. Thisresults in the frequency being constrained by:

where:

is the speed of light, and

is the largest grid edge length in the space.

This prevents the waveform from introducing energy at frequencies too high for the FDTD


78

method to produce accurate results. The pulse width should also not be so large (and thusthe frequency content so low) that the calculation is not excited at frequencies for whichaccurate and useful results might be obtained. There must be sufficient energy in thepulse in the frequency range of interest for results to be above the numerical noise. Forsimulations with very small cells compared to the shortest wavelength of interest, pulsewidths may be set much larger than the default size to reduce the number of timestepsneeded for convergence and to increase stability.

If dielectric materials are present in the FDTD space, wavelengths are reduced insidethose materials since the velocity of propagation is less than the speed of light in freespace. A reasonable rule of thumb for lossless or low-loss dielectrics is that the maximumfrequency for the spectrum of the pulse should be reduced by the square root of therelative permittivity (!img212.png! ) times the relative permeability (!img312.png! ), orequivalently, the_Pulse Width_, , should be increased by this factor:

where :

represents the corrected pulse width, and

represents the uncorrected pulse width.

For more lossy dielectrics or conductors, the frequency and pulse width should be adjustedproportionally to the change in wavelength in the material relative to the free space. Ofcourse, the maximum frequency for reliable results is also reduced as the frequencyspectrum of the excitation pulse is reduced.

For example, suppose part of the simulation space is free space and part is a low-lossdielectric with a relative permittivity of 4.0. Further suppose the cell size is 1 cm. At10 cells per wavelength, one would expect reasonable results up to 3 GHz with a Gaussianpulse 32 timesteps wide (for a uniform grid). But since part of the space has a dielectricmaterial, with the same cell size the Gaussian Pulse Width should be doubled to64 timesteps to reduce the frequency spectrum bandwidth. Correspondingly, themaximum frequency for reliable results would be decreased from 3 GHz ;to 1.5 GHz.

When using the Broadband waveform, EMPro chooses (in the case of automatic pulsewidth selection) or limits the pulse width (in the case of user-specified frequency content)of the Gaussian pulse based on the restrictions above. When using a waveform with user-specified pulse widths, it checks to make sure the frequency limit, as described above, isnot exceeded. If it is, an error is given when the simulation is created.


79

Defining the Grid and Creating a MeshIn this section, you will learn about:

Tools available for you to specify the properties of the EMPro grid.How to set localized grid properties, right down to individual objects or single cells.How to tell EMPro which objects to include in the mesh and in what order to considerthem.How to adjust the way touching objects are meshed.

The EMPro grid provides the main interface between a dimension-based geometry and thefinalized FDTD mesh space. EMPro provides a several tools for creating and customizingan effective grid so that accurate results can be retrieved without allocating excessivememory and calculation time. This section details this process.

NoteRefer to Grid Appendix for a thorough discussion of grid theory, concepts, and recommendations withinEMPro.

General grid definitions are assigned to the entire grid within the Grid Tools dialog of theGeometry workspace window. Here definitions such as target cell size, boundingparameters, and limits are defined. Customizable grid definitions such as fixed points andgrid regions can also be added here.

The Gridding Properties Editor governs grid definitions that are applied to individualobjects. This editor is useful for objects whose characteristics are not adequatelyconsidered by the general grid definitions set within the Grid Tools editing tabs.

Finally, the Meshing Properties Editor governs how each object within the boundaries ofthe defined grid will be meshed. By default, every object in the project is meshed withuniform priority, but these settings can be adjusted within this editor so that importantobjects are meshed more carefully than other less-important objects.

The following sections explain how to enter the grid properties and specify the various gridelements using these editors.

Grid ToolsThere are two ways to access the Grid Tools dialog. With the geometry workspace windowopen, either select Grid Tools from the drop-down menu in the window, or double-click onthe FDTD: Grid branch of the Project Tree. Once the dialog is open, press the Edit Gridbutton to define the grid parameters for the project.

The tabs located within grid tools follow the step-by-step editing process that is used todefine the grid. The first tab, Size, specifies general definitions that are applied to thewhole grid, whereas the second and third tabs, Fixed Points and Grid Regions, enable theuser to customize the grid by manually placing gridlines and changing the cellcharacteristics of different regions in the mesh.


80

NoteFor a thorough discussion of fixed points and grid regions,refer to Grid Regions vs. Fixed Points.

The fourth tab, Limits, applies constraints to the grid so that memory requirements andadjacent cell size ratios do not exceed a reasonable value. Since grid definitions largelyinfluence the memory requirement of a particular calculation, two bars are always visiblein order to monitor how much memory the intended grid definitions will utilize. TheCurrent bar shows how much memory the last applied mesh required, while the New barshows how much memory a new mesh requires before changes are applied to the project.The Limits tab also enables you to specify the smallest desired cell size. If a grid definitionrequires a cell smaller than the smallest desired size, the grid becomes invalid and awarning message will appear.

The fifth tab, Info, provides data about both the current and the proposed grid. (The twosets of values will be the same if any proposed changes have been applied.) The grid'sdimensions, upper and lower boundaries, cell counts, and minimum and maximum edgelengths are shown. A value is shown for each axis, X, Y and Z.

Size Tab

The Size tab enables users to define the cell size and the placement of the grid at theboundaries of the project. There are two options for defining the outer boundaries. Thefirst option is to specify the padding around the project's lower and upper boundaries interms of discrete numbers of cells. The second option is to specifying a grid bounding boxby its physical coordinates, independent of the geometry. Each option is detailed below.

Specify Padding

The next figure displays the available options for specifying the padding of the outerboundaries of an EMPro cell.

Defining a project's boundaries by cell padding within the Size tab

Selecting Specify Padding enables you to add fixed points for the bounds of a geometry.The actual padding cells will begin at those fixed points and proceed outward. Within thegeometry, the area between the fixed points will be divided into grid cells. If there are nomanual grid regions, manual fixed points, part grid regions, or part-associated automaticfixed points in the project, the area between the fixed points bounding the geometry willbe divided into cells with even spacing. Each evenly-spaced cell will have an edge length


81

no greater than the target cell size.

NoteTo have EMPro select the "fixed points" automatically, choose Use Automatic Fixed Points, which is foundin Project Tree under the Parts section of a geometry, in the Gridding Properties Editor.

The Base Cell Sizes dialog allows entry of the target cell size when generating the EMPromesh in areas that are not part of a grid region. Note that this is the target cell size, andmay not reflect the actual cell size in the mesh. The actual cell size may be slightly lowerdepending on the constraints placed on the mesh, but it will never be higher, ensuringthat the desired level of detail is not compromised. When there are two fixed points orgrid region bounds of any type, the region between them will be divided into a number ofcells. The number of cells that is required depends upon the target cell size. No cells largerthan the target will be generated. The target cell size may not divide evenly into thedistance between the fixed points or grid region bounds, The gridding process prefers togenerate cells of a constant size. First the minimum number of cells required to span thedistance is calculated and then a cell size, less than or equal to the target size, iscalculated so that the distance is spanned exactly.

NoteFor more details and for exceptions to the general rules, refer to the Grid Appendix.

The Free Space Padding (In Cells) dialog enables users to define the extent of the paddingspace around the geometry. The extent is specified as a number of cells, each of which istaken to be the target cell size. In simple projects this means that the number of cellsspecified will be added as freespace around the geometry. In more complex projects, gridregions close to the boundary may cause transition regions containing smaller cells tointrude into the freespace padding. The freespace padding's extent will remain the same,whether it is made up of the specified number of cells each exactly the target cell size, orwhether the padding extent contains some number of smaller-sized transition region cells.

There are also two checkboxes available in the Size Options drop-down list: IncludeUnmeshed Objects and Include Circuit Components. Selecting either of these options willensure that any unmeshed object and/or circuit component is included within the paddedregion. Otherwise, only meshed objects will be used to determine the size of the paddedregion. These options may also affect the size of the region bounded by free spacepadding.

Specify Bounds

The following figure displays the available options for defining the outer bounding box.

Defining a project's boundaries by a bounding box within the Size tab


82

The Base Cell Sizes dialog box is analogous to the Base Cell Sizes definition under SpecifyPadding.

There are some instances when defining a Grid Bounding Box rather than a padding offree space cells may be advantageous in placing the outer limits of the grid. The gridbounding box, unlike cell padding, is independent of the geometry, so it is possible todefine a grid bounding box that does not include the entire geometry.

Fixed Points Tab

Fixed points are essentially gridlines placed at discrete locations in the mesh, andtherefore control the placement of cell boundaries in the mesh. They are used to ensurethat important parts of the geometry are considered at cell edges which cannot becontrolled by uniform meshing. For example, it may be useful to place a fixed point at theedge of a Parts object in the geometry so that its edge does not fall between cell edges.

The Fixed Points tab stores the definitions associated with the placement of fixed points inthe geometry. To add a fixed point to the mesh, select the Add button and define theFixed Point Properties by typing in the physical coordinates of interest. Note that any ofthe three principle directions may be checked or unchecked (in the Fixed checkbox)depending on how many points are to be defined. For example, if only the X-coordinate isdefined, a fixed point will be added that intersects at that location on the X-axis, whichdefines a plane in YZ. Defining fixed points in all three principle directions will result inthree fixed points, intersecting at the defined point.

The figure below displays the dialog associated with defining fixed points. In this instance,only X and Y are checked, so two fixed points spanning the X- and Y-planes will be addedin the X (at X = -0.2 m) and Y (at Y = -0.2 m) plane, respectively.


83

NoteA fixed point can also be defined by checking the X, Y, and/or Z coordinate check-boxes of interest andclicking on the desired location of the plane in the simulation space with the Selection tool.

Grid Regions Tab

A Grid Region is a bound region in the mesh that is assigned a cell size different from thecell size assigned in the Size tab. Grid regions may be desirable in locations that requirefiner meshing and thus require smaller cells. It is also advantageous to use them in areasthat are not important to the EMPro calculation by defining them with larger cells to savememory and calculation time. Simply define the Region Bounds and the Cell Sizes in eachdirection of interest, and a region will be placed on the grid.

The figure below displays the dialog associated with defining grid regions. In this example,the cells within the particular grid region are defined to be 0.001 m. Automatic fixedpoints will be merged to be spaced 60% of this distance as defined in the Merge boxes.

NoteFor more information about fixed point merging, refer to Grid Appendix.

Limits Tab

The Limits tab contains several required and several optional limits used to restrictcharacteristics of the grid.

The Maximum Cell Step Factor definition restricts the rate at which the cell size mayincrease between two adjacent cells. For example, if a cell is 1 mm long and the maximumcell step factor is set to 2, the next cell cannot exceed 2 mm, because that would causethe cell ratio between these two cells to exceed 2. Therefore, if a grid region has cells of1 mm, and the default global grid target is 5 mm, additional cells are needed, forming atransition region between the grid region and the Global grid, since 1 mm:5 mm exceedsthe 1:2 ratio.

NoteThe ratio of 1:2 is used in this example because it is the largest ratio recommended in Chapter 11, Section2 of the Taflov and Hagness in Bibliography text.

The second required limit, Minimum Cell Size, represents the global minimum cell sizeallowed in the project. No cell in the project can fall below this limit or the project will notbe valid. There are other Minimum Cell Size definitions within the Grid Tools and Gridding


84

Properties Editor, but they must be greater or equal to this global definition to beconsidered a valid definition.

The two optional limits aid in restricting the size of the EMPro project so that thecalculation time and memory requirements do not exceed a defined limit. These limits canbe applied to the project by restricting the number of Maximum Cells that are containedwithin the mesh. If this definition is applied to a project and exceeded, an error messagewill appear to indicate this.

The following figure displays the Limits tab.

Gridding Properties EditorThe Gridding Properties Editor window is used to define grid regions and fixed pointsassociated with an individual object. To open the Gridding Properties Editor window, right-click on the appropriate object in the Parts branch of the Project Tree and select GriddingProperties as seen below.

The Use Automatic Grid Regions and Use Automatic Fixed Points checkboxes enable eitherof these definitions when they are selected. The figure below shows the GriddingProperties Editor window with both grid regions and fixed points enabled. The followingsections detail the definitions associated with grid regions and fixed points.


85

Use Automatic Grid Regions

A grid region is a region assigned a specific cell size within a specified lower and upperboundary. A custom cell size can be applied in this region to any or all of the threeprincipal directions. They are desirable in instances in which the Cell Sizes defined to theentire grid under Grid Tools: Grid Regions is not sufficient for a particular object.

Grid regions can be added to the grid within Grid Tools by defining each region's physicalbounds, or a grid region can be applied to a single object within this editor. In this case,the grid region will be applied everywhere within the object (plus any extended regiondefined in the Boundary Extensions definition).

NoteFor more information on defining a region's physical bounds, refer to the Grid Regions Tab.

Cell Sizes

The Cell Sizes definition refers to the maximum target cell size that may occur betweenthe lower and upper boundaries. The actual cell size between these boundaries may beslightly lower depending on the location of these boundaries. For instance, if a Cell Size isdefined to be 1 m in the X-direction and the upper and lower boundaries are defined at X = 1 m and X = 4.9 m, respectively, the actual cell size would be 0.975 m since thedistance between the upper and lower boundaries would not permit four 1 m cells within3.9 m. The actual cell size will never be adjusted to a value higher than the Cell Sizes


86

definition, however.

NoteFor more information on choosing an appropriate cell size, refer to Choosing an Appropriate CellSize.

Boundary Extensions

The upper and lower boundaries of the grid region usually occur at the edges of theobject, but in the case that these bounds are not sufficient, they can be moved inward oroutward in any X, Y, and/or Z direction by defining values in the Lower BoundaryExtensions and Upper Boundary Extensions dialog boxes. If a positive distance is defined,the boundary will be shifted at that distance away from the object and conversely, if anegative distance is defined, the boundary will be shifted into the object.

Use Automatic Fixed Points

Fixed points are analogous to gridlines whose locations are explicitly defined to controlwhere the edges of meshed cells fall in the geometry. Automatic fixed points are giventhat name because their positions are automatically determined by examining thegeometry of the part with which the automatic fixed points are associated. Cell sizes areadjusted to flow as evenly as possible between the fixed points while never exceeding thetarget cell size at the location of each fixed point.

NoteFor more on the Limits tab Limits Tab. For a discussion of how fixed planes vary from grid regions, refer to Appendix Grid Regions vs. FixedPoints.

Fixed points can be placed anywhere in the grid by specifying their location within theFixed Points tab of the Grid Tools dialog, or they can be added within this editor to aparticular object based on the object's geometry.

NoteFor more on the Fixed Points tab,refer to the Fixed Points Tab section.

The following settings control where fixed planes are placed in reference to the object'sgeometry.

Lower and Upper Boundaries

Places fixed point(s) at the lower or upper boundary of the object in the directionspecified.

Wire Endpoints

Places fixed point(s) at any discontinuity existing in the object. Refer to the followingfigure for an example of a fixed point placed at an edge.

Edge Corners


87

Places fixed point(s) anywhere that there is a discontinuity in the derivative of a object'sedge (i.e., points that do not have tangents).

The figure below also shows the fixed points placed at wire endpoints (yellow points) andedge corners (green points) that occur within the simple wire body (shown in white).

NoteNotice how the cell sizes vary to accommodate the placement of the four fixed planes (shown inred).

Fixed points placed at wire endpoints and edge corners

Endpoints of axis-aligned wires

Places fixed point(s) along any axis-aligned straight edges that exist in the objectgeometry.

Below, fixed planes (in red) are placed at axis-aligned straight edges.

NoteNotice that straight edges that are not axis-aligned do not have fixed planes.

Fixed points placed at axis-aligned straight edges


88

Edge Loops Bounding Box

Places fixed point(s) at the edges of loops or closed circuit of edges. Any loop that smallerthan the Minimum Size will be ignored. Thus, to consider all loops that exist in the object'sgeometry, simply define the Minimum Size as 0.

In the figure below 12 fixed points placed at the edges of loop boundaries.

NoteNotice that the Minimum Size has been set to 0 so all loops are considered. If the minimum size waslarger than one or more loops, no fixed points would be placed at its edges.

Fixed points placed at edges of loops (minimum size = 0)


89

Open Ellipse Center Points

Places fixed point(s) at the center of open ellipses.

The figure below shows four fixed planes placed at the center of four filleted edges, whichare treated as open ellipses.

Fixed points placed at the center of open ellipses


90

Closed Ellipse Center Points

Places fixed point(s) at the center of closed ellipses.

Meshing Properties EditorThe Meshing Properties Editor is used to adjust the properties of the mesh for individualobjects. This gives the user greater control over the meshing process, when necessary, inorder to create a more accurate mesh. In most cases, the default settings are suffcient. Amesh is generated automatically when calculation files are written, or when griddefinitions are applied within the Grid Tools dialog and the project is viewed with the MeshView button in the View Tools menu.

The meshing properties editor displays different setting configurations depending on thetype of part being modified. The following figure shows the Meshing Properties Editor for amodel part, with the menu of meshing modes displayed.

To open the Meshing Properties Editor window, right-click on the Parts: Modelnamebranch of the Project Tree and select Meshing Properties.

NoteWhen editing the grid, a preview of the grid is created in the Mesh View mode, but it is not actuallymeshed until the grid definitions have been applied.

When Automatic Remeshing is turned on, the mesh will be updated automatically, so thatthe cutplanes visible within Mesh View will be current. If this feature is not turned on, theicon will indicate the mesh is out-of-date. In this case, press the Automatic Remeshingbutton and select Remesh Now to generate the current version of the mesh.

The meshing properties editor displays different setting configurations depending on thetype of part being modified. The figure below shows the Meshing Properties Editor for amodel part, with the menu of meshing modes displayed.


91

To open the Meshing Properties Editor window, right-click on the Parts: Modelname branchof the Project Tree and select Meshing Properties.

Meshing Modes

There are five available meshing modes in EMPro. Each is described below. These can bespecified in the Meshing Properties Editor or in the Project Tree by right-clicking on theobject, as seen below.

Assigning meshing modes within the Project Tree

Meshing Disabled: Select Meshing Disabled to disable meshing for objects that donot need to be considered during meshing operations. Selecting this option willreduce memory usage.


92

Electrical Meshing: Select Electrical Meshing to include the object in the electricmesh only.Magnetic Meshing: Select Magnetic Meshing to include the object in the magneticmesh only.Dual Meshing: Select Dual Meshing to force the object to be included in bothelectrical and magnetic meshes.Automatic Meshing: The Automatic Meshing mode is selected by default, indicatingthat EMPro will determine whether or not to include the magnetic grid based on thepresence of magnetic materials in the geometry. Under most circumstances it is bestto keep this default, since the engine will automatically apply the most memory-efficient meshing option that also considers the electrical and magnetic properties ofthe object.

