Getting Started with ABAQUS/Explicit Trademarks and Legal Notices Conversion Tables, Constants, and Material Properties HKS

Offices and Representatives 1. Introduction ABAQUS is a suite of powerful engineering simulation programs, based on the finite element method, that can solve problems ra challenging nonlinear simulations. ABAQUS contains an extensive library of elements that can model virtually any geometry. It ha the behavior of most typical engineering materials including metals, rubber, polymers, composites, reinforced concrete, crushab and rock. Designed as a general-purpose simulation tool, ABAQUS can be used to study more than just structural (stress/displa as heat transfer, mass diffusion, thermal management of electrical components (coupled thermal-electrical analyses), acoustics piezoelectric analysis. ABAQUS is simple to use even though it offers the user a wide range of capabilities. Even the most complicated analyses can b the engineering data, such as the model's geometry, material behavior, boundary conditions, and applied loads. ABAQUS autom continuous adjustments during the analysis to ensure accuracy as well as efficiency. The user rarely has to control the solution a 1.1 The ABAQUS modules ABAQUS consists of two main analysis modules--ABAQUS/Standard and ABAQUS/Explicit. There are also two special-purpose ABAQUS/Aqua and ABAQUS/Design. In addition, ABAQUS/Safe provides fatigue postprocessing; while ABAQUS/ADAMS, ABAQ CATIA, and C-MOLD, respectively. ABAQUS/CAE is the complete ABAQUS environment that includes capabilities for creating AB jobs, and evaluating results. ABAQUS/Viewer is a subset of ABAQUS/CAE that includes just the postprocessing functionality. Th Figure 1-1 ABAQUS modules. ABAQUS/Standard ABAQUS/Standard is a general-purpose analysis module that can solve a wide range of linear and nonlinear problems involving components. ABAQUS/Standard is not discussed in this guide but is described in the companion publication Getting Started wit ABAQUS/Explicit ABAQUS/Explicit is a special-purpose analysis module that uses an explicit dynamic finite element formulation. It is suitable for blast problems, and is also very efficient for highly nonlinear problems involving changing contact conditions, such as forming s ABAQUS/CAE ABAQUS/CAE (Complete ABAQUS Environment) is an interactive, graphical environment for ABAQUS. It allows models to be cre the structure to be analyzed and decomposing the geometry into meshable regions. Physical and material properties can be ass conditions. ABAQUS/CAE contains very powerful options to mesh the geometry and to verify the resulting analysis model. Once control the analysis jobs. The visualization module can then be used to interpret the results. Only the visualization module of ABAQUS/CAE (ABAQUS/Viewer) is covered in this guide. If you want to learn how to use other a can obtain a copy of the ABAQUS/CAE training material. ABAQUS/Viewer ABAQUS/Viewer is the visualization module of ABAQUS/CAE; it is an interactive, graphical postprocessor that supports all of the range of options for interpreting the results. ABAQUS/Viewer is discussed in this guide. ABAQUS/Aqua ABAQUS/Aqua is a set of optional capabilities that can be added to ABAQUS/Standard. It is intended for the simulation of offsho capabilities include the effects of wave and wind loading and buoyancy. ABAQUS/Aqua is not discussed in this guide. ABAQUS/ADAMS ABAQUS/ADAMS allows ABAQUS finite element models to be included as flexible components within the MDI ADAMS family of p synthesis formulation of ADAMS/Flex. ABAQUS/ADAMS is not discussed in this guide. ABAQUS/CAT ABAQUS/CAT allows ABAQUS analyses to be set up and postprocessed entirely in CATIA. ABAQUS/CAT is not discussed in this ABAQUS/C-MOLD ABAQUS/C-MOLD translates finite element mesh, material property, and initial stress data from a C-MOLD mold filling analysis t this guide. ABAQUS/Design ABAQUS/Design is a set of optional capabilities that can be added to ABAQUS/Standard to perform design sensitivity calculatio ABAQUS/Safe ABAQUS/Safe is the fatigue analysis module in ABAQUS. Using results from ABAQUS analyses, it determines the fatigue life of a 1.2 Getting started with ABAQUS/Explicit This is an introductory text designed to give new users guidance in analyzing continuum, shell, and framework problems with AB not need any previous knowledge of ABAQUS to benefit from this guide, although some previous exposure to the finite element 1.2.1 How to use this guide