Meshing OrderThe Meshing Order setting specifies the level of priority that each object has. If objectsoverlap in a particular region of space, the object with a higher meshing priority willoverwrite the other objects. Each object is assigned a default priority of 50. This should bechanged to a higher or lower value depending on whether the object should be given moreor less consideration, respectively, when meshed. Choosing Set Priority prompts the userto enter a numerical priority definition.

Additionally, where numerical priority values are equal, priority can be specified within theProject Tree, as displayed below. Moving the selected Parts object priority UP or DOWN,increases or decreases its importance in the meshing calculation, respectively. Likewise,selecting Move To Top assigns that object the highest priority of all of the objects in theproject; Move To Bottom will assign it the lowest priority.

Assigning meshing priority in the Project Tree


93

Solid Meshing Options

Under most circumstances, the solid meshing options are not necessary. These optionsmay be advantageous in projects that have small features that are not adequatelyconsidered by default meshing operations. The following figure shows the solid meshingoptions in the Meshing Properties Editor.

The Meshing Properties Editor for a solid model part

Only Mesh Object Surface


94

The Only Mesh Object Surface option meshes the shell of the physical object rather thanits entire volume.

Invert The Direction Of Surface Rounding

If the edge of an object falls in the center of a cell rather than at a cell edge, the defaultbehavior of EMPro is to round the object away from the center of the object. SelectingInvert The Direction Of Surface Rounding will invert this behavior and round the edgeinward, towards the center of the object. This usually is not necessary but may be usefulin cases where an object is very close to another. In this instance, rounding a surfaceoutward can cause the objects' edges to touch and result in a short circuit that may causeinconsistent results.

Aggressive Surface Meshing

Aggressive Surface Meshing should only be enabled for an object that is very smallcompared to the projects' cell size. This option will create a more detailed mesh aroundsmaller features to ensure that they are surrounded with cell edges. This would benecessary for small feature whose electrical conductivity may be compromised withoutdetailed meshing.

Volume Meshing Options

Several additional settings are available when assigning mesh properties for volumetricparts, such as voxels. The figure below the volume meshing options in the MeshingProperties Editor are shown.

Volume Meshing Method

There are two Volume Meshing Method algorithms available for converting the volumetricdata to cell edge data. Both methods use an intersection method to determine how muchof a given volume is contained inside of a cell edge. Therefore, it is applied to everyvolume that intersects the cell edge. The data is combined to measure the percentage thateach material fills in each cell edge. These percentage fill values are in increments of ahalf of a percentage point. This information is then used based on which algorithm isselected.


95

The first method, referred to as the Dominate Material meshing algorithm, evaluatesthe list of percentage fill values, and then chooses the material that has the largestpercent fill to apply as the final material for that cell edge.The second method, referred to as the Averaged Material meshing algorithm, usesthe percent fill values for each cell edge to generate an "averaged material." Theseaveraged materials are maintained in a separate list from the original materials of thegeometry.

Degradation Level

The Degradation Level is available in the Averaged volume meshing method. It dictateshow non-averaged materials intersecting the volume of the cell edge are to be groupedand considered for averaging, by defining the size of a set of bins that represent apercentage of the total cell edge volume. For example, a degradation level of 4 subdividesthe total volume (100%) by 4, meaning each bin represents 25% of the volume.Therefore, for consideration in the averaged material, a non-averaged material must fill atleast half of the bin, or 12.5% of the total volume. By defining a course resolution formaterial consideration we are able to condense the number of averaged materials,keeping memory and calculation speed reasonable while providing a user-defined level ofaccuracy.

Volume Calculation Method

The Volume Calculation Method determines the accuracy of the rendered volume. Itdirectly relates to the Degradation Level, as it determines the amount of material volumethat fills a cell. Approximate is sufficient for most calculations, and runs most efficiently.Exact is a better choice when it's important to resolve structures that are changing rapidlywithin the size of a cell. By definition, it is more accurate but also requires more memory.

Touching Objects

The FDTD mesh is a discretization of a continuous 3-D space into cells of user-specifiedsize. The process of transferring these geometric representations into this discretizedspace (meshing) can lead to unintended artifacts. One common situation that arises isthat two objects in the geometry that arevery close together, but not touching, end up intersecting in the mesh. Such an artifactcan lead to inaccurate results due to a short-circuiting effect in the calculation. The usertraditionally has two options in this case. The first is to move the two objects furtherapart, which may not be possible given the design of the geometry. The other is to createcells small enough in that region to resolvethe separation between the objects, which may lead to an extremely large number of cellsand a dramatic increase of computation resources required to run the simulation.

EMPro provides a third option to alleviate the work required by the user. This is the role ofthe Touching Objects functionality. There are two tools available for the user to identifyand correct any unintended artifacts caused by meshing or geometric positioning.

The first, Test Objects in CAD Space, tests two objects for a CAD space intersection.It will report to the user if two objects are actually touching in non-discretized space.


96

The second, Separate Objects in Mesh, enable you to create a relationship betweentwo objects that will ensure they are separated in the final mesh by a user-definednumber of cells.

These tools enable the user to maintain a separation while not sacrificing computationalcomplexity or geometric design.

Test Objects in CAD Space

In order to find out whether two objects are actually touching, EMPro provides the TestObjects in CAD Space function to test the selected objects. This can be done by holding |Ctrl| and selecting two objects in either the Geometry workspace window or under theParts branch of the Project Tree. Then test the objects by selecting Touching Objects >Test Objects in CAD Space, as shown in the following figure.

A dialogue will appear with the results. The user will also have the option of ensuring thatthe objects are separated in the mesh, as seen in the following figure. If Yes is selected,the Separate Objects in Mesh dialog will appear, as described in the following section.

The Touching Object Test dialog

Separate Objects in Mesh


97

Defining a separation operation between two objects ensures that a gap of user-definedsize is maintained despite discretization or geometric positioning. To perform thisoperation, select two objects and navigate to Touching Objects > Separate Objects inMesh, (seen in the figure above). The dialog shown in the following figure will appear.

The Primary Model is separated from the Secondary Model. Ensure they are identifiedappropriately. Pressing Swap Inputs will reverse their labels.Define the Separation Distance (cells) between the two objects. This is the number ofcells that is "taken away" from the Secondary Model in the meshing operation.Specify the Fill Material to use, if any. If no material is specified then the materialexisting in the mesh under the removed object will be used.

NoteThis is a very powerful concept, as it means that any complex configuration of materials may exist in thegap between two objects, and that information is not lost by performing the operation.

The Separate Objects in Mesh dialog

To illustrate this concept, consider the case of two touching cubes made of differentmaterials. A mesh separation of 2 cells and a different (red) material is applied, with thesmaller (purple) cube as the Primary Model and the larger (blue) cube as the SecondaryModel. Though the geometry is unaffected, the mesh separation is visible when viewingthe geometry in Mesh View, as seen inthe following figure.

Touching objects geometry (left) and mesh view (right) with a 2-cell mesh separation applied


98

Once this relationship is specified it will be maintained and updated appropriately in themesh and nothing further is required by the user. If the user decides to delete one of theoperands, the operation will be automatically removed; and if the user re-orients orchanges materials on the operands, the operation will reflect that appropriately. The meshseparation is managed from the FDTD > Mesh branch of the Project Tree.


99

Defining Outer Boundary ConditionsIn this section, you will learn about how to:

Distinguish absorbing and reflecting boundaries in EMProChoose which boundary type to use for your project calculation

Specifying an outer radiation boundary is necessary to indicate how the calculation treatsthe boundaries of the problem space. During an EMPro calculation, the fields updated atevery cell location are dependent upon the neighboring fields. However, due to memorylimitations, the fields on the outer edges of the grid cannot be updated correctly becausethe grid must be a finite size. To correct this situation, outer radiation boundary conditionsare applied at the edges of the EMPro grid. Thus, the performance of the outer boundariesis a significant factor in the accuracy of the calculation, and it is important to use themcorrectly. This chapter details several available options for defining the outer boundariesof an EMPro project.

Outer boundaries are defined in the Outer Boundary Editor, located in the Fdtd: OuterBoundary branch of the Project Tree, as shown below.

Outer Boundary Branch

Absorbing Boundaries vs Reflecting BoundariesThe outer radiation boundary is a method for absorbing fields propagating from the EMProgrid toward the boundary. By absorbing these fields, the grid appears to extend infinitely;however, it is actually finite in order to fall within reasonable memory usage. There aretwo numerical absorbers designed to allow electromagnetic fields radiated or scattered bythe FDTD geometry to be absorbed with very little reflection from the boundary. Theseinclude a Uni-Axial Perfectly Matched Layer (PML) and a second-order, stabilized Liaoradiation boundary.

In some cases a reflecting boundary rather than an absorbing one is preferred. A perfectlyconducting boundary (either electric, PEC, or magnetic, PMC) may be used in these cases,for example, to provide a ground plane, or to image the fields in an EMPro calculation.

The Liao and PML boundaries may not be mixed together in the same calculation.Furthermore, PML may not be used with the PMC boundary. The Liao boundary may beused with both PEC and PMC boundaries.


100

NoteThe default boundary condition for EMPro is PML.

In addition, EMPro has Periodic boundary conditions that enable periodic structures to bemodeled. These boundary conditions equate the corresponding outer surfaces of themesh.

The figure below shows the Outer Boundary Editor.

The Outer Boundary Editor

Liao BoundaryThe Liao outer boundary condition is an estimation method, which is makes itfundamentally different from PML boundary conditions. By looking into the FDTD spaceand back in time, it estimates the electric fields just outside the limits of the FDTD mesh.These estimated values are then used in the FDTD equations inside the space. The Liaoestimation assumes that waves are allowed to travel outward from the space but notreflect back in. This method works well provided that there is enough space between theradiating geometry and the outer boundary. Typical limits are at least 10 cells of spacingto ensure that instability does not occur.

NoteFor more on calculation instability, refer to the section on Calculation Stability (using).

A homogeneous dielectric may be located against the Liao boundary. For example, in alossy earth or strip line calculation, the earth or dielectric layer may touch the outerboundary. Liao will usually function well in this situation provided that there are no airgaps within five cells of the Liao boundary. Liao assumes homogeneous material withinfive cells, and if this is not the case then the EMPro calculation will usually be unstablewith rapidly rising field amplitudes.

Since Liao is an estimation method, the size of the FDTD mesh is not increased by usingit. Some storage is needed for saving electric values at previous timesteps, but this isusually negligible in a typical calculation.


101

PML BoundaryThe Perfectly Matched Layer (PML) boundary condition is offered as an alternative to Liao.PML is an artificial absorbing material that absorbs the incident energy as it propagatesthrough the PML layers. Better absorption, that is, smaller reflection, is obtained byadding more layers at the expense of increasing the size of the FDTD mesh. For example,consider an EMPro calculation on a mesh using the Liao absorber that is 50 x 60 x 70 cellsor a total of 210,000 cells. There is a 15 cell free space border all around the geometry sothat the Liao boundaries can provide small reflections. If the Liao is changed to eight PMLlayers, the geometry mesh will not change. However, outside of this defined mesh region,eight additional FDTD mesh layers are added on each side of the geometry. This meansthat the actual number of FDTD cells that must be calculated grows to 66 x 76 x 86 or431,000 cells, more than double. Since PML cells require more arithmetic operations thannormal cells, the time penalty is actually greater.

This time penalty for PML is also increased because the PML cells have special equationsfor both electric and magnetic fields. For an EMPro calculation with no magnetic materialspresent, the magnetic fields are computed very quickly. However, when PML is added, themagnetic field update equations are more complicated even when no actual magneticfields are present and this adds to the time penalty.

The benefit of using the PML layers is that they provide better absorption than Liao evenwith only a five-cell border of free space, and perhaps only six PML layers would providethis. In such a situation, calculation time would be saved. Making this comparison wouldrequire meshing the object again with a smaller free space margin to the outer boundary.This can be done easily in EMPro using the mesh tab and choosing a smaller paddingaround the geometry.

Both PML and Liao boundary conditions are offered to provide flexibility. Both methodsshould provide similar results when properly used although in some cases, particularlywhen low frequencies (compared to the cell size) are used, PML is superior. It is alsorecommended that PML boundary conditions are used wherever possible when using theadaptive meshing feature.

PEC BoundaryIn some situations, terminating one or more faces of the FDTD geometry space with aPerfect Electric Conductor (PEC) outer boundary is advantageous. For example, theconducting ground plane of a microstrip could be located on one face of the FDTD space.

If all of the outer boundaries of the calculation are not absorbing, a plane wave should notbe used to excite the calculation and the far-zone transformations will not provide correctresults for far-zone fields. The sole exception is in the case of one PEC boundary and fiveabsorbing boundaries, which will compute far zone over infinite PEC ground.

NoteAn edge of the FDTD space should be set to PEC using the PEC Boundary Condition. Do not set FDTD cellsto PEC material in the geometry and set the outer boundary to absorbing, as this will cause instabilities inthe calculation.


102

PMC BoundaryThe Perfect Magnetic Conductor (PMC) outer boundary condition may be useful in reducingthe size of the FDTD mesh, memory requirements, and calculation time by takingadvantage of symmetries in the geometry. For example, this condition would be a goodchoice in a symmetric problem space where magnetic fields are strictly normal to a plane.

Periodic BoundarySimilar to the PMC boundary condition, the Periodic boundary condition may be useful intaking advantage of geometry/field symmetry to reduce the size of the FDTD mesh andtherefore the memory and calculation time required. In this case the upper and loweredges of the mesh are forced to be equal during the analysis. This may be useful for caseswhen small geometries are repeated over and over (i.e., optics examples).

NoteFor more information about using the periodic boundary with Plane Wave excitations, refer to Creating aNew Simulation (using).


103

Saving Output Data with SensorsIn this section, you will learn how to:

Use sensors to save the results of your EMPro project calculationChoose the correct sensor to use depending on the type of data you want to saveMesh voxel objects

Sensors are objects that save data during a simulation. Any type of data that can besaved in EMPro is saved with a sensor. The type of data that is saved by a sensor isdependent on the sensor type, as well as the specific data that is requested within varioussensor editors. There are various types of sensors that are available within EMPro,including:

Port sensorsNear Field sensors, including:

Point sensorsSurface sensorsRectangular sensorsPlanar sensorsSolid Part sensorsSolid Box sensors

Far Zone sensorsSpecific Absorption Rate (SAR) sensorsHearing Aid Compatibility (HAC) sensors

Result objects are generated based on the sensor objects that are defined in the project.After a sensor has been placed, an editor is used to define its characteristics based on theoutput data. Each type of sensor has its own respective editor window. This section detailsthe process of adding sensors into an EMPro project and requesting specific results witheach type of sensor.

Sensor ToolsIn general, sensors added to the project within the Sensor Tools dialog (with theexception of Port and SAR sensors). There are two ways to open this dialog. The first is tochoose Sensor Tools from the drop-down list in the Geometry workspace window. Thesecond is to right-click on the Sensors branch of the Project Tree and select the branchthat corresponds to the desired sensor type.

NotePort sensors are added by setting a component property. For more information on how to add a Portsensor, refer to Adding a New Component. SAR sensors are added by double-clicking on the sensor in the Project Tree. For more information on SARsensors, refer to SAR Sensors.

Sensor Tools menu


104

The following sections detail the process of adding each type of sensor within Sensor Tools, defining its associated characteristics, and requesting the desired output data to becalculated by the sensor.

Port SensorsA Port sensor saves near-zone voltage and current data at the location of a circuitcomponent. Port sensors are automatically added to the project when a circuit componentis added and the This Component Is A Port property box is checked in the Properties tabof the Circuit Component Properties dialog.

Each Port sensor can have a different source resistance. For more information onspecifying the source resistance, refer to Specifying the Source Resistance.

S-Parameter, Group Delay, VSWR and Reflection CoefficientCalculations

When S-parameter computation is enabled in the Simulations workspace window, Portsensors will also save data used to compute S-parameters, Group Delay, VSWR, andreflection coefficient. When multiple Active Feeds are used in the simulation, S-parameterswill be computed at each Port sensor with respect to each active feed.

NoteS-parameters at each Port sensor are calculated using the characteristic impedance retrieved from thecircuit component definitions of that port and the active feed. For more information on defining S-Parameter calculations with single or multiple ports, refer to S-parameters Simulation Setup.

Near Field SensorsNear Field sensors are used to save time-domain and/or frequency-domain near zone fieldquantities at specific points within the bounds of calculation space. In general, field data isretrieved using the Point, Surface, or Solid sensors, and hearing aid field values arerecorded using the Hearing Aid Compatibility (HAC) sensor. Solid sensor results can beviewed as 2-D plots, and both Solid and Surface sensors can be viewed as 3-D fieldsequences (excluding Point sensor data).


105

Field Retrieval

The field quantities of X -, Y -, and Z -directed electric (E) and magnetic (H) fields may besaved at a specific point, across a surface, or throughout a volume with a Point sensor,Surface sensor, or Solid sensor, respectively. Additionally, X -, Y -, and Z -directedcurrent density (J) may be collected with any of these sensors. Current densities aredetermined by multiplying the conductivity of the material at the specified location by theelectric field in the given direction. When a PEC material is present, the current densitywill be computed by the loop of magnetic fields surrounding that cell edge. Thus, thecurrent density only includes the conduction current. When a near-zone source is used asthe input, the total field values are available. With an incident Plane Wave input, thescattered and total electric and magnetic fields may be saved in addition to the totalcurrent density.

Samplings of near field data may be saved by specifying Sampling Time Range in any ofthe near field sensor definition windows. Near field data will be collected in specific planesof the geometry during the EMPro calculation at every interval specified within thedefinition. A field file containing the electric and magnetic fields and the current will becreated for each timestep specified. For example, setting an entry beginning at timestep100, ending at timestep 1000, with an increment of 100 will create 10 field files whichmay be viewed as a movie after the EMPro calculation is performed.

NoteBe aware of the number of field slices to save, as they can store enormous amounts of data. Single fieldfiles may contain megabytes of data depending on the number of cells in the specified plane.

Point Sensors

A Point Sensor is positioned at a specific point-location in the simulation space, and can bedefined by the location of a specific vertex in a part object or by a Cartesian 3-Dexpression. The sensor records data as it occurs at the specified point in space.

Point sensors record data by means of field interpolation or geometric "snapping". Whenusing the interpolated sampling method, the field components are interpolated to theexact location of the point sensor. This is performed by linear interpolation among thesurrounding eight appropriate field value sample points (e.g., when measuring Ex, the

eight surrounding X -directed edge centers are used for the interpolation, and whenmeasuring Hx , the eight surrounding X -directed cell face center points are used for the

interpolation). When using the snapped sampling method, the location of the point sensoris snapped to the nearest E-grid cell vertex. Field components for snapped point sensorscome from the cell whose lowest-index corner is defined by the snapped location of thesensor. The sensor location is thus dependent on the grid definition.

Point Sensor Properties

To define a Point Sensor, open the point sensor properties dialog under Sensor Tools. Inthe Location tab, define the sensor location manually by typing in its coordinates, orautomatically by clicking on the intended location in the simulation space with theSelection tool. In the Properties tab, enter the name of the sensor, select the desired Point


106

Sensor Definition, and choose the sampling method as described above.

Point Sensor properties dialog

Point Sensor Definition Editor

The Point Sensor Definition Editor window is used to assign definitions associated with aPoint Sensor.

To access the editor, double-click on an existing Point Sensor Definition in the Definitions:Sensor Data Definitions branch of the Project Tree. If no point sensor definition is present,right-click on this branch and select New Point Sensor Definition.

In the Fields to Save region of the editor, select the desired point sensor output data tosave:

E: Electric Field Intensity timeH: Magnetic Field Intensity vs timeB: Magnetic Flux Density vs timeJ: Current Density vs timeScattered E: Scattered Electric Field vs timeScattered H: Scattered Magnetic Field vs timeScattered B: Scattered Magnetic Induction Field vs time

NoteScattered field values can be retrieved only if a Gaussian beam or a scattered field plane wave externalexcitation is used to excite the simulation.

Point Sensor Definition Editor


107

Define the Sampling Time Range by entering the Start Time and End Time, or by simplychecking Start of Simulation and End of Simulation to automatically assign the samplingtime range to these values. Choose a Sampling Interval to indicate how often data issaved within this time range.

Surface Sensors

Surface sensors collect data on one or more faces of a geometric object in the simulationspace. Like Point Sensors, they can be interpolated or mesh-snapped.

There are three types of surface sensors in EMPro:

Sensor On Part SurfaceRectangular SensorPlanar Sensor

NoteRefer to the Surface Sensor Definition Editor section to reference the output data that can be retrieved bya surface sensor after it has been created within Sensor Tools.

Sensor on Part Surface Properties

To define a Sensor On Part Surface, select the object in the simulation space that thesensor will be attached to by clicking on it in the Pick Model tab. In the Pick Faces tab,select the specific face to attach the surface sensor. Finally, in the Properties tab, andassign the new sensor a Name, Definition and Sampling Method.

NoteDefinition, as mentioned here and in the following two sensor descriptions, refers to definitions stored inthe Definitions: Sensor Data Definitions branch of the Project Tree.

Sensor on Part Surface properties dialog


108

Rectangular Sensor Properties

Define a Rectangular Sensor by first using the Orientation tab to choose the plane inwhich the rectangle is defined. Then, use the Rectangle tab to define the rectangle's twoopposite corners. Finally, under the Properties tab, assign the sensor a Name, Definitionand Sampling Method.

NoteFor an explanation of the Orientation tab, refer to the section Specify Orientation Tab.

Rectangular Sensor Properties dialog

Planar Sensor Properties

The Planar Sensor uses a point and normal direction defined in the Orientation tab todefine an entire plane (within the boundaries of the simulation space) to collect sensordata. Select the Properties tab and assign the sensor a Name, Definition and SamplingMethod.

Planar Sensor Properties dialog


109

Surface Sensor Definition Editor

The Surface Sensor Definition Editor window is used to assign definitions associated with aSurface Sensor.

To access the editor, double-click on an existing Surface Sensor Definition in theDefinitions: Sensor Data Definitions branch of the Project Tree. If no surface sensordefinition is present, right-click on this branch and select New Surface Sensor Definition.

In the Fields to Save area of the editor, select the desired surface sensor output data tosave:

E: Electric Field Intensity timeH: Magnetic Field Intensity vs timeB: Magnetic Flux Density vs timeJ: Current Density vs timeScattered E: Scattered Electric Field vs timeScattered H: Scattered Magnetic Field vs timeScattered B: Scattered Magnetic Induction Field vs time

Scattered field values can be retrieved only if a Gaussian beam or a scatteredfield plane wave external excitation is used to excite the simulation.

Steady E: Steady Electric Field vs timeSteady H: Steady Magnetic Field vs timeSteady B: Steady Magnetic Induction Field vs timeSteady J: Steady Current Density Field vs time

Surface Sensor Definition Editor


110

Define the Sampling Time Range by entering the Start Time and End Time, or by simplychecking Start Of Simulation and End Of Simulation to automatically assign the samplingtime range to these values. Choose a Sampling Interval to indicate how often data issaved within this time range.

Solid Sensors

Solid sensors collect data by capturing mesh-snapped fields within a volumetric space(interpolated data is not available).

There are two types of solid sensors in EMPro:

Solid Part SensorSolid Box Sensor

NoteRefer to Solid Sensor Definition Editor to reference the output data that can be retrieved by a solid sensorafter it has been created within Sensor Tools.

Solid Part Sensor Properties

A Solid Part Sensor simply assumes the shape of the part that is selected in the Pick Modeltab. Assign the sensor a Name and Definition in the Properties Tab.

NoteDefinition, as mentioned here and in the following sensor description, refers to definitions stored in theDefinitions: Sensor Data Definitions branch of the Project Tree.

Solid Part Sensor Properties dialog


111

Solid Box Sensor Properties

A Solid Box Sensor assumes the shape of a 3-D box. This shape is dictated by the Originlocation defined in the Orientation tab, and its farthest corner is defined in the OppositeCorner Tab. Assign the sensor a Name and Definition in the Properties Tab.

NoteRefer to the section Specify Orientation Tab for an explanation of the Orientation tab.

Solid Box Sensor Properties dialog

Solid Sensor Definition Editor

The Solid Sensor Definition Editor window is used to assign definitions associated with aSolid Sensor.

To access the editor, double-click on an existing Solid Sensor Definition in the Definitions:Sensor Data Definitions branch of the Project tree. If no solid sensor definition is present,right-click on this branch and select New Solid Sensor Definition.

In the Fields to Save area of the editor, select the desired solid sensor output data tosave:

E: Electric Field Intensity timeH: Magnetic Field Intensity vs timeB: Magnetic Flux Density vs timeJ: Current Density vs time


112

Scattered E: Scattered Electric Field vs timeScattered H: Scattered Magnetic Field vs timeScattered B: Scattered Magnetic Induction Field vs time

Scattered field values can be retrieved only if a Gaussian Beam or a scatteredfield Plane Wave external excitation is used to excite the simulation.

Steady E: Steady Electric Field vs timeSteady H: Steady Magnetic Field vs timeSteady B: Steady Magnetic Induction Field vs timeSteady J: Steady Current Density Field vs time

Solid Sensor Definition Editor

Define the Sampling Time Range by entering the Start Time and End Time, or by simplychecking Start of Simulation and End of Simulation to automatically assign the samplingtime range to these values. Choose a Sampling Interval to indicate how often data issaved within this time range.

Hearing Aid Compatibility Sensors

Hearing Aid Compatibility (HAC) sensors gather data on a 5cm by 5cm arbitrarily-orientedrectangle in freespace. They are used to determine if a wireless device (such as acellphone) will generate electrical and magnetic fields large enough to interfere with ahearing aid. In these cases, they are useful for evaluating the wearer's ability to adjustthe position of the phone to a better location.

The HAC sensor is centered at the origin of the coordinate system described in theGeometry tab.


113

This sensor collects steady-state E and H fields at grid points near the HAC plane at eachfrequency of interest. These values can be then interpolated onto the plane at a user-defined spatial resolution.

HAC Sensor properties dialog

Far Zone SensorsFar Zone Sensors are located at theoretical infinite distance from the simulation geometry.They are only available in the absence of PMC or periodic outer boundary conditions, orwhen more than one PEC boundary is used.

To create a Far Zone Sensor, under the Geometry tab, choose its coordinate system:

THETA/PHIALPHA/EPSILONELEVATION/AZIMUTH

Far zone sensors can be created with one of the following geometries:

A range of theta/alpha/elevation over a constant (single) phi/epsilon/azimuthA range of theta/alpha/elevation over a range of phi/epsilon/azimuthA range of phi/epsilon/azimuth over a constant (single) theta/alpha/elevation

The following figures demonstrate the transformation of the far zone sensor based on the


114

defined geometry in the Theta/Phi coordinate system.

Several Far Zone sensor geometries


115

As seen under the Properties tab, EMPro has the ability to compute both a steady-statefar-zone (near-to-far) and a transient far-zone transform. These two options are describedbelow.


116

Steady-state Far-zone Transformations

Steady-state transformations are particularly advantageous because the calculationoverhead is minimal. They do not require the definition of specific far-zone angles beforethe FDTD computation, since all patterns are computed in post-processing using data thatis automatically stored by the EMPro calculation engine. Instead, the calculation saves thetangential electric and magnetic fields on the far-zone transformation surface at twotimesteps, near the end of the calculation when the system should be in steady state. Thissampling determines the complex tangential fields on the far-zone surface at theexcitation frequency. These fields are then used in post-processing to provide radiationgain or bistatic scattering in any far-zone direction at any pattern increment. This savesconsiderable computer time and memory if many far-zone directions are required.

Additionally, the selection of a steady-state far-zone transformation computes the singlefrequency input impedance, total input power, radiated power, and antenna efficiency. Allvalues computed require that the calculation has reached steady state.

Defining a Far Zone sensor

The far-zone gain is displayed in units of dBi. This is the number of decibels of gain of anantenna referenced to the zero dB gain of a free-space isotropic radiator. This value iscalculated based on the net power available at the source voltage output. Directivity is notavailable.

Transient Far-Zone Transformations

The transient far-zone calculation should be used when broadband results are desired,since the steady-state transform is only performed for a single frequency. The broadfrequency range, therefore, can be determined at a few points in space. An additionalfeature of the transient far-zone transformation is that the time-domain far-zone electricfields are also generated and may be plotted.

The transient far-zone calculation requires extra calculation time for each far-zone anglespecified, and unlike steady-state far-zone transformations, all far-zone angles must bedefined before running the calculation engine. The transient far-zone calculation isintended for use in cases where the far-zone results at a few points are desired since it iscomputationally intensive. This calculation may be desired in instances when far-zonetime-domain fields are needed.

If detailed gain patterns versus angle are necessary, you may reduce calculationtime by enabling only steady-state data collection for your sensor and specifyinga DFT frequency for your simulation at each frequency you are interested in.


117

SAR SensorsEMPro includes several features that fall under the category of biological applications. Forcompliance with regulations on field absorption in human tissue, the Specific AbsorptionRates (SARs) can be computed and averaged. Detailed human body meshes are availablefor simulations related to effects on realistic heterogeneous models of the body. For somewireless applications, the Specific Anthropomorphic Mannequin (SAM) head is used inaddition to the heterogeneous human head (see the figure below).

The Specific Anthropomorphic Mannequin (SAM) head

The Specific Absorption Rate, or SAR, is the unit of measure commonly used to determinethe interaction of electromagnetic fields with human tissue. Most regulations involvingdevices producing electromagnetic fields must not exceed some exposure limits, typicallydefined in terms of the SAR averaged over a cubical volume of tissue.

NoteAs an example, the IEEE sets exposure levels in terms of 1 g averaging volumes for most of the body,with a 10 g averaging volume applying to extremities such as the ears and fingers.

SAR is defined in terms of the room mean square (RMS) of the electric field magnitude bythe relation

Where:

is the electrical conductivity in , and


118

is the material density (defined in in EMPro )

Since the FDTD grid defines the electric fields at the edges of the cells, a single SAR valueis formed by summing and averaging the contributions of the 12 electric fields on theedges of the cells. The SAR is then referenced to the center of the FDTD cell.

In EMPro, the SAR is measured with the SAR sensor and may only be computed in normaldielectric materials. Frequency-dependent materials have a loss term formed by theimaginary part of the permittivity rather than simply by the conductivity, and are notsupported for SAR calculations.

The SAR values are saved only in complete voxels (closed FDTD cells) where all 12 edgesof the cell are lossy dielectric material (non-zero conductivity) with a non-zero density,therefore steady-state values for SAR and conduction currents will not exist in all planes.To exclude certain materials from a SAR calculation, simply leave the material density aszero. Saving the SAR in a plane of free-space will not produce any useful output as allvalues will be zero.

NoteFor more information on voxels, refer to the section on Voxels.

Un-averaged SAR Calculation

Un-averaged SAR is measured in EMPro using the SAR Sensor. Note that mostspecifications which involve SAR limits are defined in terms of constant-mass regions, sothey will require averaged SAR.

Averaged SAR Calculation

The averaged SAR calculation is more meaningful under most circumstances. Thiscalculation is defined by regulations from organizations such as the IEEE and variousgovernment bodies. It is computed over cubical volumes of voxels where no face of theaveraging volume is external to the body (and thus full of air or other non-tissuematerial). In certain cases, particularly at the surface of the body, the cubical volume rulecan not be satisfied. In those situations, special rules exist for setting the SAR value in agiven voxel. Refer to the IEEE published standards for regulating SAR calculations andsetting SAR values in the Bibliography.

NoteOnly one SAR averaging region can be defined per calculation run. Additional averaging can be performedas a post-processing step, given that sufficient un-averaged SAR was collected for the region of interest.

Averaged SAR is measured in EMPro using the SAR Averaging Sensor. There are severalways to compute average SAR values in EMPro, as shown in the SAR Averaging tab. Oneway is to save 1 gram or 10 gram average SAR regions over the Full Grid. During thecalculation the averaged SAR values will be computed for all appropriate voxels. Thisprocess is time consuming, and since the 10 gram SAR is only applicable to the extremitytissues, it is not necessary to compute it for the entire geometry.


119

As an alternative to computing values over the entire grid, the EMPro interface also has atool for computing the average SAR over a Box Region. If a subregion of the whole objectis defined, then an option is available to allow all data outside of that subregion to beconsidered as free space. The figure below displays the Box Region dialog.

Another alternative is to select the Auto Subregion option. In this case, the Max/Min SARRatio is defined in decibels so that the requested 1 gram or 10 gram average is performedonly where applicable, thus saving a great deal of calculation time. This quantity must beentered as a unitless ratio (amplitude) or in dBp (a decibel unit with suffix to indicate anabsolute unit of electric power). For example, in a typical application, the extremitytissues would be identified by different material types from the body tissues, so indicatingthis value in the Max/Min SAR Ratio would isolate the calculation to that specific region.The figure below displays this dialog.

Requesting averaged SAR statistics in an automatic subregion

EMPro also offers tissue selection control under the Tissue Materials tab. You can computeaveraged SAR for All Tissue Materials, or for Selected Extremity Tissue Materials. Inchoosing the latter option, a dialog box will appear with a list of available pre-definedmaterials to include in the calculation.


120

Running CalculationsIn this section, you will learn:

How to create and run a simulationHow to start the EMPro calculation engineThe main factors to consider before beginning a calculation

After the EMPro project setup is complete, it is time to run calculations on the geometry.The Simulations workspace window stores the project simulation(s). From this window,the user creates, queues and runs the simulations.

The actual electromagnetic calculations are not made by the EMPro GUI. Rather, theelectromagnetic calculations are run by a separate program called EMPro FDTD once thefinished project file has been saved. Usually this process is run from the EMPro GUI, whichcalls EMPro FDTD as needed. However, the user may also run EMPro FDTD directly fromthe command line, or from a remote computer. Once the simulations have been run andthe desired calculations have completed, the results can be viewed within the EMPro GUI.

Simulations Workspace WindowThe Simulations workspace window provides the interface to queue projects to be run withEMPro FDTD, the calculation engine.

The Simulations workspace window

The Simulations workspace window lists the name of every simulation that has been


121

created for the project. Its Status column shows whether the simulation has been created,queued, or completed.

This window also provides the user with the ability to choose how to run the simulation ontheir computer. Under the Queue drop-down list, specify whether to run the simulation onthe CPU or on the hardware card with Acceleware hardware acceleration.

New simulations are created by pressing the New Simulation button in the upper-leftcorner of this window. The associated definitions are described below.

Creating a New Simulation

Choosing a Source

There are two main options for choosing a source in in EMPro, as seen in the figure below.When a voltage or current source is used as an input, check Use Discrete Sources.Alternatively, the user may also Select External Excitation to use as the source. Use NoSources should be selected in special cases with the Static Solver. Discrete sources andexternal excitation sources are briefly described below.

NoteFor more information on discrete sources, refer to Circuit Component Definition Editor. For moreinformation on external excitations, refer to External Excitation Editor.

Choosing a source for a simulation


122

Discrete Source

A discrete source is a cell edge on which the electric field is modified by the addition ofsome type of input waveform. The cell edge can be modified to behave like a voltage orcurrent source. All calculations with discrete source input are performed in total field.Antenna and microwave circuit computations are examples of calculations that may beperformed using discrete sources.

Plane Wave

An incident Plane Wave source is assumed to be infinitely far away so that the constantfield surfaces are planar and normal to the direction of propagation. All calculations withthis plane wave source are performed in scattered-field. Total field values may be savedand are typically more desirable than scattered-field plane wave. Calculations of radarcross section or scattering may be performed using this input.

NoteFor important considerations when choosing between scattered and total-field plane wave sources, refer toPlane Wave Excitations.

Gaussian Beam

This choice allows for a focused Gaussian beam source in which the incident electric field


123

has a two-dimensional, radially-symmetric Gaussian distribution in planes normal to theincident direction and converges to maximum intensity at the focus point. As with theplane wave source, all calculations with a Gaussian beam source are performed inscattered field, though total field values may also be saved and displayed also. Unlike theplane wave and discrete sources, the Gaussian beam source requires that the sourcewaveform be sinusoidal. Examples where this type of source is useful include structuresused at optical frequencies and situations where it is desired to illuminate only a portion ofthe geometry.

Parameter Sweep Setup

A parameter sweep can be set up so that a script will loop through a particular parametermultiple times and run a calculation at each iteration. For example, if the parameter isantenna length, a script can loop through various antenna lengths, and run a calculationat each length. The Sweep Type is defined by the values it will collect (Start, Incr., Countvalues, or Start, End, Count values), in order to generate the Ending Value or Incrementvalues, respectively. A third option is to select Explicit Values for the parameter sweep,which do not necessarily have to be at evenly spaced intervals.

NoteFor more information on scripting and to see a sample script for a simple parameter sweep, refer toScripting Workspace Window.

Setting up a parameter sweep

S-parameters Simulation Setup

To calculate a S-parameter for a project with multiple ports defined, the simulation willconsider one port at a time to be the active port. The computation with that source activewill provide data for a column of the S-parameter matrix. For example, if port one isexcited in a three-port circuit, EMPro will determine S11, S21 and S31. If different ports

are to be excited, a separate calculation must be performed with each port active. Forexample, if the full S-parameter matrix for a two-port problem is desired, two calculationsmust be performed with a different port active in each. EMPro will save the S-parametersfor each run in separate files, differentiated by the active port number.


124

NoteFor more information about the data saved with port sensors, refer to Sensor Tools.

If a parameter sweep is specified in the Setup Parameter Sweep tab for multipleparameters in addition to specifying multiple ports within the Setup S-Parameters tab, the parameter sweep will be performed for each individual port.