There are seven chapters in this guide, each of which introduces one or more topics. Most chapters contain a discussion of per work through the examples carefully since they contain a great deal of practical advice on using ABAQUS. The capabilities of AB results. This chapter is a short introduction to ABAQUS and this guide. Chapter 2, "ABAQUS Basics," which is centered around a simple Chapter 2, "ABAQUS Basics," you will know how to prepare input data for an ABAQUS simulation, check the data, solve the anal "Overview of Explicit Dynamics," introduces the concept of explicit dynamics and explains the concept of "stability" using a wave Chapter 4, "Finite Elements and Rigid Bodies," presents an overview of the main element families available in ABAQUS/Explicit a mesh. Material modeling, with an emphasis on elastic-plastic metals and hyperelastic rubbers, is presented in Chapter 5, "Mater plate. Chapter 6, "Contact," explains contact modeling, which is illustrated with two examples: a circuit board, housed in protect square tube crushed between two rigid plates. Using ABAQUS/Explicit to solve quasi-static problems is presented in Chapter 7, sheet metal forming simulation, which requires importing between ABAQUS/Explicit and ABAQUS/Standard to perform the formin 1.2.2 Conventions used in this guide Different text styles used in the tutorial examples follow: nput in COURIER FONT should be typed into the input file, ABAQUS/Viewer, or your computer exactly as shown. For example, *SOLID SECTION, ELSET=LUG, MATERIAL=STEEL would be entered into an input file. abaqus job=skew_nl would be typed on your computer to run ABAQUS. tems in <italics> indicate that user data such as a name or value should be substituted: *SOLID SECTION, ELSET=<element set>, MATERIAL=<material name> enu selections, tabs within dialog boxes, and labels of items on the screen in ABAQUS/Viewer are indicated in bold: View->View Options->Hardware Deformed Shape Plot Options 1.3 ABAQUS documentation The documentation for ABAQUS is extensive and complete. You will need to refer to the manuals regularly as you use ABAQUS. HKS, unless otherwise specified, in printed form and through our online documentation server. For more information on accessin procedures in the ABAQUS/Explicit User's Manual. ABAQUS/Explicit User's Manual This is the ABAQUS manual that you will use most often. It is the reference manual for all of the capabilities in ABAQUS/Explicit. so you should have a copy available as you work through the examples. ABAQUS Keywords Manual This manual contains a complete description of all the input options available in ABAQUS, including any parameters that may be ABAQUS/Viewer User's Manual This basic reference document for ABAQUS/Viewer includes an introductory tutorial as well as a complete description of how to Using ABAQUS Online Documentation This online manual contains instructions on using the ABAQUS online documentation server to read the online manuals. Other documentation available from HKS: ABAQUS Example Problems Manual This manual contains more than 75 detailed examples designed to illustrate the approaches and decisions needed to perform m motion of an elastic-plastic pipe hitting a rigid wall; inelastic buckling collapse of a thin-walled elbow; explosive loading of an e buckling of a composite shell with a hole; and deep drawing of a metal sheet. The description of each problem includes a discu When you want to use a feature that you have not used before, you should look up one or more examples that use that feature. usage of the capability. To find an example that uses a certain feature, search the online documentation or use the abaqus findk keyword/problem database,'' Section 3.2.8 of the ABAQUS/Explicit User's Manual, for more information). All the input files assoc installation. A utility that gives you access to these files, called ABAQUS/Fetch, is provided with every ABAQUS release. The syn abaqus fetch job=<file name> You may fetch any of the example files from the installation directory, so that you can run the simulations yourself and review th Input Files electronic book. ABAQUS Benchmarks Manual This volume (available online and, if requested, in print) contains benchmark problems and standard analyses used to evaluate t calculations and other published results. These problems, like the example problems, offer a good starting point for learning abo benchmark problems are included in this manual. The input files are also supplied as part of the installation and can be accesse ABAQUS Verification Manual This online-only volume contains basic test problems that provide quality assurance testing of each individual program feature ( these problems when learning to use a new capability. The input files are also supplied as part of the installation and can be acc

ABAQUS Input Files This online manual contains all the input files that are included with the ABAQUS release and referred to in the ABAQUS Example ABAQUS Verification Manual. They are listed under the title of the problem that refers to them. The input file references in the ma this book cannot be saved directly to a file, they can be saved by copying and pasting them into an editor. ABAQUS Theory Manual This volume (available online and, if requested, in print) contains detailed, precise discussions of the theoretical aspects of ABA background and is not required on a routine basis. ABAQUS/CAE User's Manual This reference document for ABAQUS/CAE includes three comprehensive tutorials as well as detailed descriptions of how to use evaluation. ABAQUS/Standard User's Manual Split into three volumes, this is the reference manual for all of the capabilities in ABAQUS/Standard. Release Notes This document contains brief descriptions of the new features available in the latest release of the ABAQUS product line. Site Guide This document, which is provided with each ABAQUS license, describes how to install ABAQUS and how to configure the installa most relevance to users, is also included in the User's manuals. Quality Assurance Plan This document describes HKS's QA procedures. It is a controlled document, provided to customers who subscribe to either HKS Lecture Notes Notes and workshops are available for many features and applications for which ABAQUS is used, such as metal forming or hea offers to help users improve their understanding and usage of ABAQUS. While they are not intended as stand-alone tutorial mat that mode. Frequently, the lecture notes on a topic serve as a good introduction to the topic, making the User's Manual easier to Documentation Price List or can be found under Products on the ABAQUS Home Page at www.abaqus.com. Newsletters HKS publishes ABAQUS/News and ABAQUS/Answers on a regular schedule. ABAQUS/News includes topical information about includes technical articles on particular topics related to ABAQUS usage. These newsletters are distributed at no cost to users w office if you wish to be added to the mailing list for these publications. They are also archived in the Reference Shelf on the ABA ABAQUS Web server HKS has a home page on the World Wide Web (www.abaqus.com), containing a variety of useful information about the ABAQUS requently asked questions BAQUS systems information and machine requirements enchmark timing documents rror status reports BAQUS documentation price list raining seminar schedule ewsletters User's manuals for ABAQUS/ADAMS, ABAQUS/CAT, ABAQUS/C-MOLD, and ABAQUS/Safe are also available. 1.4 Support HKS offers both technical (engineering) support and systems support for ABAQUS. Technical and systems support are provided support as an important part of the service we offer and encourage you to contact us with any questions or concerns that you h telephone, fax, electronic mail, or regular mail. Information on how to contact each office is listed in the front of each ABAQUS ABAQUS Home Page on the World Wide Web. When contacting your local support office, please specify whether you would like ABAQUS analysis) or systems support (ABAQUS will not install correctly, licensing does not work correctly, or other hardware-re We welcome any suggestions for improvements to the support program or documentation. We will ensure that any enhancement wish to file a complaint about the service or products provided by HKS, refer to the ABAQUS Home Page. 1.4.1 Technical support HKS technical support engineers can assist in clarifying ABAQUS features and checking errors by giving both general informatio analyses. If you have concerns about an analysis, we suggest that you contact us at an early stage, since it is usually easier to correct an analysis at the end. Please have the following information ready before calling the technical support hotline, and include it in any written contacts: he version of ABAQUS that are you using. -The version numbers for ABAQUS/Standard and ABAQUS/Explicit are given at the top of the data (.dat) file. -The version numbers for ABAQUS/CAE and ABAQUS/Viewer can be found by selecting Help->On version from the main menu -The version number for ABAQUS/CAT is given at the top of the input (.inp) file as well as the data file.

-The version numbers for ABAQUS/ADAMS and ABAQUS/C-MOLD are output to the screen. -The version number for ABAQUS/Safe is given under the ABAQUS logo in the main window. he type of computer on which you are running ABAQUS. he symptoms of any problems, including the exact error messages, if any. orkarounds or tests that you have already tried. For support about a specific problem, any available ABAQUS output files may be helpful in answering questions that the support The support engineer will try to diagnose your problem from the model description and a description of the difficulties you are ha which can be faxed to HKS or sent in the mail. Plots of the final results or the results near the point that the analysis terminated problem. If the support engineer cannot diagnose your problem from this information, you may be asked to send the input data. The data ABAQUS Home Page at www.abaqus.com for the media formats that are currently accepted. All support calls are logged into a database, which enables us to monitor the progress of a particular problem and to check that know the log number of your particular call for future reference, please ask the support engineer. If you are calling to discuss an mention it so that we can consult the database to see what the latest action has been and, thus, avoid duplication of effort. In a (or include it at the top of any e-mail correspondence). 1.4.2 Systems support HKS systems support engineers can help you resolve issues related to the installation and running of ABAQUS, including licensin You should install ABAQUS by carefully following the instructions in the ABAQUS Site Guide. If you encounter problems with the ABAQUS Site Guide to ensure that they have been followed correctly. If this does not resolve the problems, look on the ABAQUS known installation problems. If this does not address your situation, please contact your local support office. Send whatever info an aborted analysis or a detailed explanation of the problems encountered. Whenever possible, please send the output from the 1.4.3 Support for academic institutions Under the terms of the Academic License Agreement, we do not provide support to users at academic institutions unless the ins for more information. 1.5 Review of the finite element method A review of the fundamental concepts related to the finite element method can be found in ``A quick review of the finite element Keywords Version. The extension of the finite element method to explicit dynamics is covered in ``Finite element method for exp 2. ABAQUS Basics A complete ABAQUS/Explicit analysis usually consists of three distinct stages: preprocessing, simulation, and postprocessing. T Preprocessing (ABAQUS/CAE) In this stage you define the model of the physical problem and create an ABAQUS input file. The model is usually created graph ABAQUS input file for a simple analysis can be created directly using a text editor. Simulation (ABAQUS/Explicit) The simulation, which normally is run as a background process, is the stage in which ABAQUS/Explicit solves the numerical prob analysis include displacements and stresses that are stored in files ready for postprocessing. Depending on the complexity of th range from seconds to days. Postprocessing (ABAQUS/Viewer) When the simulation has completed or while it is running, you can evaluate the results interactively using ABAQUS/Viewer or ano for displaying the results, including color contour plots, animations, deformed shape plots, and X-Y plots. 2.1 Components of an ABAQUS model An ABAQUS model is composed of several components that together describe the physical problem to be analyzed and the resu following information: geometry, element section properties, material data, loads and boundary conditions, analysis type, and ou Geometry Finite elements, nodes, and rigid bodies define the basic geometry of the model. Each element in the model represents a discre by many interconnected elements. Elements are connected to one another by shared nodes. The coordinates of the nodes and which elements--comprise the model geometry. The collection of all the elements and nodes in a model is called the mesh. Ge actual geometry. The element type, shape, and location, as well as the overall number of elements used in the mesh, affect the results obtained f the number of elements in the mesh), the more accurate the results. As the mesh density increases, the analysis results converg time required for the analysis increases. The solution from the numerical model is generally an approximation to the solution of t approximations made in the model's geometry, material behavior, boundary conditions, and loading determines how well the num Element section properties The geometry of many elements in ABAQUS/Explicit is not defined completely by the nodal coordinates. For example, the layers not defined by the nodes of the element. Such additional geometric data are defined as physical properties of the element and a 4, "Finite Elements and Rigid Bodies," describes the necessary section properties for each element type.

Material data Material properties for all elements must be specified. The validity of results from ABAQUS/Explicit is limited by the quality of the Loads and boundary conditions ABAQUS/Explicit provides a variety of loading options, the most common of which include: oint loads; ressure loads on surfaces; ody forces, such as the force of gravity; and hermal loads. Boundary conditions are used to constrain portions of the model to remain fixed (zero displacements) or to move by a prescribe Analysis type The most common type of simulation performed in ABAQUS/Explicit is an explicit dynamic analysis, where the dynamic response thermal stress analyses, which simulate the coupled thermal-mechanical response of a body, and annealing analyses, which sim metals are heated to a high temperature, can also be performed in ABAQUS/Explicit. Only explicit dynamic analyses are discuss Output requests A finite element analysis can generate a large amount of output. Making use of the variety of available output options, you can r adequately. 2.2 Format of the input file The ABAQUS input file is the means of communication between the preprocessor, usually ABAQUS/CAE, and the solver, ABAQU of the numerical model, is a text file with an intuitive, keyword-based format. It is easily modified using a text editor. Indeed, the into an editor. If you are using ABAQUS/CAE, you will not need to see the input file. However, by understanding the input file, yo We use the example of an overhead hoist shown in Figure 2-1 to illustrate the basic format of the ABAQUS input file. The hoist hand end and mounted on rollers at the right-hand end. The members can rotate freely at the joints. Since the model is two-dim simulation is performed to determine the structure's deflection and the peak stress in its members when a 10 kN load is applied Figure 2-1 Schematic of an overhead hoist. Since this problem is very simple, the ABAQUS input file is compact and easy to understand. The complete ABAQUS input file fo ``Overhead hoist frame,'' Section A.1, is split into two distinct parts. The first section contains model data and includes all the in contains history data that define the model's loading history. The history can be further subdivided into a sequence of steps, ea The input file is composed of a number of option blocks that contain data describing a part of the model. Each option block beg data lines. These lines cannot exceed 256 characters. Figure 2-2 Input for overhead hoist model. 2.2.1 Keyword lines Keywords (or options) always begin with a star or asterisk (*). For example, *NODE is the keyword for specifying the nodal coor connectivity. Keywords are often followed by parameters, which may be required or optional. The *ELEMENT option requires the when defining elements. For example, *ELEMENT, TYPE=T2D2 indicates that we are defining T2D2 elements (two-dimensional truss elements with two nodes). Many parameters are optional a *NODE, NSET=PART1 indicates that all the nodes defined in this option block will be put into a set called PART1. It is not essential to put nodes into se Keywords and parameters are case independent and must use enough characters to make them unique. Parameters are separat to associate the value with the parameter. Occasionally, so many parameters are required that they will not all fit on a single 256-character line. In this case a comma is pl continuation line. For example, *ELEMENT, TYPE = T2D2, ELSET = FRAME is a valid keyword line. Details of the keywords are documented in the ABAQUS Keywords Manual. 2.2.2 Data lines Keyword lines are usually followed by data lines, which provide data that are more easily specified as lists than as parameters on coordinates; element connectivities; or tables of material properties, such as stress-strain curves. The data required for particul For example, the option block defining the nodes for the overhead hoist model is: *NODE 101, 0., 0., 0. 102, 1., 0., 0. 103, 2., 0., 0. 104, 0.5, 0.866, 0.