Setting up S-parameters

Frequencies of Interest

Within this section, the Frequencies tab specifies whether the simulation is a broadband(transient) or steady-state calculation. For broadband calculations, uncheck the CollectSteady-State Data box at the top of this tab. For steady-state calculations, check this boxand choose whether the calculation is to only Use Waveform Frequency or to Use SpecifiedFrequencies of interest. By specifying more than one frequency of interest, the calculationengine will essentially run a separate calculation at each discrete frequency by runningDFT, saving each as its own run. This will therefore increase the calculation time incomparison to using only the waveform frequency.

In the Data Storage tab, the user has the ability to specify whether to save temporarydata In Memory or On Disk. Saving the data in memory will speed up the calculationbecause there is no file saving or loading from disk, but it increases the memoryrequirements.

Also use this tab to designate which data to save for steady-state far zone post-processing. For a broadband excitation, the user has several options. Checking NormalizeFields will match calculated values to a sinusoidal run. Checking Compute DissipatedPower will calculate dissipated power based on electric field and magnetic field samplings.It is recommended to leave this box unchecked unless there is specific interest indissipated power, because it can increase run time significantly due to sampling data overthe entire geometry. Sampling Interval specifies how often to sample a data type. Asampling interval of one provides the most accurate results because it reduces the effectsof aliasing.


125

NoteFor more information on viewing far zone post-processing results, refer to Post Processing.

Setting up Frequencies of Interest under the Frequencies and Data Storage tabs

Specifying Field Formulation

The Specify Total/Scattered Field Interfaces definitions are only applicable with PlaneWave sources and are dependent on boundary condition specifications. They regulate howto perform the calculation in certain regions, specifically where a region of total-field issurrounded by a region of scattered-field. The interface between the two regions must befree space. Scattered fields cannot be sampled inside the total-field region and vice versa.Calculations of radar cross section or scattering are based on fields inside the scattered-field region. For non-periodic boundaries, the six sides of the total-field region are definedas eight cells into the FDTD mesh. For these conditions, there is no option to turn theinterface off.

Setting up Total/Scattered Field Formulation


126

There are two sets of X, Y, and Z boundaries listed within this dialog. The checkboxeslisted in the Desired dialog are available for users to indicate which interfaces will beturned on. Depending on the boundary conditions, however, the selected Desireddefinitions may not be applicable or may have to be applied in conjunction with otherdefinitions. Thus, the Effective dialog displays the actual definitions that will be appliedduring the calculation.

When Periodic boundaries are specified, certain sides for the interface may be turned offand the total-field region may extend to the boundary using this definition. Periodicboundaries may be useful for applications such as optics where small geometries arerepeated over and over again. The figure below illustrates an example in which the outerboundaries have been set to Periodic in the the Y and Z directions, and the total/scatteredfield interface has been turned off (unchecked) in the lower X , upper and lower Y , andupper and lower Z directions.

The interface may be turned off for problems that use periodic boundary conditions

Specifying Termination Criteria

Convergence and stability are essential in determining whether a calculation will yieldusable results. Convergence in a broadband calculation is met when all electromagneticenergy has dissipated to essentially zero. There are several options in EMPro to definetermination criteria to ensure proper convergence had been reached.

Selecting termination criteria within the Simulations workspace window


127

The most basic method of ending a calculation is defining a value in the MaximumTimesteps definition. Once the defined number of timesteps has completed, the calculationwill stop. It is important to note that the calculation will terminate regardless of whetheror not convergence has been met, so setting this definition to a proper value is important.If it is too low, results will be of no use.

Selecting the Detect Convergence check-box will automatically stop the calculation if slowconvergence is detected, regardless of whether the number of maximum timesteps havecompleted. Due to numerical noise in the calculation, there may be a trivial amount ofelectromagnetic energy, even after the calculation has converged. The value defined inthe Threshold dialog dictates when the calculation has reached an acceptable value toassume convergence.

NoteA general rule of thumb is that the values should have diminished by at least 30 dB or 1/1000th from thepeak values.

For sinusoidally-excited problems, typical values for this setting range from -55 dB to -25dB depending on the level of accuracy versus runtime desired. For instance, if highaccuracy S-parameters are the goal, then the convergence threshold should be set tolower than -30 dB. If however, the user wants to view antenna patterns, -30 dB to -35 dBis suitable. The trade-off here is run time for accuracy. In general, for sinusoidally excitedproblems, using automatic convergence with a threshold of about -35 dB will produce veryaccurate results and will run in the shortest possible time to reach this level.

If a calculation is finished but convergence has not been reached, the output from mostcalculations will be meaningless. The only option is to decrease the convergence Thresholdor increase the number of Maximum Timesteps in the calculation and run it again. If aresonance is occurring at a frequency beyond the range of interest, and a broadband inputis used, the input waveform can be modified to limit the frequency content and eliminate


128

the resonance. If the resonance is in band, or a sinusoidal input is used, then this is notapplicable and more timesteps must be run or a lower convergence threshold must bechosen.

Advanced Options

There are several options available in the Advanced button of this window that allow formore specific termination criteria.

Selecting the Flatline Detection check-box will stop the calculation if a slow convergence isdetected. This may occur if the user sets the convergence threshold to a very low value(e.g. < -50 dB). In this case the calculation may converge but to a level higher thanspecified.

NoteTo prevent false convergence, "slow" convergence can only be detected once the convergence level hasreached at least -40 dB.

For a steady-state calculation, convergence is reached when near-zone data shows aconstant amplitude sine wave - when all transients have died down and the only variationleft is sinusoidal. In this case "convergence" is tested on the average electric field in thespace for its deviation from a pure sine wave. If Detect Convergence is turned on, EMProautomatically places points throughout the space for this purpose. It is particularlyimportant to monitor the results inside high permittivity dielectrics since the fieldpropagation in these materials is much slower than in free space.

To ensure that steady-state calculations converge, EMPro will enable you to control theSampling Interval and Sampling Density of the sample points. The temporal SamplingInterval definition is used to control the interval for which convergence is tested duringcalculations with broadband (pulse) excitations.

NoteSetting this value to 100 or 200 timesteps is typical. Setting this value to much less than that increasesthe computational overhead a small amount.

Sampling Density is used to control how many spatial samples are used to determineconvergence. The sample points are equally spaced in all three dimensions of the grid.Low density samples every 4th point in each dimension, while High density samples everypoint in each dimension. For very low frequency problems or where the number oftimesteps per RF cycle is greater than 200 (e.g., very small cells with a low frequencyexcitation), this should be set to LOW. For moderate frequencies or where the number oftimesteps per RF cycle is less than 200 but greater than 100 this should be set to Medium.For high frequency problems where the number of timesteps per RF cycle is less than 100,a High setting gives the best accuracy. This setting is for both broadband as well assinusoidal excitations.

Notes

The Notes section is simply a tool for users to add any notes to be attached to a project.The notes will be available in the Simulations workspace window after the simulation iscreated.


129

Queuing and Running Simulations

After defining the necessary components of the simulation in the New Simulation dialogdescribed above, select Create And Queue Simulation and the main window will show allof the queued and completed simulations as seen in the figure below. Only one simulationcan be run at one time, so as soon as a calculation is complete, another queuedsimulation will begin.

Running a calculation

Below this dialog are three tabs:

Summary, where a basic summary of the calculation is provided, as seen in below.Notes, which simply documents any notes that were added by the user in the NewSimulation dialog.Output, which provides the output generated by EMPrFDTD. Statistics such as thepercentage of completion, current timestep, convergence, time elapsed, etc, arelisted for every simulation in this tab (including output information for everyparameter swept).

The calculation Summary tab


130

Starting the Calculation EngineThere are two ways to start the EMPrFDTD engine. The first way is to launch it from EMProin the Simulations workspace window. The second way is to start the calculation from acommand line. This is preferable for calculations that require a large amount of memory inorder to free memory used by the interface.

Running EMPrFDTD Within the EMPro Interface

Calculations launched from EMPro are sent to the EMPrFDTD engine from the Simulationsworkspace window. A list of every simulation that has been created within the NewSimulation dialog is listed in the main window. When a simulation is ready to run, send itto EMPrFDTD by selecting the simulation and clicking the Add To Queue button, located tothe right of Selected Simulation at the top of the window. Although multiple simulationscan be queued at once, only one simulation is run in EMPrFDTD at a time. Once the Playbutton is pressed, each simulation will run one at a time until all of the queued simulationshave terminated.

NoteFor more on creating a new simulation, refer to Creating a New Simulation.

Running EMPrFDTD From the Command LineIn some cases, it is useful to run EMPrFDTDfrom a command line. This will prevent EMProfrom overloading since it will not be allocating memory for EMPrFDTD in addition to itsnormal memory requirements. This is not an issue for most calculations, but it may causeproblems when the memory required to run a calculation approaches the limitations ofcomputer memory. In this case, closing EMPro and running EMPrFDTD from the commandline will free up any memory used by EMPro.

To run EMPrFDTD, navigate to the appropriate project folder and run the command:

Windows: emprfdtd.exe [options]


131

Mac OS X/Linux: emprfdtd [options]

For normal operation no [options] need be specified, but the table below lists severaloptions that may be of interest.

Command Line Options

option Function

v Verbose mode, prints progress of calculation.

proc N Sets the number of processors to use for a calculation with the multi-processormodule (optional).

loadstatice Loads fields saved by the static voltage solver.

savestaticeonly Saves fields by the static voltage solver and terminates before time-stepping.

savestatice Saves fields by the static voltage solver and continues timestepping.

hardware Uses Acceleware hardware acceleration; in the case that an error occursinitializing Acceleware, the engine uses the software to run calculation.

hardwareonly Uses Acceleware hardware acceleration; in the case that an error occursinitializing Acceleware, the calculation is terminated.

forcetemporarydatatomemory Forces steady-state data to be stored in memory rather than in an external file.

saronly Calculation skips time-stepping and only performs post-processing of SAR data.

Running the Calculation RemotelyRunning a remote calculation is an alternative way to potentially speed up calculation timeand free local computer resources for other operations. This section will explain how tocopy files and launch the calc engine on a remote computer.

NoteIf your EMPro project is stored on a file system which is remotely accessible, you can avoid the process ofcopying files back and forth from the remote machine. Simply log into the remote machine, navigate tothe simulation directory and launch the calc engine.

Running Simulations Remotely

Create the simulation. After creating the simulation from the Simulations1.workspace window, click the Create Simulation Only button to save the simulation tothe Simulations folder where your EMPro project is stored.

You can find the path to this directory in the first line of text under theSummary tab

NoteFor more information about setting up a simulation in EMPro, refer to Simulations WorkspaceWindow.

Copy files to the remote machine. For simplicity, you can copy the whole2.Simulations directory (you'll need to copy it recursively, since it containssubdirectories). However, only a few files in this directory are required by thecalculation engine:


132

project.xsimRun*/geometry.inputRun*/mesh.inputRun*/project.input

NoteThe * in the file name represents the simulation number, (e.g., Run0001/geometry.input,Run0002/geometry.input, etc).

For Windows users, it may be easiest to zip each file individually and transfer them tothe remote machine over an ftp connection.For Mac OS X users, you can save these files in a compressed archive with thefollowing command:

tar -czf inputFilesForCalcEngine.tar.gz project.xsim Run*/geometry.input

Run*/mesh.input Run*/project.input

Then copy the inputFilesForCalcEngine.tar.gz file to the remote machine and extractit using:

tar -xzf inputFilesForCalcEngine.tar.gz

Run the simulation Log in to the remote machine, change directories to the3.Simulations folder, and run EMPrFDTD for the project.

NoteFor more information on running the calc engine from the command line, refer to RunningEMPrFDTD From the Command Line.

Copy files back to local machine. In order to view your results from the EMPro4.GUI, you'll need to copy all of the Simulations files from the remote machine to yourlocal machine, with the exception of the files listed above. That is, after running thesimulation on the remote machine, all new files should be copied back into theSimulations folder on your local machine. The tar utility makes this easy. On theremote machine, change directories to the Simulations directory and run

tar -czf outputFilesFromCalcEngine.tar.gz -exclude=project.xsim -

exclude=Run*/geometry.input -exclude=Run*/mesh.input -

exclude=Run*/project.input *

For Windows users, zip the outputFilesFromCalcEngine.tar.gz file and send it back toyour local machine over the ftp connection. You can then extract the files usingWinzip. Overwrite any old files with newer versions of the same file.For Mac OS X users, copy the outputFilesFromCalcEngine.tar.gz file to your localmachine, change directories and extract it using:

tar -xzf outputFilesFromCalcEngine.tar.gz

Refresh results. To make your results available from the EMPro GUI, click Refresh5.in the Results workspace window.

Remote SAR Post-processing


133

Setting up post-processing. After requesting post-processing from the RESULTS1.workspace window, in the subsequent dialog box, tell the application to not run thecalculation right away.Copy raw SAR data, calc engine input files, and SAR request files. The2.following files are required for post-processing in the Simulations folder on theremote machine:

project.xsimRun*/geometry.inputRun*/mesh.inputRun*/project.inputRun*/SteadyStateOutput//SAR_Raw_Sensor..sar.gzRun*/SteadyStateOutput/*/SAR_Raw_Sensor.statsRun*/request.sar

The * in the file name represents the simulation number, (e.g.,Run0001/geometry.input, Run0002/geometry.input, etc).

For Windows users, it may be easiest to zip each file individually and transfer them tothe remote machine over an ftp connection.

For Mac OS X users, change the directory to Simulations and run the followingcommand on your local machine:

tar -czf sarRawData.tar.gz project.xsim Run*/geometry.input Run*/mesh.input

Run*/project.input Run*/SteadyStateOutput//.sar.gz Run*/request.sar

Then copy the sarRawData.tar.gz file to the remote machine and extract it using:

tar -zxf sarRawData.tar.gz

Run SAR averaging on the remote machine. On the remote machine, change3.directories to the Simulations directory, and execute the calc engine application usingthe -saronly command line flag.

NoteFor more information on running the calc engine from the command line, refer to RunningEMPrFDTD From the Command Line.

Copy SAR results back to the local project. If you've already copied your other4.results back to the local project, you'll need to package up your SAR results and copythem back. From the remote machine, copy the following files to the Simulationsdirectory on the local machine:

Run*/statusRun*/SteadyStateOutput/*/*gsar.gzRun*/SteadyStateOutput//.infosar1gRun*/SteadyStateOutput//.infosar10gRun*/SteadyStateOutput//.infoseqRun*/SteadyStateOutput//.*gssqRun*/SteadyStateOutput//.stats

Then run the following command:


134

tar -czf sarResults.tar.gzRun*/status Run*/SteadyStateOutput/*/*gsar.gz

Run*/SteadyStateOutput//.infosar1g

Run*/SteadyStateOutput//.infosar10g Run*/SteadyStateOutput//.infoseq

Run*/SteadyStateOutput//.gssq Run/SteadyStateOutput//.stats

NoteNotice that this command sequence retrieves all of the SAR averaging results, including those donewith your initial calc engine run.

For Windows users, zip the sarResults.tar.gz file and send it back to your localmachine over the ftp connection. You can then extract the files using Winzip.Overwrite any old files with newer versions of the same file.

For Mac OS X users, copy the sarResults.tar.gz file to your local machine, changedirectories and extract it using:

tar -xzf sarResults.tar.gz

Refresh results. To make your results available from the EMPro GUI, click Refresh5.in the Results workspace window.

Calculation ConsiderationsThe calculation portion of EMPro may be quite lengthy depending on the application. A fewguidelines are provided here for estimating computer resources, monitoring the progressof the calculation, and avoiding calculation instability.

NoteFor information about defining proper termination criteria to ensure that the calculation has finished, referto Creating a New Simulation .

Computer Resources Estimation

EMPrFDTD will give a time estimate while the calculation is running. This is recalculatedevery time EMPrFDTD updates its status based on how much time passed since the lastupdate and the remaining number of timesteps. It is not a completely accurate estimationsince it does not consider data such as near-field samplings which may only be savedduring certain portions of the calculation. Also, the estimate does not include any post-processing which may occur. Of special note are the SAR averages since they arecomputed in post-processing and may require a significant amount of calculation time.

NoteA quick way to estimate the amount of memory that EMPrFDTD will need for a given problem is to multiplythe number of cells in the geometry by 27. If magnetic materials are included in the geometry (anymaterial with non-free space permeability), multiply by 30, rather than 27. The resulting number is theapproximate number of bytes needed to calculate the project.

Multi-Processing ModulesEMPrFDTD has the ability to do both threaded and Message Passing Interface (MPI)calculations. Both of these capabilities are optional features which may be added to the


135

calculation engine.

Multi-threaded calculations use shared memory and are intended for computers withmultiple processors and/or cores on a single motherboard. The overhead for the multi-threading routines can cause very small calculations to run slower when more than onethread is selected.

NoteIn general, the multi-threaded option should not be used when the number of FDTD cells in the geometryis less than one million.

The number of cells is computed simply from the X , Y , and Z dimensions (in cells) of thegeometry space. The number of threads used for a calculation is defined in the Queuedrop-down box of the Simulations workspace window, as seen below.

Specifying calculation engine threads

Specify the number of processors that are to be used for the calculation under Maximum# Of Threads.

To specify the number of threads from the command line, the "-proc N" option should beused, where N is the number of threads.

NoteFor a summary of command-line options, refer to Running EMPrFDTD From the Command Line.

MPI calculations can be executed on cluster of computers that are connected by anetwork. At present, only Linux computers are supported and each computer must havethe same MPI libraries installed. To run an MPI calculation, see the documentation for theMPI library that is installed on all machines in the cluster. This documentation will giveinstructions on how to start an application using the MPI tools.

Acceleware Hardware AccelerationEMPro has optional Acceleware hardware acceleration. The hardware is available in theform of graphics cards, which can replace existing graphics cards or act from a stand-alone computer. When this hardware is available, it may be enabled within theSimulations workspace window under the Queue drop-down box. To enable it from the


136

command line, the "-hardware" option should be used.

Monitoring Calculation ProgressWhile EMProFDTD is running, its progress will be updated periodically. When launchedfrom the EMPro interface, the progress of the calculation will be printed in the Output tablocated within the Simulations workspace window, as seen in the figure below. Whenrunning from a command line, the progress will be printed to the window that was used tostart the calculation if the "-v" option is used.

Monitoring calculation progress in the Output tab

Calculation StabilityImproper application of the outer boundaries can lead to unstable calculations. Typicallythe stability of a calculation depends on a few simple guidelines involving boundaries andsource placement. Because an absorbing boundary condition like LIAO tries to simulatefree space, it requires that a certain amount of continuity be present in the cells leading tothe boundary. The cross section at a boundary must be the same for at least 10 cells in


137

from the boundary.

NoteFor many problems, a free-space border of 10-20 cells is the best way to ensure stability and accurateperformance of the outer boundary.

Another rule for stability is that no source can be placed within 10 cells of an absorbingboundary. An unstable calculation is easily determined by viewing the line plots of timedomain data or by viewing the field snapshots. When automatic convergence is enabled,EMProFDTD can automatically detect an unstable simulation and terminate it.

In some cases instability can be introduced by a frequency-dependent material. If suchmaterials are used in the calculation and an instability results, it may be necessary tochange the material parameters or reduce the calculation timestep.


138

Viewing OutputIn this section, you will learn about:

How to read and filter your calculation resultsThe available tools to analyze 2-D (plotted) and 3-D (field display) results

Following an EMPro calculation, the results may be reviewed in the Results workspacewindow. The results that are available in this window depend on the characteristics of theproject such as discrete sources, sensors, external excitations, and other project criteriaspecified in the Simulations workspace window.

This section details the review of results available in this window. Some results will be inthe form of numerical values. These are typically single-frequency results performed witha near-zone source. Other results will be displayed in the form of plots. There are severaltypes of plots available to view results based on whether they are time-dependent,frequency-dependent, or angle-dependent. Finally, some results will be available to reviewas colored field displays. Broadband results collected by Surface sensors and Solid sensorsare viewed as individual "Field Snapshots" or "Field Sequences" (strings ofsnapshots). Three-dimensional far-zone fields may also be available for view depending onsimulation criteria.

Results Workspace Window The Results workspace window

The Results workspace window stores all of the data collected by the Sensors during thecalculation. Once the simulation is queued for calculation, the project will be listed in thiswindow so that the results can be viewed. Additionally, the results of any other savedproject can be loaded by pressing the List Project button in the upper-left of the dialog.This makes it possible to view and compare the results of multiple projects without havingto load several projects individually. Similarly, the results of any project may be closed byselecting the project and clicking the Unlist Project button.

While the calculation progresses, several options may be selected to control how data isrefreshed in the Results workspace window as more and more results become available.The list of results in the workspace window may be manually updated by selecting theRefresh button. The list will be automatically updated if Auto-Update Results check box isselected. This check box will also automatically update plotted data when the specific plot


139

window is opened as the calculation progresses.

Filtering Data Results

It is possible to filter the data within the Results workspace window by searching or bycategorizing. The Search box in the top right corner enables you to to search for the nameof any field or data visible within the window. The results will automatically appear in thelist below.

The user is also able to customize the four columns at the top of the window, which filterdata according to the specified categories. The column headings are controlled by right-clicking on any of the current column headings and selecting one of the availablecategories. Each category is described in the table below.

Data Filtering Options

Data Filter Description

Project Id Displays the EMPro project ID, which references the location of the loaded project in thefile directory.

Project Name Displays the project name indicated by the user.

Simulation Id Displays the simulation ID that refers to the simulation's location in the file directory.

Simulation Number Displays the simulation number that is automatically generated based on how manysimulations have been created in a specific project.

Simulation Name Displays the simulation name specified in the Simulations workspace window.

Run Id Displays the run ID which references the location of the run in the file directory.Multiple runs are created in the case that a simulation collects data for more than onevariable or location (i.e., during parameter sweeps, multiple ports, etc.).

Run Number Displays the run number.

Result Type Filters results by type. Any of the following results may be viewed depending on theproject, simulation criteria, and the type of data that was requested:

Electric Field (E)Magnetic Field (H)Conduction Current (Jc)Magnetic Flux Density (B)Poynting Vector (S)Voltage (V)Current (I)ImpedanceAvailable PowerInput PowerInstantaneous PowerFeed LossReflection CoefficientS-ParametersSAR (Specific Absorption Rate)Maximum SAR ValueAverage SAR in Exposed ObjectAverage PowerNet Input PowerNet Feed LossNet Available PowerSystem Efficiency


140

Radiation EfficiencyDissipated PowerDissipated Power in TissueDissipated Power in Non-TissueDissipated Power per MaterialRadiated PowerDissipated Power per Electric Material ComponentDissipated Power per Magnetic Material ComponentDimension (time, position, etc.)Axial RatioRadar Cross SectionGainVoltage Standing Wave Ratio (VSWR)

OUTPUT OBJECT Filters results according to its Output Object , which refers to a specific sensor by name.

DATA TYPE Filters results according to the type of sensor that collected the data. The sensor isreferenced by its general type, rather than by its user-defined name (Refer to OutputObjects for filtering data by sensor name). The following is a comprehensive list of thesensors that may be listed within this filter:

Point Sensor - Retrieves data from any Point sensor in the project.Surface Sensor - Retrieves data from any Surface sensor in the project.Solid Sensor - Retrieves data from any Solid sensor in the project.Far Zone Sensor - Retrieves data from any FAR ZONE sensor in theproject.Raw SAR Sensor - Retrieves raw SAR data.Averaged SAR Sensor - Retrieves averaged SAR data.Circuit Component - Retrieves data from a Ciruit ComponentHAC Sensor - Retrieves hearing aid compatibility data from HAC sensors.System - Retrieves ambient result data (not associated with a sensorobject).External Excitation - Retrieves data on the External Excitation waveform.Raw Steady-State Far-Zone Data - Contains information which can beused to generate new steady-state far zone patterns after a simulation isrun.

Project: Simulation : Run

Displays the Project, Simulation, and Run name in one column.

Domain Filters data according to Time, Frequency or Discrete Frequencies domains.

Field Type Filters data according to total-field or scattered-field.

Status Displays whether a result is complete or still being calculated while the simulation isrunning. This status can be refreshed manually by pressing the Refresh button orautomatically by selecting Auto-Update Results .

Misc Displays query-specific information. For example, for the circuit component voltagewhen collecting S-Parameters, it could contain the active port. For S-parameters, itcould contain the S number or simply the active port. For System numbers, it couldcontain the number (such as efficiency) or net input power.

Viewing Numerical ResultsWhen a single-frequency calculation has been performed with a near-zone source (voltageor current), parameters such as input impedance, S-parameters, VSWR and the reflectioncoefficient are displayed as numerical values rather than as line plots since the data onlyis relevant for the input frequency. Other numerical values are collected by means of asystem sensor, which is automatically present in every EMPro project.


141

Numerical Data Collected by EMpro

Results Displayed as Numerical Data

Sensor Type Time Dependence Result Type

Port Sensor Single-Frequency Available Power

Current

Impedence

Input Power

Reflection Coefficient

S-Parameters

Voltage

VSWR

System Sensor Single-Frequency Dissipated Power

Dissipated Power in Non-Tissue

Dissipated Power in Tissue

Dissipated Power Per Electric Material Component

Dissipated Power Per Magnetic Material Component

Dissipated Power Per Material

Net Available Power

Net Feed Loss

Net Input Power

Radiated Power

Radiation Efficiency

System Efficiency

Impedance

All input impedances are calculated by the ratio of the complex V over complex I for theFDTD mesh edge at the port location. The sign convention is positive for power flow intothe antenna or other structure. Thus for a port which delivers power to the antenna, theimpedance will have a positive real part, and for a port which absorbs power from theantenna, the impedance will have a negative real part.

When only one active port is present, the feed point impedance at that port is the self-impedance at that port. If more than one port is active, the port impedance values listedrepresent the ratio of the complex voltage and current at each port including the effects ofthe sources at all active ports. Thus these impedance values are not the self-impedancesat each port, but rather terms of an impedance matrix.

S-Parameter Calculations

When an S-parameter calculation is made with the steady-state far-zone transformationselected, then both the steady-state antenna data and the S-parameter data will bedisplayed. This may be useful, for example, when making calculations for a microstripantenna when both S11 and input impedance are of interest.


142

Net Available Power

The available power is calculated as the total power delivered by the active port or portsinto a matched load.

Radiated Power

The radiated power is computed as the difference between the total net input powerdelivered by the active ports and the dissipative losses from conductive materials andresistive loads in inactive ports.

System Efficiency

The system efficiency is calculated as the ratio of radiated power to available power. Thus,it includes radiation efficiency and mismatch efficiency.

Power Scaling Factor

When viewing single-frequency results collected by the system sensor, the user canspecify a scaling factor of the available power to the calculation and determine the scaledresult. The absolute input power can be adjusted as desired or an overall scaling can beapplied. For example, if the user wanted to know what the output power would have beenwith an input of one mW, they can enter this value in the available power (will be subjectto mismatch loss) or in the net power (after mismatch loss) to see the overall effect.Similarly, any of the output results may be scaled to determine the effect on input andoutput. Clicking reset will return the values to the un-scaled state.

Input power, Input Impedance, and Loss

When making calculations that include the input voltage, current and/or power in thecalculation formulas, such as antenna gain or input impedance, the input voltage, currentand/or power will be that provided at the terminals of the mesh edge. Referring to thefollowing figure, the impedance at the port would be the (complex) mesh edge voltage Vdivided by the (complex) mesh edge current I. The complex values would be obtainedfrom an FFT for a broadband calculation or from two samples of the voltage and current(electric and magnetic fields at the port mesh edge) for a sine wave excitation. If the portis delivering power, then the impedance at that port will have a positive real part. If theport is absorbing power, then the impedance at that port will have a negative real partand the input power will be negative. This is determined by the direction of current flowand voltage polarity for the FDTD mesh edge.

Feed schematic, including FDTD mesh edge voltage, V, and current, I


143

Active ports (those with active voltage or current sources) are treated differently thaninactive ports for some calculations. For an antenna calculation, the input power to theantenna is the algebraic sum of all powers delivered by active ports. Power absorbed byactive ports will not reduce the antenna efficiency; however, power absorbed by inactiveports will reduce antenna efficiency.

To clarify this, consider two different situations. An antenna composed entirely of perfectconductor includes 2 ports, the first containing a 1-V source and a resistor, the second apassive port containing only the resistor. Assuming that some current flows in the passiveresistor, the antenna efficiency will be less than 100%. If we repeat the same calculation,but with a 0.00001-V source added to the formerly passive port, and with both sourcesset active, the antenna efficiency will now be calculated as 100% even though the0.00001-V source will have negligible effect on the antenna currents and radiation. Thisdiscrimination is done so that active ports may utilize lumped circuit elements to match toan antenna without changing the antenna efficiency, impedance, and gain, while passiveelements may be added to an antenna with their effects included in the antenna efficiencyand gain results.

Similarly, for input impedance calculations, the source resistance, capacitance andinductance values will not be included in the input impedance. For example, the inputimpedance of an antenna as calculated using EMPro should not change regardless of anychanges in the active port components (source/R/L/C). This is as it should be, since theantenna impedance is a function of the antenna geometry/materials and not of how theantenna is fed.

Viewing 2-D Plotted Results

Some results are viewed as 2-D plots. There are three basic categories of plots in EMPro,depending on the X -axis (abscissa) type:

XY PLOTS - visualizes data in a Cartesian coordinate system.POLAR PLOTS - represents data with an angle-independent axis.SMITH CHARTS - displays complex data vs. frequency, such as S-parameters andreflection coefficient.

The 2-D plots are viewed by right-clicking on the desired result type in the Resultsworkspace window and selecting Create Line Graph, as seen in the following figure.

Results right-click menu


144

Before the graph is displayed, the user has the option of adjusting the graph propertiessuch as the component, data transform, and complex part. The Target Graph optionenables you to view and edit plots that were previously created (from the same dataselected in the Results window).

The following figure shows samples of the create graph dialog for each plot.

Create graph dialog for XY, Polar, and Smith plots

For all calculations, the most important quantities are the time-domain plots ofthe fields in the problem space. Always perform a quick review of these valuesto ensure that a calculation has converged. Without convergence, most otherresults will be meaningless, particularly any plots converted to the frequencydomain such as S-parameters or impedance.

NoteNote that input impedance and S-parameters may be plotted in rectangular form vs. frequency or as aSmith chart.

Results Displayed as Plotted Data

Plotted Data Collected by EMPro


145

Sensor Type Time Dependence Result Type

Port Sensor Broadband Current

Impendence

Input Power

Instantaneous Power

Reflection Coefficient

S-Parameters

VSWR

Point Sensor/Surface/Volume Broadband E-Field (E)

H-Field (H)

B-field (B)

Conduction Current (Jc)

Poynting Vector (S)

System Sensor Broadband Net Feed Loss

Net Input Power

Far Zone Sensor Broadband Radar Cross Section

Gain

E-Field (E)

Far Zone Sensor Single-Frequency Radar Cross Section

Axial Ratio

E Theta, E Phi

Circular Polarization

Ludwig-2 Az, El

Ludwig-2 Al, Ep

Ludwig-3

Far Zone Post Processor Single-Frequency E-Field (E)

Customizing Plots

There are several ways to customize plotted data after it is opened from the Resultsworkspace window. The following sections detail the various tools that are available tomodify plotted results.

NoteKeep in mind that the tools available within the plot window depend on the type of graph you are viewing(i.e., XY Plots, Polar Plots, or Smith Charts).

Export Data

Select this option to export graphical data point values to a text file in a specifieddirectory.

Export Image Tool

Select this icon to save an image of the current plot to a specified directory.

Pan Tool


146

Select this tool, and drag the mouse within the plot to pan to the desired view of the plot.Press the |Ctrl| key and drag to pan along the independent axis. Press the |Shift| keyand drag to pan along the dependent axis.

Zoom Tool

Select this tool to zoom-in and to zoom-out of the plot. The mouse wheel as well as theright and left mouse buttons, can be used to perform zoom operations as described below.

Using mouse-wheel:

Roll the center wheel of the mouse forward to zoom-out of both axessimultaneously.Press |Ctrl| and roll the center wheel forward to zoom-out of the independentaxis.Press |Shift| and roll the center wheel forward to zoom-out of the dependentaxis.Roll the center wheel of the mouse backward to zoom-in to both axessimultaneously.Press |Ctrl| and roll the center wheel backward to zoom-in to the independentaxis.Press |Shift| and roll the center wheel backward to zoom-in to the dependentaxis.

Using mouse buttons:

Right-click and drag the the mouse anywhere in the plot window to zoom-out ofboth axes simultaneously.Press |Ctrl| and right-click/drag to zoom-out of the independent axis.Press |Shift| and and right-click/drag to zoom-out of the dependent axis.Left\click and drag to define a rectangular view-window in the plot window tozoom-in to both axes.Press |Ctrl| and left-click/drag to zoom-in to the selected domain ofindependent axis values.Press |Shift| and and left-click/drag to zoom-in to the selected range ofdependent axis values.

Legend Visible

This button toggles the display of the legend within the graph.

Graph Properties Tool

The following illustration highlights the three tabs that are available for editing theproperties of any graph in EMPro. Each tab is detailed below.

Tabs available for editing graph properties


147

Title properties

Define the graph name and title as well as the background color of the graph in this tab. Acheckbox also toggles the legend display on and off. This tab is displayed in the figureabove.

Axes properties

Define the title of the axes and the limits of the axes in this tab. The Auto checkbox maybe selected to auto-select these limits.

NoteIf a graph only contains continuous data, the X -axis must be specified (this will not happenautomatically).

The Units drop-down menu is used to specify the units and to apply a log scale ifnecessary.

For certain Smith plots, you will also have the option of modifying the ReferenceImpedance.

The figure below shows the axes properties editor for a 2-D XY graph.

Editing the axes of graphs


148

Plot properties

Define the characteristics of the plotted lines in this tab. A list of every dependent variableis listed with a customizable line color, line width, and line style. Any unwanted variablescan also be deleted in this tab.

Selection tool

Select this tool to move, delete, or edit a marker's properties. To move a marker, click onthe marker with the selection tool (once selected, it will turn yellow) and roll the mouse-wheel forward or backward to move an attached marker along its plot. Pressing |Ctrl| androlling the mouse-wheel will speed up the movement of the marker.

To delete any marker, simply select the unwanted marker, right-click and press DeleteMarker . All markers may be deleted at one time by right-clicking anywhere in the plotarea and choosing Delete All Markers .

To edit a marker's properties, click on the desired marker so that it turns yellow, thenright-click and select Marker Properties . A window will appear with several differentediting options. The location coordinates of the marker can be adjusted by manuallytyping in the desired values in the Requested Location section of the dialog box.

Marker properties dialog

NoteIf the coordinate box you desire to edit is disabled, select the appropriate option in the Attached plot drop-down menu.

The marker may also be attached to a particular plot by selecting its name in the AttachedPlot drop-down menu and selecting an Interpolation Method for EMPro to use in order toplace the point. Depending on this definition, the marker will be shifted to the nearestpoint on the selected function or linearly interpolated based on the independent axis thatis entered by the user in the Requested Location dialog box. Finally, the type of markercan also be redefined in this editor window in the ) Type drop-down list.

Additionally, this tool can be used to move or close the legend in the graphical space. (Anytool, however, can be used to perform this function.)


149

Select this tool by clicking on its icon or select it from the Markers menu, as shown inbelow.

Marker drop-down menu

Point Marker Tool

Select this tool to mark any point on the plot by clicking on the desired marker location. Amarker with the location coordinates will appear above the point, depending on the type ofplot:

XY Plot: (X -location, Y -location)Polar Plot: (radius, angle)Smith Plot: (real part of location, imaginary part of location, frequency)

When the mouse moves close to the plotted curve, it is snapped to the closest location onthe interpolated line or sampled point. Holding the |Ctrl| key will disable the snappingaction, allowing a point to be placed anywhere. Holding the |Shift| key will snap themarker to sampled points only. Note that markers placed on sampled points are blue, andmarkers placed on interpolated points are black.

Crosshair Marker Tool

Select this tool to mark the location of a single point by two intersecting cross-hairs. Themarker is placed at the right edge of the plot. (Snapping actions are the same as thePoint/Tracker Marker described above.)

The following figure shows an XY Plot with a Crosshair Marker (1.8873e-08 s, -6.0381e-06 A) and Point Marker (1.4636e-08 s, 1.0573e-05 A).

2-D XY graph with Crosshair and Point markers (highlighted in red)


150

The figure below shows a polar graph with a Crosshair Marker along a radius.

Polar graph with Crosshair marker (highlighted in red)


151

The following figure shows a smith graph with a Crosshair Marker along a radius.

Smith graph with Crosshair markers (highlighted in red)


152

Vertical Marker Tool

Select this tool to place a vertical line that intersects with the X -axis. The marker (Y = constant) will be placed along the top-edge of the plot area. (Snapping actions are thesame as the Point/Tracker Marker described above.)

Horizontal Marker Tool

Select this tool to place a horizontal line that intersects with the X -axis. The marker (X = constant) will be placed along the right-edge of the plot area. (Snapping actions arethe same as the Point/Tracker Marker described above.)

The figure below shows a 2-D XY Graph with a Vertical Marker (at X = 1e-08 s) as well asa Horizontal Marker (at Y = 1.5e-05 A).