105, 1.5, 0.866, 0. The first piece of data on each data line is an integer that defines the node number. The second, third, and fourth entries are flo The data can consist of a mixture of integer, floating-point, or alphanumeric values. Floating point values can be entered in a va as the number four: 4.0 4. 4 4.0E+0 .4E+1 40.E-1 Data items are separated by commas, as in Figure 2-2 , which allows fairly arbitrary spacing of the input values on the data line comma. 2.3 Creating an input file The simulation of the pin-jointed, overhead hoist in Figure 2-1 is used to illustrate the creation of an ABAQUS input file using a data into a file using one of the editors available on your computer. The ABAQUS input file must have a .inp file extension. For c which can be chosen to identify the analysis, is called the jobname. In this case use the jobname "frame_xpl" to easily associate All of the other examples in this guide assume that you will be using a preprocessor, such as ABAQUS/CAE, to generate the me for all the examples are available. See ``Execution procedure for ABAQUS/Fetch,'' Section 3.2.9 of the ABAQUS/Explicit User's M However, since the purpose of this example is to help you understand the structure and format of the ABAQUS input file, you sh or copy the input file that is provided. 2.3.1 Units Before starting to define this or any model, you need to decide which system of units you will use. ABAQUS has no built-in syst Some common systems of consistent units are shown in Table 2-1 . Table 2-1 Common systems of consistent units. Quantity SI SI (mm) US Unit (ft) US Unit (inch) Length m mm ft in Force N N lbf lbf Mass kg tonne (103 kg) slug lbf s2/in Time s s s s Stress Pa (N/m2) MPa (N/mm2) lbf/ft2 psi (lbf/in2) Energy

J mJ (10-3 J) ft lbf in lbf Density kg/m3 tonne/mm3 slug/ft3 lbf s2/in4 The SI system of units is used throughout this guide. Users working in the systems labeled "US Unit" should be careful with the u properties are multiplied by the acceleration due to gravity. 2.3.2 Coordinate systems You also need to decide which coordinate system to use. The global coordinate system in ABAQUS is a right-handed, rectangu to be the horizontal axis of the hoist and the global 2-axis to be the vertical axis (Figure 2-3 ). The global 3-axis is normal to th left-hand corner of the frame. Figure 2-3 Coordinate system and origin for model. For two-dimensional problems, such as this one, ABAQUS requires that the model lie in a plane parallel to the global 1-2 plane. 2.3.3 Mesh You must select the element types and design the mesh. Creating a proper mesh for a given problem requires experience. For th member of the frame, as shown in Figure 2-4 . Figure 2-4 Finite element mesh. A truss element, which can carry only tensile and compressive axial loads, is ideal for modeling pin-jointed frameworks, such as "Finite Elements and Rigid Bodies," and also in the ABAQUS/Explicit User's Manual, which describes every element available in A ABAQUS/Explicit User's Manual makes locating a particular element easy. Whenever you are using an element for the first time, connectivity and any element section properties needed to define the element's geometry. The connectivity for the truss elements used in the overhead hoist model is shown in Figure 2-5 . Figure 2-5 Connectivity for the 2-node truss element (T2D2). Node and element numbers are merely identification labels. They are usually generated automatically by ABAQUS/CAE or anothe numbers is that they must be positive integers. Gaps in the numbering are allowed, and the order in which nodes and elements associated with an element are removed automatically and are not included in the simulation. In this case we use the node and element numbers shown in Figure 2-6 . Figure 2-6 Node and element numbers for the hoist model. 2.3.4 Model data The first part of the input file must contain all of the model data. These data define the structure being analyzed. In the overhead eometry: Nodal coordinates. Element connectivity. Element section properties. aterial properties. Heading The first option in any ABAQUS input file must be *HEADING. The data lines that follow the *HEADING option are lines of text de accurate description to allow the input file to be identified at a later date. Moreover, it is often helpful to specify the system of un the *HEADING option block for the hoist problem contains the following: *HEADING Two-dimensional overhead hoist frame SI units (kg, m, s, N) Data file printing options By default, ABAQUS will not print an echo of the input file or the model and history definition data to the printed output (.dat) file history definition in a datacheck run before performing the analysis. The datacheck run is discussed later in this chapter. To request a printout of the input file and of the model and history definition data, add *PREPRINT, ECHO=YES, MODEL=YES, HISTORY=YES to the input file. Nodal coordinates The coordinates of each node can be defined once you select the mesh design and node numbering scheme. For this problem nodes are defined using the *NODE option, and the value of the NSET parameter defines the name of the node set to which the