2-D XY graph with Vertical and Horizontal markers (highlighted in red)


153

3-D Field DisplaysColored 3-D field displays may be viewed in slices of the geometry by saving eitherbroadband or single-frequency field quantities. The fields may be viewed in with thegeometry (solid or meshed), or by themselves.

Three types of fields may be viewed with the geometry:

Time Domain Snapshots - available for any calculation as they are simply snapshots of the near-zone fields at specific steps in time.Complex Fields or Derived Quantities (such as SAR) - available at specificfrequenciesThree Dimensional Far-Zone Fields - can be requested either before thecalculation with a far-zone sensor, or after a calculation through post-processing.

Data Collected With Near-Field Sensors

Field snapshots are listed in the field control panel as either single slices or as fieldsequences. The field sequences are movies of the individual slices as the fields progresswith time in a particular slice of the geometry. Depending on what was saved, each fieldsnapshot may have electric and magnetic fields, current densities, Poynting vectors storedfor each direction (X , Y and Z ) and a display of the combined magnitude. Field snapshotsare collected by surface sensors and solid volume sensors for broadband data.


154

NoteIn order to view single-frequency data, the Collect Steady-State Data check box must be checked in theSimulations workspace window (under the Frequencies Of Interest tab).

Results Plots Available For Each Sensor Type

Sensor Type Result Type Time Domain Discrete-Frequency Broadband

Point Sensor E-Field (E) X X

H-Field (H) X X

B-field (B) X X

Poynting Vector (S) X X

Conduction Current (Jc) X X

Scattered E X X

Scattered H X X

Scattered B X X

Average Power X

Surface/Solid Sensor E-Field (E) X X

H-Field (H) X X

B-field (B) X X

Conduction Current (Jc) X X

Scattered E X X

Scattered H X X

Scattered B X X

Poynting Vector (S) X

Average Power X

SAR Sensor SAR (Specific Absorption Rate) X

HAC Sensor E-Field (E) X

H-Field (H) X

HAC max E-Field (E) X

HAC max H-Field (H) X

(Scattered) Electric Fields (E): magnitude, normal, or X , Y or Z components of the(scattered) electric field data at each cell edge.

(Scattered) Magnetic Fields (H): magnitude, normal, or X , Y or Z components of the(scattered) magnetic field data at each cell edge.

(Scattered) Magnetic Flux Density (B): magnitude, normal, or X , Y or Z componentsof the (scattered) B-field computed from the magnetic fields at each cell edge and theassociated permeability for that cell edge.

Average Power Density (SAV): magnitudes of the average power density computedfrom the electric and magnetic fields at each cell edge.

Conduction Current Magnitude (Jc): conduction current at an electric field cell edge.

SAR (Specific Absorption Rate): computed for each complete cell containing a lossydielectric with a non-zero material density.


155

Data Collected with Three Dimensional Far-zone Fields

The Results workspace window also displays the data collected within 3-D far-zone fields.Any 3-D far-zone request generated using EMPro's post-processing engine (run separatelyfrom the calculation engine) is automatically added once the post-processing is complete.If steady-state far-zone data is enabled for the sensor, the 3-D results will include the E-field and axial ratio in the discrete-frequency domain. Gain or discrete-frequency radarcross section (RCS) will also be available when using a feed or external excitation,respectively.

NoteNote that the polarization (Theta/Phi, Ludwig-2, etc.) is selected through the Setup tab of the Field EditingToolbar.

Viewing 3-D Field Displays

EMPro displays 3-D field data in the Geometry workspace window. The followingsubsections discuss how to configure and analyze the display, using the Scale Bar, theField Reader Tool, rescaling, and the field editing toolbar.

The Scale Bar

The Scale Bar, located at the top of the Geometry workspace window, "paints" the viewwith a range of colors which correspond to the range of values displayed. By default, theScale Bar color palette is shown in continuous mode within a default range of values. Youcan adjust these properties by right-clicking on the scale bar and selecting Discrete Mode,to change the palette to discrete colors, or Automatic Range, to change the range ofvalues to that which is actually present.

There is also a Properties option under the right-click menu. Clicking this will bring up theScale Bar Editor. Using this editor, you can manually set the Scale Bar limits, units andcolors to your preference. Take note of several settings that may not be intuitive for thefirst-time user:

Under the Limits section, it is possible to "clamp" values outside the defined scalebar limits to the nearest color. This makes it possible to view outliers that otherwisewould not be colored with the Scale Bar.Under the Scale section, when Relative dB is selected, you can define its referencevalue or use the value automatically selected by EMPro.Checking Use Discrete Colors makes it easier to view contours in the 3-D results.Font Color and Background Color change the colors of the Scale Bar itself, not thecoloration of the results.

The following two figures show the Scale Bar Editor, and the Scale Bar_ with the FieldReader tool.

The Scale Bar Editor


156

The Field Reader Tool

The Field Reader Tool is located in the toolbar to the right of the Geometry workspacesimulation space. After selecting this icon, wheel the mouse over the geometry object toidentify its field values. A marker in the Scale Bar will display the nearest known fieldvalue to the location of the mouse. This location is represented by a small dot on thescreen.

The Field Reader tool


157

Rescaling

In the Results workspace window, filter the results by the output object System. Double-clicking on results such as Dissipated Power or System Efficiency will bring up the SystemSensor Output dialog. Changing any value in this table will rescale the other valuesshown.

NoteThe Show Scaled Values box must be checked to enable editing in the System Sensor Output dialog.

Scaling only affects results which are in the discrete frequencies domain.When you change the scaling by editing values in the System Sensor Output dialog,you only scale the results for that particular calculation engine run at that particularDFT frequency. Any other tables of data and plots associated with this run andfrequency will also be affected.

System Sensor Output dialog


158

The Field Editing Toolbar

Located at the bottom of the Geometry workspace window, the field editing toolbar is usedto configure the properties of the view. In the upper-left corner of this toolbar, a drop-down list will display any view that you have opened from the Results window. You canuse the Hide Others and Unload buttons to single out certain view(s) if necessary. Thefollowing figure shows a drop-down list of such results in the field editing toolbar.

The results drop-down list of the field editing toolbar

The tabs and configuration options below will change depending on the active view. Acomprehensive list of the available options are described below.

The Setup Tab

There are two main configurations for the Setup tab, depending on the type of sensor youare evaluating.

Keep in mind that not all fields listed will be available for every view.


159

Surface, Solid SAR Sensor Configuration

Sequence Axis - Select the axis along which to view the sequence (from theSequence tab).Display Mode - The available display modes are Vector Field, Point Cloud, Flat, and3D.Size Factor - Used to scale the size for point clouds and vector display modesDisplayed Field - Lists the available components of a vector field that can be shownin a plot.Decimation - Used to sample a subset of points for solid sensorsSurface Resolution - Used to sample a subset of points for Surface sensors. Thiscontrol is only available if the Surface sensor sampling method is configured as FieldInterpolation.Complex Part - Used to single out real/imaginary/magnitude values for single-frequency dataAxis Ranges - Set the independent axis during the sequence

Axis - defines the axisFull - runs along the full range of the axisMin And Max - subsets non-sequence axes

Enable Scaling - enable the scaling factor

The Setup tab for a solid sensor configuration

Far-Zone Sensor Configuration

In the first section, set up your viewing preferences.Viewing - choose among the available result typesComplex Part - view the Magnitude, Real, or Imaginary part of the results.Size - enter the size of the maximum radius.

In the second section, choose the orientation of the Center point.Enter the (X, Y, Z) coordinates of the center, orChoose the center using the Pick tool.The Reset Center On drop-down enables you to center the plot at the center ofthegeometry or at the location of any of the ports in your active project.

Select the following viewing options:Display Axes Theta/PhiDisplay Main Lobe Direction (the direction of maximum gain)Show Scaled Values

The Setup tab for a far-zone sensor configuration


160

The Sequence Tab

This tab is available when viewing an SAR, Solid, or Volume sensor. Configure the settingshere to "play through" a simulation.

Minimum/Maximum - values entered here set limits on the range of the sequenceaxis.Showing - shows the currently plotted value of the sequence axis.The scroll bar at the bottom of the window can be used to view any point along thesequence.The simulation buttons enable you to move through the sequence:

The forward and back buttons play the sequence forwards and backwards,respectively.The pause button pauses a playing sequence.The fast forward and rewind buttons fast forward and rewind the sequence,respectively.The jump forward and jump backward buttons jump to the end or beginning ofthe sequence, respectively.

Compute Bounds - when this box is checked, EMPro will compute the bounds duringthe sequence.- performs the Compute Bounds calculationThe Advanced tab

Checking the Auto-Repeat Sequence box will automatically replay thesequence continuously.The Step Size determines the number of indices to change for the simulationbuttons.

The Sequence tab of the field editing toolbar

The Rotations TabThe Rotations tab, shown in Figure 13.20, oers several operations to adjust the orientationof the far-zone results by setting the Up Vector of the view. Changing the Up Vector(which is the Z-axis by default), will set the reference point for the spherical coordinatesystems (e.g., theta/phi, alpha/epsilon, etc.) This affects field plots of single polarization


161

components, partial power effciency and power computations, mean eective gaincomputations, and antenna diversity. This is useful incases where the geometry is aligned with an axis of the computation grid, but the real-world position of the geometry is not axis-aligned.

The Up Vector Presets list enables you to select a pre-defined Up Vector in the X-, Y-, or Z -direction.The Pick Up Vector tool enables you to click and drag within the simulation space tocreate your own Up Vector.The Undo button will undo an operation.The Redo button will redo an operation.The Reset button will reset the Up Vector.

The Rotations tab of the field editing toolbar

The PDF Tab

The PDF tab controls the settings used in the computation of mean eective gain andantenna diversity far zone results. The following illustration shows the PDF tab.

XPD is cross-polarization discrimination of the incident multipath field. It can beexpressed by the ratio of time-averaged vertical power to time-averaged horizontalpower in the fading environment. Users can input a linear value or a decibel value.PDF, or Probability Density Function, is used to model communication channels. Thisfunction "weights" different directions in the far zone sphere such that certain farzone direction are taken into account more than others in the mean effective gainand antenna diversity calculations. It has a maximum value of one and a minimumvalue of zero. EMPro provides the following choices for PDF's:

Uniform Value: in this case, PDF is simply one.Gaussian: this is most commonly used as a model of channels. Theta Max andPhi Max are the angles of the maximum incoming field. Sigma Phi and Sigma

Theta are the standard deviations for the and directions, respectively.User-Defined: this enables you to import a PDF from a text file. The file mustcontain a grid of theta/phi angles and the corresponding probability density ateach angle. For example, to define the PDF at ten theta angles with twenty phiangles, the text file data would look something like the following:

theta0radians phi0radians pdfAtTheta0Phi0


...




162

...


The PDF tab in Gaussian mode

The Statistics Tab

There are two main configurations for the Statistics tab, depending on the type of sensoryou are evaluating.

NoteKeep in mind that not all fields listed will be available for every view.

Far-Zone Sensor Statistics

The next figure shows the Statistics tab for a far-zone sensor configuration.

When selected, Power/Efficiency gives the user the option of choosing from among thefollowing radiation patterns:

• Full Pattern - specifies the use of a full far zone sphere.• Upper Hemisphere - specifies the use of theta = [0, 90o] and phi = [0, 360o].• Open Sky - specifies the use of theta = [0, 80o] and phi = [0, 360o].• Partial Pattern - enables you to specify an arbitrary solid angle. These angles are definedrelative to the coordinate system defined under Rotations.

Each pattern will display its statistics to the right. In the case of Partial Pattern, the usercan redefine the min and max values of Phi and Theta by typing them in to thecorresponding boxes. The statistics will update accordingly.

The Other Statistics option presents a list of additional statistics.

The Statistics tab for a Far-Zone sensor, showing Power/Efficiency and Other


163

statistics

SAR Sensor Statistics

The next illustration shows the Statistics tab for an SAR sensor configuration.This tab is available when looking at SAR data for both the SAR Sensor and the SARAveraging Sensor. The following figure displays the dialog that appears when the View AllSAR Stats button is pressed.

The SAR Maximum Value displays a max SAR value dependent on the properties ofthe SAR Sensor, (i.e., the maximum 1-gram Averaged SAR value, the maximum 10-gram Averaged SAR value, or the maximum raw SAR value). A value entered hereaffects the power scaling for this run and frequency, in the same manner as adjustingthe Power Scaling Factor in the System Sensor Output dialog.SAR Maximum at Location displays the global coordinates of the location where theSAR maximum value is found.The Avg. SAR in Exposed Object field is available when looking at raw SAR sensordata only. It displays the whole body average SAR value. An adjustment enteredhere aects the power scaling in the same way as adjusting the SAR Maximum Value.The View Slice With Max Value button takes you to the planar slice of SAR data whichcontains the maximum value.The View All SAR Stats button opens a separate window displaying the statistics forthe Raw SAR, 1g Averaged SAR, and 10g Averaged SAR together in a table. Throughthis window, the scaling factor can be adjusted, which affects all SAR values in thiswindow, including data displayed in the SAR Sensor Statistics tab. The scaling factorand SAR statistics values will also automatically update when the user manuallychanges (rescales) any other statistics available through the Results workspacewindow, such as power and efficiency values.

NoteFor more information on rescaling values in the System Sensor Output dialog, refer to Rescaling.

The Statistics tab for an SAR sensor


164

The complete list of SAR statistics

The Diversity Tab

The Diversity tab is used to compute antenna diversity metrics between two far zonepatterns. To perform the diversity computations, you must load the data for both far zonepatterns in the Results workspace. Then go to the diversity tab for one of the twopatterns. The following figure shows the Diversity tab.

The Pattern 2 drop-down list enables you to select the far zone result which yourcurrent far zone result is compared against. The only choices listed here are resultswhich have been loaded into the Results workspace window.The statistics next to From 3D Pattern show the diversity computation for the full farzone sphere. The statistics next to From 2D Pattern show the diversity computationat theta = 90o.Under Diversity Options, you can set the Antenna 1 Phase and Antenna 2 Phasecontrols to specify the phase difference between the two far zone patterns.

NoteNote that the XPD and PDF settings under the PDF tab impact this computation, as well as thesettings under the Rotations tab.

The Diversity tab of the Field editing toolbar


165

The Hearing Aid Compatibility TabThis tab, as the name suggests, is only available when viewing (HAC) sensors.

Frequency - Select the frequency of interest for the hearing aid.Display Mode - The available display modes are Point Cloud and Flat.Show Scaled Values - When checked, the displayed results are scaled by thescaling factor entered by the user on the system sensor output table.Surface Resolution - This option appears whenever there is a HAC or interpolatedsurface sensor. It controls the resolution at which field data is shown for the sensor.The higher the resolution setting, the more data points are shown on screen (and thecloser they are together).M-Rating - Displays the M-rating of the sensor, or the suitability of the hearing aidused with a wireless device. The rating range is 1-4, with 4 being the best rating.Threshold - This is the maximum allowed field value given the current M-rating.Band - Displays the band type. Under the Edit button, you can configure theproperties of the band, or define a pre-set list of properties using the Manage Presetsbutton. An editor dialog will appear where you can add a new preset or importpresets from a text file. Each line in the file must follow this format:

Lines in the file starting with # or ! are comments and are ignored.Otherwise, the line must contain six semicolon-delimited items in this order:name; waveform modulation factor; E-field probe modulation factor; H-fieldprobe modulation factor; articulation weighting factor, in dB (must be either 0 or-5); and a boolean value which determines whether RMS conversion should beused (1 to enable, 0 to disable).

An example of this format is as follows:

! This is a comment

Band 1 Name;0.85;0.9;0.8;-5; 0

Band 2 Name; 1.0; 0.92; 0.33; 0; 1

The 3x3 Grid Region - This shows the maximum field values in each of the 9 squaresof the HAC grid. The double-line on the top-level of the readout helps you orient inreference to the white grid on the displayed field (it also has a double-line in the top-left corner). Three of the cells in the 3x3 grid will have a dotted red border aroundthem. These are the three cells chosen to be excluded from the HAC ratingcomputation as allowed by the IEEE HAC standard. Also, any cells with red text arecells which exceed the field value threshold amount for the current M-rating.

The Hearing Aid Compatibility tab of the field editing toolbar


166

Post-Processing

Post-processing gives the user the option of running additional computations after thecalculation engine run is finished to save certain results to disk. Two types of post-processing are available.

Far Zone Post-Processing

Far-zone post-processing is enabled by checking the Save Data for Post-simulation FarZone Steady-state Processing box under the Frequencies of Interest: Data Storage tab ofthe Simulations workspace window. When this configuration is set, there will be an entryfor Raw Steady-State Far Zone Data in the Results workspace window. Double-clickingthis entry will enable you define a new far zone geometry at which the steady-state farzone pattern will be computed. Edit the geometry exactly as you would edit a far zonesensor. When you click Done in the far zone sensor editor, the computation of the farzone pattern will begin. Once complete, you can find a new result in the Resultsworkspace window which corresponds to the newly defined far zone pattern.

SAR Post-Processing

Right-clicking on any Raw SAR Sensor entry in the Results workspace window will show anoption in the context menu to Postprocess Results. This enables you to perform SARaveraging. Edit the geometry and averaging parameters exactly as you would in the SARAveraging Sensor editor. Once you click Done in the editor, you will be given the option toautomatically run the post-processor. If you say yes, the post-processing operation isqueued as if you had created a new simulation. Otherwise, you will have to do the post-processing run manually. Once the SAR averaging is complete, new SAR average data willappear in the Results workspace window.


167

Additional Tools for Customizing andOrganizing ProjectsIn this section, you will learn how to:

Use scripts to streamline tasks in EMProDefine your own universal parametersStore object definitions externally to reuse them in multiple projectsOrganize objects in your EMPro project for more convenient access

EMPro provides several tools to facilitate the creation and organization of projects.

Parameterization is a powerful tool that makes it easy to define variables andfunctions in one convenient workspace window, which can be referenced anywherewithin the EMPro interface. Additionally, Parameterization can be used in conjunctionwith scripting to sweep through a series of parameters (i.e., multiple antennalengths) to run a calculation at every swept point. In EMPro, parameters are definedin the Parameters workspace window.

The Scripting workspace window makes it convenient to write scripts to accomplishtasks that are specific to an EMPro project. It provides users with the ability to createfully-customizable functionality within the EMPro interface that is specific to their owntasks. Scripting may be useful for quickly performing repetitive tasks, referencingexternal files, employing a series of modeling operations at once, or virtually any taskthat is tedious with the standard EMPro tools.

Libraries provide useful means of storing definitions and any types of objects createdwithin an EMPro project. They are saved in the Libraries workspace window as filesthat are not attached to a specific project so that they can be referenced again andagain. They are very useful for creating new projects that reuse definitions andobjects from past projects. Rather than having to rebuild a project from scratch,pertinent libraries can be imported so that time is not wasted redefining similarobjects and properties.

EMPro also provides various features that help in grouping and organizing definitionsand objects. The Groups branch of the Project Tree functions to store shortcutgroups, which are groups of objects that are added and organized by the user.Similarly, Assemblies are user-defined groups of geometric objects that are added tothe Parts branch of the Project Tree. They are convenient, especially for projects thatcontain a large number of parts, so that objects in the tree remain organized andeasy to access.

Parameters Workspace WindowThe Parameters workspace window enables you to define parameters that can be used inother places in the project to parameterize their project. A parameter can be referenced inany dialog box in EMPro. It can have a simple numeric value or a mathematicalexpression, and can reference other parameters.


168

The Parameters workspace window can be accessed at any time by right-clicking in thetabbed workspace or selecting the window in the View menu of the Application Menu Bar.

The Parameters workspace window

Defining ParametersThe Parameters workspace window contains four fields: Name, Formula, Value andDescription. Note that a value named 'timestep' is already present upon opening thiswindow.

A new parameter is added by clicking the button above the table of parameters. A newline with default values will be added to the list of parameters. These values can edited bydouble-clicking on any value. The |Tab| key will scroll through the columns of the tableand the |Escape| key will cancel any changes that have been made to a parameter entry.

The Formula column is where a user will input a mathematical formula or a simplenumeric value that will define the value of a given parameter. This formula can referenceother parameters that have already been defined.

The Value column is a read-only column that displays the evaluated value of theparameter. If an invalid formula is entered, an error message will appear within this fieldwith a description of the invalidity. Simply hold the mouse over the error message to viewthis description.

Similarly, a parameter is deleted by selecting the unwanted parameter and clicking thebutton above the table. If a parameter is deleted that is referenced within anotherparameter's definition, an error message will appear since the parameter that isreferenced is no longer defined.

Each parameter is referenced by its assigned name defined in the Name column of thiswindow. For instance, a parameter named "length" can be referenced at any time bytyping "length" into any dialog box within EMPro, and it will assume this defined value.


169

NoteA sweep over a parameter may be set up in the Simulations workspace window. For more information,refer to Simulations Workspace Window.

Scripting Workspace WindowThe Scripting workspace window enables you to create, edit, manage, and execute user-defined scripts, which are capable of gathering and reporting information from the EMProproject or making changes to the project. Scripts are blocks of Python scripts that useEMPro as module. They are typically used to automate repetitive or tedious tasks (thatcould otherwise be done through the EMPro GUI) with greater speed and precision.

The Scripting workspace window

Parameters that have been defined in the active parameter list of the Parametersworkspace window may be referenced within any script. A consequence of this ability isthat changing a parameter in the parameter list may change the behavior of user-definedfunctions.

There are a dozen of example scripts included. You can launch the demo by using themenu item Help > Scripting Demo.

Demo Scripting Window


170

EMPro Scripting Language

All scripts in EMPro are written in the Python scripting language.

NoteFor more information on Python, visit the Python Software Foundation homepage athttp://www.python.org or use the Help functionality within EMPro.

New Python Script

Select this icon to bring up a New Python Script tab where a new Script can be defined.The script is not executed until the Execute Python icon is pressed.

Commit Script

Select this icon to commit a change to a script after an edit.

Revert

Select this icon to "revert" or abandon an edit. Any changes that have been made in theeditor will be lost, and the editor will revert back to the script that was stored in the EMProproject or application.

Clear Output Window

Select this icon to clear the text in the output window where error messages and scriptoutput are written.

Search and Replace

Select this icon to search for text within scripts. By selecting the Replace With check-box,

http://www.python.org

http://www.python.org


171

text can be located and replaced with the desired text.

Execute Script

Select this icon to execute a macro from one of the user-defined scripts.

Libraries Workspace WindowThe Libraries workspace window enables you to create libraries or collections of objectsgrouped by category so that they can be easily referenced during a project and can beaccessed in subsequent projects. This makes it very easy to access commonly usedobjects and definitions so that they do not have to be recreated during every project.

The Libraries workspace window

Creating a New Library

To create a new library directory or subdirectory, click on the button above the Librariesspace in the workspace window. Specify the name of the location to store the new libraryfile.

Accessing Existing Libraries

To access an existing library, click on the button, navigate to the appropriate directory,and select the desired library to load into the project.

Adding Objects and/or Definitions to a Library

To add objects of definitions to a library, simply drag the desired object from the Project


172

Tree into lower workspace entitled Object/Notes. The object will be placed in a librarygrouping that corresponds to its original position in the Project Tree. For example, aMaterial dragged from the Project Tree will be placed in the Materials folder of the Library.Additionally, filters can be applied to library objects to control the visibility of each group.They are controlled in the Filters section of the Libraries workspace window.


173

Appendix of Geometric ModelingIn this appendix, you will learn how to use the geometric modeling tools available withinEMPro.

Editing Cross-Sections for 2-D and 3-D Models

Shapes

Edge tools

Edge tools are used to create lines of various shapes within the EMPro interface. Thefollowing figure displays the Edge Tools including the Straight Edge tool (upper left),Polyline Edge tool (upper right), Tangent Line tool (lower left) and Perpendicular Edge tool(lower right).

NotePressing |Tab| while using these tools will bring up the Specify Position dialog, which is used to enterrelevant properties to the tool being used.

The Edge Tools

Straight Edge

Creates a simple straight edge. To use this tool, click the Straight Edge button and clicktwo points in the sketching plane where the endpoints should be located.


174

Polyline Edge

The Polyline Edge is similar to the Straight Edge tool except it allows multiple points tocreate a series of connected straight edges. Click a starting point in the sketching planeand continue clicking on the locations of subsequent endpoints to create desired polylineedge. Click on the first vertex or press |Return| to finish.

Perpendicular Edge

Creates a straight edge perpendicular to an existing edge. To use, select the PerpendicularEdge button and click on the existing edge that will define the perpendicular direction.This can be a straight or curved edge. Then click on the location of the first and secondendpoints of the perpendicular straight edge.

Tangent Line

Similar to the Perpendicular Edge tool, but instead draws a line tangent to a pre-existing,non-linear edge. To use, select the Tangent Line tool, and click on the existing curve thatwill define the tangential direction. Then click on the location of the first and secondendpoints of the tangential straight edge.

Closed Polygon Tools

The following illustration displays the Closed Polygon tools including the Rectangle,Polygon and N-Sided Polygon tools.

The Closed Polygon tools

Rectangle

Creates a simple rectangle. Click the desired location of the first vertex of the rectangleand drag the mouse to the location of the second vertex.

Polygon

Creates a polygon specified by the user. (For regular polygons, see N-Sided Polygon). Itfunctions like the Polyline Edge tool. Click the starting point and all subsequent points,then press |Return| to close the polygon. This will draw a line from the last selectedendpoint to the first endpoint.


175

N-Sided Polygon

Creates a regular, N-Sided Polygon of a user-specified number of sides. Click the locationof the center of the polygon. Then press the left-bracket key '[', or the right-bracket key']' to decrease or increase the number of sides, respectively. Once the correct number ofsides is selected, drag the mouse until the desired size and orientation around the centerpoint is achieved and click again to finish the N-sided polygon.

Arc Tools

The figure below displays two of the arc tools: the 3-Point Arc and 2-Point Arc tools.

The Arc tools

3-Point Arc Tool

Creates an open arc from three points. Click on the location of the first endpoint. Click asecond location to specify a point between the two endpoints (which helps determinesize), and a third location to specify the other endpoint.

2-point Arc Tool

Creates a semi-circle from two points. Click on the first endpoint location and drag themouse until the desired semi-circle size and orientation is achieved. Click this second endpoint location to finish.

Arc center, 2 points Tool

Creates an open arc from three points. First, click on the location of the center of the arc.Secondly, click a point to specify the radius of the arc. Finally, click the location of theendpoint to specify the length of the arc.

Circle and Ellipse Tools

The following figure displays an example of a circle drawn with the Circle Center, Radiustool.

A circle drawn with Circle Center, Radius tool


176

Circle Center, Radius

Creates a circle defined by its center point and radius. Click the location of the circle'scenter point, then select another point to define the radius and finish the circle.

3-point Circle

Creates a circle based on three user-specified points, similar to the 3-Point Arc tool. Clickthe first two points to set the location of the circle and the third to specify its size.

2-point Circle

Creates a circle based on the distance between two points. After selecting the first point,choose the second to define the diameter and finish the circle.

Ellipse

Draws an ellipse from three points: the center and two perpendicular radii. Click thecenter point of the ellipse, then select the desired location of the first radii. Finally, selectthe desired length of the second radii, perpendicular to the first.

Tools

Select/Manipulate

Selects anything within the sketch. This is the default tool when no other tool is selected.It can be used to:

Move an object, edge, or vertex to a new position, by clicking-and-draggingSelect a vertex or edge and lock or edit its position, by right-clicking and selectingLock Position or Edit Position.Edit the value of an angle or distance constraint, by right-clicking and selecting theedit option.Delete an edge or constraint, by right-clicking and selecting the delete option.

Select/Manipulate tool


177

Trim Curves

Deletes segments of curves until they intersect with other curves. To use this tool, click onthe section of the curve that is to be deleted.

Trim Curves tool

Insert Vertex

Inserts a vertex onto an already existing edge. Click the desired location of the new vertexon the existing edge.

Insert Vertex tool

Fillet Vertex

Converts a sharp corner into a rounded corner between two curves. Click on any sharpcorner and drag until the desired fillet radius is achieved and click to finalize fillet.


178

Fillet Vertex tool

Constraints

The geometry Constraints tools are used to modify pre-drawn shapes to the desiredspecifications.

NoteSome of the "before" images below have been marked with white arrows to show which edges areconstrained in the "after" image on the right.

Horizontal Constraint

Constrains a segment to the horizontal direction.

Polygon before (left) and after (right) two sides are constrained horizontally

Vertical Constraint

Constrains a segment to the vertical direction.

Polygon before (left) and after (right) two sides are vertically constrained


179

Collinear Constraint

Constrains two straight segments so that they are in line with each other.

Polygon before (left) and after (right) after two sides are constrained to be collinear

Parallel Constraint

Constrains two straight segments so that they are parallel to each other.

Polygon before (left) and after (right) two sides are constrained in parallel

Perpendicular Constraint

Constrains two straight segments so that they are perpendicular to each other. Thefollowing figure displays a polygon before (left) and after (right) two sides are


180

perpendicularly constrained

Tangent Constraint

Constrains a straight segment so that it is tangent to a circular segment at a point. In thefollowing figure, Circle and polygon before (left) and after (right) a side of the polygon isconstrained tangentially with reference to the circle:

Concentric Constraint

Constrains two circular segments so that they are centered upon the same point. In thefollowing figure, two circles before (left) and after (right) are made concentric:

Angle Constraint

Constrains an angle to a user-specified value between two straight lines. Click once to


181

select angle, then click a second time to place label and enter the angle size. In the figurebelow, the polygon before (left) and after (right) an angle has been constrained to a user-defined value.

Distance Constraint

Constrains the distance between two points, the distance between a point and a line, orthe length of a line to a user-specified value. After selecting the object(s) to constrain,click a final time to place label and enter distance.

As shown in the figure below, there are three different constraint "modes": parallel,vertical and horizontal. The mode is determined by the location of the mouse cursor whenyou click to specify where the constraint should be drawn.

The polygon (A) before line has been constrained, (B) with a parallel distance constraint,(C) with a vertical distance constraint and (D) with a horizontal distance constraint.


182

Equal Length Constraint

Constrains selected segments to an equal length (assumes the length of the segmentselected second). Polygon before (left) and after (right) two sides are made equal lengthto one another.

Equal Distance Constraint

Constrains two pairs of points so that each pair assumes a distance from each other equalto the distance between the original pair.In the following figure, polygon before (left) andafter (right) two sides are made equal distance from each other:

Radius Constraint

Constrains the radius to a user-specified value.

Equal Radius Constraint

Constrains selected radii to an equal length. In the following figure, two Circles before(left) and after (right) their radii are made equal:


183

Snapping

Snapping tools are used to snap the mouse to a specific point or edge in the EMProgeometry.

NoteThe blue lines in the images below highlight the "snap-to"landmarks.

Snap to Grid Line

Mouse is snapped to the nearest point on the nearest grid line.

Snap to Grid Line Tool

Snap to Grid/Edge Intersections

Mouse is snapped to the nearest intersection between the grid and the sketch edge.

Snap to Grid/Edge Intersections Tool


184

Snap to Vertices

Mouse is snapped to the nearest vertex of the sketch or edge mid-point within range.

Snap to Vertices Tool

Snap to Edges

Mouse is snapped to the edges of a pre-defined object.

Snap to Edges Tool


185

Snap to Edge/Edge Intersections

Mouse is snapped to the vertices of intersecting edges.

Snap to Edge/Edge Intersection Tool

2-D Modeling Options

The 2-D Modeling tools are used to outline or fill-in a simple geometry object.

Wire Body

The Wire Body tool is the simplest geometry object. Any of the Shape tools can be used tocreate the desired wire geometry.

Sheet Body

The Sheet Body tool is similar to the Wire Body tool except its interior is filled with amaterial.


186

NoteIt is also possible to create a sheet body using advanced options with 3-D modeling operations.

Sheet Body from Faces

The Sheet Body from Faces tool enables you to create a Sheet Body from the face of apre-existing geometry object. The interface will prompt the user to select the desiredobject face.

3-D Solid Modeling Options

The 3-D Modeling tools are used to create simple solid geometry objects from 2-D forms.

NoteFor solid body creation, the 2-D sketch must be closed so that there are no lingering endpoints.

Extrude

Extrude is used to sweep a face in the normal direction from its center. Once a 2-D form ismade in the Edit Cross Section tab, select the Extrude tab to its right to perform anextrusion. For a default extrusion, define the distance in the Extrude Distance dialog boxby typing in a numerical value, parameter name (See: Section Defining Parameters), orequation.

NoteIf units are not entered next to the numerical value, the default units are assumed.

For more information about defining distances with parameter names, refer to Defining Parameters.

Additionally, the Direction dialog box specifies the axis along which the extrusion willoccur. Clicking done after the desired geometry is created will add the object to theproject. It can now be seen in the Project Tree.

Extrusion Tool

Revolve


187

Revolve is used to sweep a face in a circular path. Once a 2-D form in made in the EditCross Section tab, select the Revolve tab to perform a revolution. For a default revolution,define the angle in the Angle dialog box by typing in a numerical value, parameter name,or equation. The Axis Root Position dialog specifies the location of the root of the axisaround which the shape will revolve. The Axis Direction box specifies the direction alongwhich the revolution will occur. Clicking DONE after the desired geometry is created willadd the object to the project. It can now be seen in the Project Tree.

Revolution Tool

Creating a sphere with the Revolution Tool

Advanced 3-D Solid Modeling Operations

The Advanced 3-D Modeling tools are used to modify a pre-defined 3-D geometry object.


188

They are available within the Extrude and Revolve operations.

Twist

Twist options control how much the face is twisted as it is swept. They can be specified byangle or law.

By Angle: Specify the total number of degrees that the face will twist while it isswept.

By Law: Specify a mathematical expression to control the rate of twist as afunction of the variable X .

In the following figure, the Twist Tool defined by A) Angle (90 degrees) and B) Law(!img4.png! )

Draft Type

Draft Type options control the expansion or contraction of the edges of the face as it isswept from its initial position.

No Draft: No expansion or contraction of edges during sweep.Draft Angle: Specify the expansion or contraction angle from initial position.

A cylinder sweep with Draft By Angle (10 degrees)


189

Draft Law: Specify a mathematical law to control the shape of the sides as theface is swept from initial position as a function of the variable X .

A cylinder sweep with Draft By Law (.5sin(2x))

End Distance/Start Distance: Specify the offset distance in the plane wherethe sweep ends/begins.

A cylinder sweep with Draft By End Distance (1 mm) and Start Distance (1 mm)


190

Hole Draft Type

Hole Draft Type options control the expansion and contraction of a hole. They aretherefore only valid during sweeping operations applied to a faces that contain holes. HoleDraft Type can be defined based on the values assigned to the edges in Draft Typeoptions, or by angle.

No Draft: No expansion or contraction is applied to the hole, even if the facehas a Draft Type applied to it.

Draft Angle: Specify the expansion or contraction angle from initial position.

Hole with no Draft (left) and a defined Draft Angle (right)

With Periphery: The expansion or contraction of the hole will be the same asthe outside edges of the face as specified in Draft Type.

Against Periphery: The expansion or contraction of the hole will be theopposite to the outside edges of the face as specified in Draft Type. (i.e., thehole will contract as the face expands and expand when the face contracts.)

Hole with Draft Angle against (left) and with the Periphery (right)


191

Gap Type Modeling Operations

The Gap Type specifies how to close the gap created by an offset. The default gap type isNatural, but the following options are available for filling gaps in the geometry.

Natural: Extends the two shapes along their natural curves until they intersect.Rounded: Creates a rounded corner between the two shapes.Extended: Draws two straight tangent lines from the ends of each shape untilthey intersect.

. Illustration of gap types, showing A) the original gap, B) Natural, C) Rounded and D)Extended.

Cut Off End

Controls the orientation of a face that does not follow its normal during a straightsweeping operation. Select this option to chop the end of the swept 3-D object so that thenormal of the end face is aligned with the line used for sweeping. Original Model (Left) andModel After Cut Off End (Right)


192

Make Solid

This option makes the model entirely solid. If this option is not selected, the model will behollow.

Modifying Existing Geometry

Specify Orientation

The Specify Orientation button is used to position the selected geometry in the simulationspace. Clicking this icon will bring up the Specify Orientation tab.

NoteFor more information on using the Specify Orientation tab, refer to Specify Orientation Tab (quickstart).For descriptions of the tools used to rotate, translate and zoom into the simulation space View Tools.

Chamfer Edges

Chamfer Edges operation creates a beveled edge between two surfaces. After selectingthe edge, it will be trimmed at a 45 angle if Constant Distance is selected in the SpecifyDistance tab. Otherwise, the user enters the chamfer distance for the surfaces on the leftand right sides of the edge.

A Chamfer operation applied to a cylinder edge

Blend Edges


193

The Blend Edges operation rounds the selected edge of the geometry. Under the SpecifyRadius_tab, the user can enter the _Blend Radius to adjust the rounding factor.

A Blend operation applied to a cylinder edge

Shell Faces

The Shell Faces operation creates a shell from existing geometry. After selecting the facesto keep open, the user can enter the Shell Thickness under the Specify Thickness tab.

NoteBy definition, the shell operation is used on geometry which is intended to have volume. This operation isnot for use an object such as a Sheet Body, whose volume is insignificant in the EMPro calculation.

A Shell operation applied to a cylinder

Loft Faces

The Loft Faces operation connects two parts of an existing geometry. Under the SpecifyLoft tab, the user can adjust the Smoothness Factor to create the desired shape. In thefollowing figure, two objects within a geometry with faces selected (left) and laterconnected by a Loft (right).