<node number>,-coordinate>,-coordinate>,-coordinate> The nodes for the hoist model belong to the node set NALL, and they are defined as follows: *NODE, NSET=NALL 101, 0., 0., 0. 102, 1., 0., 0. 103, 2., 0., 0. 104, 0.5, 0.866, 0. 105, 1.5, 0.866, 0. Element connectivity The members of the overhead hoist are modeled with truss elements. The format of each data line for a truss element is <element number>, <node 1>, <node 2> where node 1 and node 2 are at the ends of the element (see Figure 2-5 ). For example, element 16 connects nodes 103 and 16, 103, 105 The TYPE parameter on the *ELEMENT option must be used to specify the kind of element being defined. In this case you will u One of the most useful features in ABAQUS is the availability of node and element sets that are referenced by name. By using th defined in the option block are added to an element set called FRAME. A set name can have as many as 80 characters and mus through element set names, all elements in the model must belong to at least one element set. The complete *ELEMENT option block for the overhead hoist model (see Figure 2-6 ) is shown below: *ELEMENT, TYPE=T2D2, ELSET=FRAME 11, 101, 102 12, 102, 103 13, 101, 104 14, 102, 104 15, 102, 105 16, 103, 105 17, 104, 105 Element section properties Each element must reference an element section property. The appropriate element section option for each element and the add described in the ABAQUS/Explicit User's Manual. For the T2D2 element you must use the *SOLID SECTION option and give one data line with the cross-sectional area of the elem assumed to be 1.0. In this case all the members are circular bars that are 5 mm in diameter. Their cross-sectional area is 1.963 ?10-5 m2. The MATERIAL parameter, which is required for most element section options, references the name of a material property definit to 80 characters and must begin with a letter. In this example all of the elements have the same section properties and are made of the same material. Typically, there will be example, different components in a model may be made of different materials. The elements are associated with material proper elements were included in an element set called FRAME. Element set FRAME is then used as the value of the ELSET parameter o your input file: Materials One of the features that makes ABAQUS easy to use is that almost any material model can be used with any element. Once the appropriate, with the elements in the mesh. ABAQUS has a large number of material models, many of which include nonlinear behavior. In this overhead hoist example we u discussion of all the material models available in ABAQUS/Explicit can be found in the ABAQUS/Explicit User's Manual. Linear elasticity is appropriate for many materials at small strains, particularly for metals up to their yield point. It is characterized as shown in Figure 2-7 . Figure 2-7 Linear elastic material. The material behavior is characterized by two constants: Young's modulus, , and Poisson's ratio, . A material definition in the ABAQUS input file starts with a *MATERIAL option. The parameter NAME is used to associate a mater Material suboptions directly follow their associated *MATERIAL option. Several suboptions may be required to complete the mate material that is listed on the most recent *MATERIAL option until another *MATERIAL option or a non-material option block is giv Without considering thermal expansion effects (which would be defined with the *EXPANSION material option), two material opti elastic material. The form of this option block is the following: *ELASTIC <>,<> *DENSITY

<>, Therefore, the complete, isotropic, linear elastic material definition for the hoist members, which are made of steel, should be e *MATERIAL, NAME=STEEL *ELASTIC 200.E9, 0.3 *DENSITY 7800., The model definition portion of this problem is now complete since all the components describing the structure have been speci 2.3.5 History data The history data define the sequence of events for the simulation. This loading history is divided into a series of steps, each def contains the following information: he type of simulation (dynamic, fully coupled thermal-stress, or anneal). he loads and constraints. he output required. In this example we are interested in the dynamic response of the overhead hoist to a suddenly applied load of 10 kN at the cente on the right-hand end (see Figure 2-1 ). This loading is a single event, so only a single step is needed for the simulation. The *STEP option is used to mark the start of a step. Like the *HEADING option, this option may be followed by data lines conta include nonlinear geometric effects in the analysis by setting the NLGEOM parameter equal to YES or NO. The default is YES. In *STEP 10kN central load Analysis procedure The analysis procedure (the type of simulation) must be defined immediately following the *STEP option block. In ABAQUS/Expli *DYNAMIC TEMPERATURE-DISPLACEMENT, EXPLICIT; and *ANNEAL. The *DYNAMIC TEMPERATURE-DISPLACEMENT procedu body, while the *ANNEAL procedure simulates the relaxation of stresses and plastic strains that occurs as metals are heated to a dynamic response of the structure over a period of 0.01 s. Thus, we will use *DYNAMIC, EXPLICIT. Add the following lines to yo *DYNAMIC, EXPLICIT , 0.01 The remaining input data in the step define the boundary conditions, loads, and desired output and can be given in any conveni Boundary conditions Boundary conditions are applied to those parts of the model where the displacements are known. Such parts may be constraine or may have specified, nonzero displacements. In either situation the constraints are applied directly to the nodes of the model. In some cases a node may be constrained completely and, thus, cannot move in any direction (for example, node 101 in our ca free to move in others. For example, node 103 is fixed in the vertical direction but is free to move in the horizontal direction. The freedom (dof). In the case of our two-dimensional hoist, each node can move in the global 1- and 2-directions; therefore, ther move out of plane, the problem would be three-dimensional, and each node would have three degrees of freedom. Nodes attac freedom representing the components of rotation and, thus, may have up to six degrees of freedom. The labeling convention used for the degrees of freedom in ABAQUS is shown in Figure 2-8 . Figure 2-8 Labeling convention for degrees of freedom. The degrees of freedom active at a node depend on the type of elements attached to that node. The two-dimensional truss elem translation in the 1- and 2-directions (dof 1 and dof 2). Constraints on nodes are defined by using the *BOUNDARY option and specifying the constrained degrees of freedom. Each da <node number>, <first dof>, <last dof>, <magnitude of displacement> The first degree of freedom and last degree of freedom are used to give a range of degrees of freedom that will be constrained. 101, 1, 3, 0.0 constrains degrees of freedom 1, 2, and 3 at node 101 to have zero displacement (the node cannot move in any of the global 1 If the magnitude of the displacement is not specified on the data line, it is assumed to be zero. If the node is constrained in one second field. For example, to constrain node 103 in the 2-direction (degree of freedom 2) only, any of the following data line fo 103, 2, 2, 0.0 or 103, 2, 2 or 103, 2 Boundary conditions on a node are cumulative. Thus, the following input constrains node 101 in both directions 1 and 2: 101, 1 101, 2