194

Remove Faces

The Remove Faces operation removes a blend or chamfer that was previously applied to ageometry edge. This operation must be applied before the user can offset the length of anobject.

NoteThis operation is useful for modifying objects that have been imported from CADfiles.

Offset Faces

With Offset Faces, the user enters a positive or negative offset distance to increase ordecrease the length of the selected model, respectively. The following figure displays acylinder with an applied negative offset (left) and positive offset (right).

Boolean Operations

Two Parts Boolean Operation

The Two Parts Boolean options perform operations on two existing geometry parts. Ineach case, one object is identified as the Tool (the part used to perform the modification),


195

and the other as the Blank (the part that is modified). There are three types of operations:

SubtractIntersectUnion

In a Subtract operation, the Tool is subtracted from the Blank. In the Intersect and Unionoperations, the part selected first is inconsequential. The following figure displays theoriginal two objects (Upper Left), objects after Boolean Union (Upper Right), objects afterBoolean Intersection (Lower Left) and objects after Boolean Subtraction (Lower Right).

Extrude Boolean Operation

The Extrude Boolean option performs an operation on an existing geometry part. In thiscase, the user chooses the Blank, and then creates the object to use as the TOOL. Theuser then specifies the orientation of the extrusion and the nature of the operation (Subtract, Intersect, or Union). In essence, this operation is a shortcut for the Two PartsBoolean operation.

A boolean extrude operation

Revolve Boolean Operation


196

The Revolve Boolean option performs an operation on an existing geometry part. The userchooses the Blank, and then creates the object to use as the Tool. The user then specifiesthe orientation of the revolution and the nature of the operation (Subtract, Intersect, orUnion).

A boolean revolve operation

3-D Patterns

Linear/Rectangular Pattern

The Linear/Rectangular Pattern option enables you to select a part in the geometry andreplicate it in a linear pattern. After selecting the part to modify, define the Spacing andNumber Of Instances in the U', V' and W' directions. Spacing refers to the distancebetween objects in the specified direction, and Number of Instances refers to the numberof objects in the specified direction. For example, if three cylinders are to be spaced at 2-mm intervals in the U'-direction, the Spacing in the U'-direction is 2 mm and the Numberof Instances in the U'-direction is 3. Additionally, the Stagger check-boxes apply a staggerin the specified direction at every other instance in that direction.

NoteSpacing refers to the distance between each object's center point in the specified object. So, for example,if the spacing between two cylinders does not exceed the distance of the cylinder's diameter, the cylinderswill overlap.

A linear pattern applied to a cylinder


197

Circular/Elliptical Pattern

The Circular/Elliptical Pattern option enables you to select a part in the geometry andreplicate it in a circular or elliptical pattern. After selecting the part to modify, navigate tothe Specify Circular/Elliptical Pattern tab and define the following fields:

Axis Point - specifies the position of the axisAxis Normal - specifies values to define the direction of the patternRoot Position (available in Elliptical Mode) - specifies a point (usually the center of apart) to use as the reference to replicate in the elliptical pattern.Major Axis (available in Elliptical Mode) - specifies the direction of the major axis.Pattern Options

Instances - specifies the number of objects in the patternAngle - specifies the angle across which the objects are patterned (i.e.,180!img463.png! means that objects are patterned across half of the ellipse)Ratio (available in Elliptical Mode) - specifies the ratio of the minor axis to themajor axis

An elliptical pattern applied to a small cylinder


198

Grid AppendixIn this section, you will learn about:

Concept of creating a grid in EMPro.How to choose a cell size to optimize your EMPro project calculation.Methods for varying the EMPro grid.How to debug the EMPro grid.

Grid Concepts OverviewThe grid consists of three sets of points, one set for each axis: X, Y and Z. At each point isa plane. For example, consider the point at X = 3, which defines a plane in Y and Z. Thepoint Y = 4 defines a plane in X and Z. The point Z = 5 defines a plane in X and Y. Wheretwo planes intersect is a line. Planes from the remaining axis cut that line into edges.These edges are called "cell edges". Building the grid consists of defining the appropriateset of plane-defining points for each axis. (Meshing, which occurs after gridding, is the actof assigning materials to each cell edge.)

The figure below displays intersecting planes in 3-D with lines and cell edges highlighted.The intersecting planes with lines (in yellow) and sample cell edges in the upper-leftcorner of the XY-plane (in red) are displayed.


199

Except for some special features associated with components, which are discussed below,the grid is made up of the following elements:

Fixed pointsGrid regionsTarget cell sizesAutomatic fixed point merge distances

A fixed point is a point on an axis at which a plane, in the other two axes, exists. Edge-on,that plane is seen as a line. That is what is meant by a "grid line". A grid region is abounded part of the grid. Grid lines are placed at the grid region boundaries. Within a gridregion the target cell size and the fixed point merge distance can be different from theproject's default target cell size and/or merge distance. One fixed point is placed at thebeginning of each grid region and another fixed point is placed at the end of each region.

The figure below displays intersecting planes in 3-D with fixed points, grid lines and a gridregion highlighted.

Intersecting planes with fixed points (in red), grid lines (in blue) and a grid region (in green)

Manual fixed points and manual grid regions are specified in the Grid Tools dialog. Othergrid regions and fixed points, associated with individual parts, may be specified using theGridding Properties Editor. Grid regions associated with a part are called part grid regions.


200

Fixed points associated with parts are called automatic fixed points, because they areautomatically extracted from the geometry of the part.

Taking these elements together, the grid is made up of manual fixed points, automaticfixed points, manual grid regions and part grid regions. Each kind of grid region contains atarget cell size and an automatic fixed point merge distance. Those values are used whencreating the grid.

NoteIt is important when building a grid to control the size of the smallest cell in the grid. The project'stimestep is derived from the smallest grid cell's edge length. Smaller timesteps result in longer runs andlarger timesteps result in faster runs. For that reason it is important to prevent the grid from having a celledge smaller than what is necessary to get the desired results.

Automatic fixed points and part grid regions are extracted from geometry and may resultin points so close that the timestep is smaller than desired. The fixed point merge distanceis used to merge automatic fixed points in order to provide control of the timestep.Automatic fixed points are merged so that they are no closer than the merge distance. If agrid region start or end boundary is too close to another grid point, the grid region may beexpanded to prevent the too-small timestep. The grid region is always expanded, nevercontracted, in these situations.

The grid is created in the following steps. Each axis, X, Y and Z, is considered separately.Note that when considering only a single axis, a cell size is really just an edge size.

Create a set containing manual and automatic fixed points and the fixed points fromthe borders of manual and part grid regions. If the Specify Padding option is chosen,then fixed points for overall bounds of the geometry are added to the set.

Editing the grid with the Specify Padding option chosen

Automatic fixed points are merged according to the merge distance. The fixed pointmerge distance can have different values at different points on the grid. There is a "main grid" fixed point merge distance that is specified on the Size tab (see thefigure above). Each grid region, including both manual and part grid regions, has itsown fixed point merge distance. The smallest fixed point merge distance for a givenpoint on the grid is chosen from all manual and/or part grid regions covering thegiven point. If no grid region covers the given point, the main grid fixed point mergedistance is chosen. Note that this allows grid regions to specify a merge distancegreater than the main grid fixed point merge distance. Notice that in the figure belowa smaller merge distance is specified in the grid region than that of the main grid.


201

Adding a grid region with a smaller cell size than the main grid

The movable points, for example automatic fixed points and grid region start/endboundaries, are moved away from unmovable points. Unmovable points includemanual fixed points and some entries associated with components, as describedbelow. Remember that grid regions never shrink in order to ensure that regionsneeding improved accuracy get it.

The set of grid points is examined. Transition regions are added to prevent adjacentcell size ratios from violating the maximum cell step factor specified on the Limitstab. For example, a transition region is generated if the maximum cell step factor is2, and 2 adjacent cells have sizes 5 mm and 1 mm. A transition region contains thefewest number of cells required to reach the desired cell size. Each cell in thetransition region has a progressively larger size. In the following figure, the MaximumCell Step Factor shown is never exceeded by adjacent cell size ratios.

The set of grid points, including the transition regions, is examined. Gaps greaterthan the target cell size at the given point are filled evenly with the fewest number ofcells required such that the cell size is less than or equal to the target. For example,consider a gap of 9.7 mm with a target cell size of 1 mm. In this case, 8 new pointsmust be added to the 2 points surrounding the gap. The distance is

, bridged by 9 cell edges. Each cell edge will be

per edge.

That completes the calculation of grid points. There are some key facts to note:

No matter what the fixed point merge distance may be, it is always possible for


202

there to be two fixed points a distance of apart after themerging has been completed. If the target cell size at the given point is less

than , then this distance must be subdivided into two ormore cell edges.

Consider a 0.7 mm merge distance with a .2 mm target size. Given that

remainder !img220.png! , there must be 3 grid lines in the0.7 mm gap. Those 3 grid lines create 4 spaces in the gap. The gaps are

each. The formula is:

The target cell size may be greater than the automatic fixed point mergedistance. Consider a merge distance of 0.7 mm and a target distance of 1 mm.If the automatic fixed points are dense then every gap is exactly 0.7 mm. Nogaps are greater than 1 mm, so the work is done. If there was at least oneregion of sparse automatic fixed points then there may be a gap greater than1 mm. That gap must be bridged using one or more extra grid lines. The

smallest such gap is , or 0.5 mm in our example.

The smallest cell in the grid will be the smaller of and .Any given grid may not encounter one or both of those situations, so that grid'ssmallest cell size may be larger.

Choosing an Appropriate Cell SizeSince smaller cells require longer calculation time, it may be advantageous to also definethe lower limit of the cell size in the Size tab of the Gridding Properties Editor window. Theminimum cell size can be defined as a Merge distance (i.e., a specific distance with units),or as a ratio of the Merge value to the Target base cell size (i.e., a Merge value of 0.8would restrict the minimum cell size from dropping to a value below 80% of the Targetbase cell size.)

NoteTo learn about the factors that affect the smallest cell size the equations at the end of the previoussection, refer to Grid Concepts Overview.

When defining the Target base cell size, ensure that the cell size is much less than thesmallest wavelength for which accurate results are desired. A commonly applied constraintis ten cells per wavelength , meaning that the side of each cell should be less than one-tenth of the wavelength of the highest frequency (shortest wavelength) of interest. If thecell size is much larger than this, the Nyquist sampling limit is approached too closely forreasonable results to be obtained. Significant aliasing is possible for signal components


203

above the Nyquist limit.

Choosing a cell size of one-tenth of a wavelength is a good starting point, but otherfactors may require a smaller cell size to be chosen, such as small geometry features andmaterial characteristics.

Grid definitions can be customized for specific objects in the Gridding Properties Editorwindow, so that smaller features are considered, without having to apply smaller,memory-intensive cells to the whole grid.

NoteFor more information about assigning grid definitions to specific object Gridding Properties Editor (using).

Material characteristics will also influence cell characteristics since EMPro 2010 is avolumetric computational method. If some portion of the computational space is filled withpenetrable material, the wavelength in the material must be used to determine themaximum cell size. Geometries containing electrically dense materials require smaller cellsthan geometries that contain only free space and perfect conductors. For this reason, amaterial definition must be applied to Parts objects to generate a valid mesh. An errormessage will appear in the case that a material is not assigned to an object.

NoteFor information on applying material definitions to objects Material Editor (using).

Grid Regions vs. Fixed PointsThere are two primary means of varying the grid in EMPro 2010. A grid region is a regionwithin the grid that is assigned its own target cell size, which is different from the defaultgrid size defined in the main Size tab of the Grid Tools button. A fixed point is a point onan axis where a grid line will be placed. Cell sizes are adjusted to flow smoothly betweenfixed points, never exceeding the Target cell size.

The target cell size can vary within different grid regions along a given axis. The maingrid's target cell size applies everywhere except within grid regions. A grid region has startand end boundaries on an axis. Grid regions can have target cell sizes and automatic fixedpoint merge distances that differ from the main grid and from other grid regions. Gridregions can overlap. When they do, the smallest target cell size and the smallestautomatic fixed point merge distance are chosen from all of the overlapping grid regionsat the given point.

Grid regions, like fixed points, can be manual or automatic. Manual grid regions aredefined on tabs associated with the main grid editor. Automatic grid regions areassociated with parts and so are also called part grid regions. Part grid regions are definedby right-clicking on a part in the tree and choosing Gridding Properties.

Any grid region includes fixed points for the grid region boundaries. Between the bounds,the target cell size and automatic fixed point merge distances can be different than theirvalues outside the grid region bounds.


204

The grid flows as evenly as possible between fixed points, using at each point theappropriate target cell size and automatic fixed point merge distance for that point.

The figure below shows a simple shape that has a uniform grid size of 1 m.

Simple extrusion with uniform grid

The following figure applies an automatic grid region to this simple shape. Note that theedges of the rectangle are now aligned with the edges of the grid. Also note that, becausethe height of the rectangle was not evenly divided by 1 m, the main grid spacing adjustedslightly to accommodate for this.

Simple extrusion with applied grid regions

The following figure applies automatic fixed points to the shape (all default settings in theGridding Properties Editor are applied). Note that the edges, like in the case of the appliedgrid regions, are aligned with the rectangle's edges. The cells within the shape, however,are auto-generated, and therefore vary from the default target cell size as little aspossible.

Simple extrusion with applied fixed points


205

Simple extrusion with applied fixed points

The next illustration applies a user-defined grid region to cover the area of the shape, withcell sizes significantly smaller than the main grid. Note the transition region of cellssurrounding the shape. This region contains cells of non-uniform size, which vary from theCell Size defined for the specific grid region to the Cell Size defined for the entire grid, at arate which is limited by the Maximum Cell Size Step Factor defined in the Limit tab of theEdit Grid dialog in Grid Tools.

Simple extrusion with a manual grid region defined in Grid Tools

Debugging the GridThere may be times when one does not understand how certain EMPro grid features in agiven grid came to exist. The individual grid features may be turned off so that the grid


206

becomes, in essence, a blank slate. Individual grid features can then be turned back onone at a time. Seeing how the grid changes in this step-by-step fashion makes it possibleto locate the cause of any given grid feature.

The following steps outline how to turn off individual grid features.

Be sure to save a copy of the project and open the copy for use in the debuggingprocess.View the mesh and use the MEASURE tool to measure the geometry's largest extent.Open the GRID TOOLS dialog:

Under the SIZE tab, set the X, Y, and Z TARGET sizes to something just largerthan the geometry's largest extent.Set the MERGE sizes for each axis to e.g. 1e-12 mm (not a ratio).Set the Free Space Padding cell values to zero or possibly one.Delete all entries under Fixed Points and Grid Regions.

If your project includes any Circuit Components, ensure that the box labeled EvenlySpaced In Orthogonal Directions is unchecked in the Circuit Component Properties:Properties tab.

Be sure to turn this property back on (where applicable) later in the debuggingprocess, as accurate results in many cases depend on even component spacing.

Right-click on a part under Project Tree: Parts and choose View Flat Parts ListClick on the GR (Grid Region) column heading to bring check-marked parts tothe top of the list.

Right-click on the part in the flat parts list and select Gridding Properties. Turn off thepart's Automatic Grid Regions and/or Fixed Points.

Click the FP (Fixed Points) column heading to bring check-marked parts to thetop of the list.

As was done for parts with grid regions above, turn off any remaining partswith automatic fixed points chosen.

Under the Grid Tools: Limits tab, set the Maximum Cell Step Factor to somethinglarge, e.g. 20000. Set the Minimum Cell Size to something very small, (e.g., 1e-12mm).

After performing those steps the grid should be relatively bare, consisting only of thegeometry bounds and padding cells, if any were retained.

Now begin turning on grid features one at a time.

Choose an important part and turn on its fixed points (in its Gridding Propertieseditor). Turn off all Automatic Discovery Options except one, and click Apply. Turn offthat extraction type and turn on the next one, examining each type in turn.

As you examine the fixed points for the part, experiment with its automatic gridregion. Turn it on and initially set its Target value to the same value as the maingrid's target (under Grid Tools). Experiment with different Merge distances,clicking apply each time. At first, leave the part grid region's target cell size thesame as the main grid's target cell size.Choose the best combination of fixed point extraction types and merge distancesfor the part.

Go to another important part, in turn, until all parts for which automatic fixed points


207

have been examined. Ensure that automatic fixed points are turned on for thoseparts and that an appropriate merge distance was chosen for each.Review each important part. Experiment with the target cell sizes for each part,setting the target cell size to the value desired for the final grid.

At this point the grid probably contains adjacent cells whose widths vary by more than isallowed by FDTD theory. Go to the Grid Tools: Limits tab and change the Maximum CellStep Factor to 2.0, being sure to tab out of the field to make the new setting take effect.Transition regions will appear. You may wish to experiment with values lower than 2.0,although the value must be greater than 1.0. If the Maximum Cell Step Factor is too low,the grid will report that a grid could not be created. Choose a higher step factor that is notgreater than 2.0.

The grid probably will contain large gaps even after the transition regions have beenadded. Change the main grid's target cell size to its final desired value to see theremaining gaps filled with grid lines spaced appropriately.


208

BibliographyK. S. Kunz and R. J. Luebbers, "The Finite Difference Time Domain Method for1.Electromagnetics". Upper Saddle River: CRC Press, 1993.A. Taflove and S. Hagness, Computational Electrodynamics: The Finite-Difference2.Time-Domain Method, Third Edition. New York: Artech House Publishers, 2005.C95.3-2002, "Recommended practice for measurements and computations of radio3.frequency electromagnetic fields with respect to human exposure to such fields,100khz to 300ghz," IEEE Standards and Coordinating Committee 28 on Non-IonizingRadiation Hazards, pp. i-126, April 2002.K. Yee, "Numerical solution of initial boundary value problems involving maxwell's4.equations in isotropic media," IEEE Transactions on Antennas and Propagation 14,pp. 302-307, 1966.C. Balanis, Advanced Engineering Electromagnetics. New York: Wiley, 1989. Section5.2.8.3.J. R. W. Simon Ramo and T. V. Duzer, Fields and Waves in Communication6.Electronics. New York: Wiley, 1994. Section 13.12.R. E. Collin, Foundations for Microwave Engineering. New York: Wiley-IEEE P, 2000.7.Section 6.7.B. Lax and K. J. Button, Microwave Ferrites and Ferrimagnetics. McGraw-Hill, 1962.8.Sections 4.1, 4.2.F. Kung and H. T. Chuah, "A finite-difference time-domain (fdtd) software for9.simulation of printed circuit board (pcb) assembly," Progress in ElectromagneticResearch, PIER 50, vol. Elsevier book series, pp. 299-335, 2005.

Date post:	25-Feb-2021
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

May 2010 EMPro 2010 - Keysightedadownload.software.keysight.com/eedl/empro/2010/pdf/... · 2014. 7....

Documents