Rather than specifying each constrained degree of freedom, some of the more common constraints can be given directly using ENCASTRE Constraint on all displacements and rotations at a node. PINNED Constraint on all translational degrees of freedom. XSYMM Symmetry constraint about a plane of constant . YSYMM Symmetry constraint about a plane of constant . ZSYMM Symmetry constraint about a plane of constant . XASYMM Antisymmetry constraint about a plane of constant . YASYMM Antisymmetry constraint about a plane of constant . ZASYMM Antisymmetry constraint about a plane of constant . Thus, another way to constrain all the active degrees of freedom at node 101 in the hoist model is 101, ENCASTRE The complete *BOUNDARY option block for our hoist problem is *BOUNDARY 101, ENCASTRE 103, 2 In this example all of the constraints are in the global 1- or 2-directions. In many cases constraints are required in directions tha option can be used in such cases to define a local coordinate system for boundary condition application. Loading Loading is anything that causes the displacement or deformation of the structure, including: oncentrated loads; ressure loads; onzero boundary conditions; ody loads; and emperature (with thermal expansion of the material defined). In reality there is no such thing as a concentrated, or point, load; the load will always be applied over some finite area. However in that area, it is an appropriate idealization to treat the load as a concentrated load applied to a node. Concentrated loads are specified using the *CLOAD option. The data lines for this option have the form <node number>, <dof>, <load magnitude> In this simulation a load of -10 kN is applied in the 2-direction to node 102. The option block is the following: *CLOAD 102, 2, -10.E3 By default, ABAQUS/Explicit assumes that the load is applied instantaneously at the start of the analysis. Output requests Finite element analyses can create vast amounts of output. ABAQUS allows you to control and manage this output so that only d produced. Three types of output are available from ABAQUS/Explicit: utput database (ODB) output (.odb file). Writes a neutral binary file used by ABAQUS/Viewer for postprocessing. estart output (.res file). Used to continue the analysis, thus allowing a simulation to be run in stages. ile output (.fil file). Writes results in a binary format; primarily used for subsequent postprocessing with third-party software. You will use the first of these in the overhead hoist simulation. By default, ABAQUS/Explicit writes a preselected set of the most commonly used output variables to the output database file. A given in the ABAQUS/Explicit User's Manual. You do not need to add any output requests to accept these defaults. For this exam configuration. Since you have now finished defining all the data required for the step, use the *END STEP option to mark the end of the step: *END STEP The input file is now complete. Compare the input file you have generated to the complete input file given in Figure 2-2 . Save 2.4 Checking the input data Having generated the input file for this simulation, you are ready to run the analysis. Unfortunately, it is possible to have errors in

data. You should perform a datacheck analysis first before running the simulation. To run a datacheck analysis, make sure that y located, and type the following command: abaqus job=frame_xpl datacheck interactive If this command results in an error message, the ABAQUS installation on your computer has been customized. You should conta command to run ABAQUS. The job=frame_xpl parameter specifies that the jobname for this analysis is frame_xpl. All the files as identifier, which allows them to be recognized easily. The analysis will run interactively, and messages similar to the following will appear on your screen: ABAQUS JOB frame_xpl Begin Solver Input File Processor Tue Oct 17 14:46:06 2000 Run /usr/abaqus/exec/pre.x ABAQUS/EXPLICIT is running on a Category A machine and has checked out 4 Network Tokens. Tue Oct 17 14:46:24 2000 End Solver Input File Processor Begin ABAQUS/Explicit Packager Tue Oct 17 14:46:25 2000 Run /usr/abaqus/exec/package.x ABAQUS/EXPLICIT is running on a Category A machine and has checked out 4 Network Tokens. . . . Tue Oct 17 14:46:33 2000 End ABAQUS/Explicit Packager Begin ABAQUS/Explicit Analysis Tue Oct 17 14:46:33 2000 Run /usr/abaqus/exec/explicit.x ABAQUS/EXPLICIT is running on a Category A machine and has checked out 4 Network Tokens. Tue Oct 17 14:46:36 2000 End ABAQUS/Explicit Analysis ABAQUS JOB frame_xpl COMPLETED When the datacheck analysis is complete, you will find that a number of additional files have been created by ABAQUS. If any er will be written to the data file, frame_xpl.dat. This data file is a text file that can be viewed in an editor or printed. Part of this dat Figure 2-9 ABAQUS input file echo and list of options processed. Figure 2-10 Model data in data file. Figure 2-11 History data in data file. Figure 2-12 Summary of problem and file sizes. Try viewing the data file in a text editor. The file can contain lines up to 256 characters long, so the editor should be able to acc page that contains information about the version of ABAQUS used to run the analysis. The header page also contains the phone office or representative who can offer technical support and advice. Following this header page is an echo of the input file (Figure 2-9 ). The input data echo is generated by adding the option *PR ECHO is set to NO. Following the input data echo is a list of the options processed by ABAQUS (Figure 2-9 ). This is the first point at which error a with ***ERROR, while warnings begin with ***WARNING. Since these messages always begin the same way, searching the data error is a syntax problem (i.e., when ABAQUS cannot understand the input), the error message is followed by the line from the in The rest of the data file is a series of tables containing all of the model data (Figure 2-10 ) and the history data (Figure 2-11 ) These tables are generated by including the option *PREPRINT, MODEL=YES, HISTORY=YES in the input file. However, these tab By default, the parameters MODEL and HISTORY are set to NO. You should always check your model by setting MODEL=YES an interpreted by ABAQUS in the way you intended. Once you have confirmed that you have the correct input data, you can reduce Any warning or error messages are written to the data file, frame_xpl.dat, or written to your screen during an interactive analysis. warnings and errors are written to the status file, frame_xpl.sta. Finally, there is a summary of the size of the numerical model (see Figure 2-12 ). The comment at the end of the data file that s CALCULATOR HAS BEEN FOUND indicates that some of the requested output needs to be converted prior to postprocessing wit

conversion using the postprocessing calculator are discussed in ``The postprocessing calculator,'' Section 4.3.1 of the ABAQUS 2.5 Running the analysis Exit the editor, and make any necessary corrections to your input file. When the datacheck analysis completes with no error mes abaqus job=frame_xpl continue interactive Messages like the following will appear on the screen: ABAQUS JOB frame_xpl Begin ABAQUS/Explicit Analysis Tue Oct 17 14:53:20 2000 Run /usr/abaqus/exec/explicit.x ABAQUS/EXPLICIT is running on a Category A machine and has checked out 4 Network Tokens. . . . ------------------------------------------------------------------------------SOLUTION PROGRESS ------------------------------------------------------------------------------STEP 1 ORIGIN 0.00000E+00 Total memory used for step 1 is approximately 34.1 kilowords Global time estimation algorithm will be used. Scaling factor : 1.0000 STEP TOTAL CPU STABLE CRITICAL KINETIC INCREMENT TIME TIME TIME INCREMENT ELEMENT ENERGY 0 0.000E+00 0.000E+00 00:00:00 1.841E-04 13 0.000E+00 ODB Field Frame Number 0 of 20 requested intervals at increment zero. 3 5.524E-04 5.524E-04 00:00:00 1.840E-04 17 1.451E+01 ODB Field Frame Number 1 of 20 requested intervals at 5.523613E-04 6 1.104E-03 1.104E-03 00:00:00 1.835E-04 13 1.459E+01 ODB Field Frame Number 2 of 20 requested intervals at 1.103952E-03 9 1.654E-03 1.654E-03 00:00:00 2.141E-04 13 4.725E+00 . . . 49 1.000E-02 1.000E-02 00:00:01 2.131E-04 13 1.132E+01 ODB Field Frame Number 20 of 20 requested intervals at 1.000000E-02 Restart Number 1 at 1.00000E-02 THE ANALYSIS HAS COMPLETED SUCCESSFULLY

Tue Oct 17 14:53:25 2000 End ABAQUS/Explicit Analysis ABAQUS JOB frame_xpl COMPLETED It is desirable to perform a datacheck analysis before running a simulation to ensure that the input data are correct. However, it simulation by using the following command: abaqus job=frame_xpl interactive If a simulation is expected to take a substantial amount of time, it is convenient to run it in the background by omitting the intera abaqus job=frame_xpl (The above commands apply for the standard ABAQUS installation on a workstation. However, ABAQUS jobs can be run in batch systems administrator how to run ABAQUS on your system. Details of the ABAQUS commands are documented in Chapter 3, "Ex 2.6 Postprocessing with ABAQUS/Viewer Graphical postprocessing is important because of the great volume of data created during a simulation. ABAQUS/Viewer allows including deformed shape plots, contour plots, symbol plots, animations, and X-Y plots. All of these methods are discussed in t

features discussed in this guide, consult the ABAQUS/Viewer User's Manual. For this example you will use ABAQUS/Viewer to do the frame. Start ABAQUS/Viewer by typing abaqus viewer at the operating system prompt. The ABAQUS/Viewer window appears. Reading the output database file To begin this exercise, open the output database file that ABAQUS/Explicit generated during the analysis of the problem. To open the output database file: 1.From the main menu bar, select File->Open; or use the tool in the toolbar. The Open Database dialog box appears. 2.From the list of available output database files, select frame_xpl.odb. 3.Click OK. Tip: You can also open the output database frame_xpl.odb by typing abaqus viewer odb=frame_xpl at the operating system prompt. ABAQUS/Viewer displays a fast plot of the model. A fast plot is a basic representation of your undeformed model shape and is a Important: The fast plot does not display results and cannot be customized; for example, to display element and node numbers appearance of the model. The title block at the bottom of the viewport indicates the following: he description of the model (from the first line of the *HEADING option in the input file). he name of the output database (from the name of the analysis job). he product name (ABAQUS/Standard or ABAQUS/Explicit) and version used to generate the output database. he date the output database was last modified. The state block at the bottom of the viewport indicates the following: hich step is being displayed. he increment within the step. he step time. The view orientation triad indicates the orientation of the model in the global coordinate system. Displaying and customizing an undeformed shape plot You will now display the undeformed model shape and use the plot options to enable the display of node and element numberin From the main menu bar, select Plot->Undeformed Shape; or use the tool in the toolbox. ABAQUS/Viewer displays the undefor Figure 2-13 Undeformed model shape. To display node numbers: 1.From the main menu bar, select Options->Undeformed Shape. The Undeformed Shape Plot Options dialog box appears. 2.Click the Labels tab. 3.Toggle on Show node labels. 4.Click Apply. ABAQUS/Viewer applies the change and keeps the dialog box open. The customized undeformed plot is shown in Figure 2-14 . Figure 2-14 Node number plot. To display element numbers: 1.In the Undeformed Shape Plot Options dialog box, toggle on Show element labels. 2.Click OK. ABAQUS/Viewer applies the change and closes the dialog box. The resulting plot is shown in Figure 2-15 . Figure 2-15 Node and element number plot. To suppress the display of node and element numbers in the undeformed shape plot, repeat the above procedure and toggle of Displaying and customizing a deformed shape plot You will now display the deformed model shape and use the plot options to change the deformation scale factor and overlay the From the main menu bar, select Plot->Deformed Shape; or use the tool in the toolbox. ABAQUS/Viewer displays the deformed Figure 2-16 Deformed model shape. The scale factor for the deformed plot is displayed in the state block. By default, the displaced shape is not scaled. For this ana shape is evident. To change the deformation scale factor:

1.From the main menu bar, select Options->Deformed Shape. 2.In the Deformed Shape Plot Options dialog box, click the Basic tab if it is not already selected. 3.From the Deformation Scale Factor area, toggle on Uniform and enter 40.0 in the Value field. 4.Click Apply to redisplay the deformed shape. The plot title displays the new scale factor. To return to the default deformed plot, repeat the above procedure and, in the Deformation Scale Factor field, toggle on Auto-c To overlay the undeformed model shape on the deformed model shape: 1.In the Deformed Shape Plot Options dialog box, toggle on Superimpose undeformed plot. 2.Click OK. By default, ABAQUS/Viewer plots the undeformed model shape in green and the deformed model shape in white. Change the lin the two plots. To change the undeformed model shape's line style: 1.From the main menu bar, select Options->Undeformed Shape. The Undeformed Shape Plot Options dialog box appears. 2.Click the Color & Style tab. 3.In the Edge Attributes area, choose the dotted line style. 4.Click OK. The customized plot is shown in Figure 2-17 . Figure 2-17 Undeformed and deformed model shapes. To return the undeformed plot's line style to solid, repeat the above procedure and select the solid line style. Alternatively, to ret can click Defaults in the Undeformed Shape Plot Options dialog box. Checking history data with ABAQUS/Viewer By default, both the model data and history data are written to the output database file during the datacheck phase. Thus, you c before running the simulation. You have already learned how to draw plots of the model and to display the node and element nu the correct mesh. The boundary conditions applied to the overhead hoist model can also be displayed using ABAQUS/Viewer. To display boundary conditions on the undeformed model: 1.From the main menu bar, select Plot->Undeformed Shape; or use the tool in the toolbox. 2.From the main menu bar, select View->ODB Display Options. 3.In the ODB Display Options dialog box, click the BC tab. 4.Toggle on Show boundary conditions. 5.Click OK. ABAQUS/Viewer displays symbols to indicate the applied boundary conditions, as shown in Figure 2-18 . Figure 2-18 Applied boundary conditions on the overhead hoist. Exiting ABAQUS/Viewer From the main menu bar, select File->Exit to exit ABAQUS/Viewer. 2.7 Summary he ABAQUS input file contains a complete description of the analysis model. It is the means of communication between the pr (ABAQUS/Explicit). If you are using ABAQUS/CAE, you will not need to edit the input file since everything can be done within AB he input file contains two sections: the model data, which define the structure, and the history data, which define the loading. ach section of the input file comprises a number of option blocks, each consisting of a keyword line that may be followed by ou can perform a datacheck analysis once you have created the input file. Along with descriptions of how ABAQUS/Explicit ha in the data (.dat) and status (.sta) files. se ABAQUS/Viewer to verify the model geometry and boundary conditions graphically, using the output database file generate t is often easiest to check the data file for mistakes in the material definition, section properties, etc. Use *PREPRINT, MODEL o run the analysis after a successful datacheck analysis without repeating the datacheck phase, use the following command: a he output database file contains results for graphical postprocessing with ABAQUS/Viewer. BAQUS/Viewer allows you to visualize analysis results graphically in a variety of ways. 3. Overview of Explicit Dynamics The explicit dynamics procedure can be an effective tool for solving a wide variety of nonlinear solid and structural mechanics p ABAQUS/Standard. From a user standpoint, the distinguishing characteristics of the explicit and implicit methods are: xplicit methods require a small time increment size that depends solely on the highest natural frequencies of the model and is generally take on the order of 10,000 to 1,000,000 increments, but the computational cost per increment is relatively small. mplicit methods do not place an inherent limitation on the time increment size; increment size is generally determined from ac typically take orders of magnitude fewer increments than explicit simulations. However, since a global set of equations must be

method is far greater than that of an explicit method. Knowing these characteristics of the two procedures can help you decide which methodology is appropriate for your problems. 3.1 Types of problems suited for ABAQUS/Explicit Before discussing how the explicit dynamics procedure works, it is helpful to understand what classes of problems are well-suit incorporated pertinent examples of the following classes of problems commonly performed in ABAQUS/Explicit: High-speed dynamic events The explicit dynamics method was originally developed to analyze high-speed dynamic events that can be extremely costly to an example of such a simulation, the effect of a short-duration blast load on a steel plate is analyzed in Chapter 5, "Materials." Sinc the structure changes rapidly. Accurate tracking of stress waves through the plate is important for capturing the dynamic respon frequencies of the system, obtaining an accurate solution requires many small time increments. Complex contact problems Contact conditions are formulated more easily using an explicit dynamics method than using an implicit method. The result is tha complex contact interaction between many independent bodies. ABAQUS/Explicit is particularly well-suited for analyzing the tran loads and subsequently undergo complex contact interaction within the structure. An example of such a problem is the circuit bo circuit board sitting in foam packaging is dropped on the floor from a height of 1 m. The problem involves impact between the p conditions between the circuit board and the packaging. Complex postbuckling problems Unstable postbuckling problems are solved readily in ABAQUS/Explicit. In such problems the stiffness of the structure changes d includes the effects of contact interactions. An illustration of both situations is the tube crushing example in Chapter 6, "Contact itself, causing self-contact on both its inside and outside surfaces. Highly nonlinear quasi-static problems For a variety of reasons ABAQUS/Explicit is often very efficient in solving certain classes of problems that are essentially static. contact such as forging, rolling, and sheet-forming generally fall within these classes. Sheet forming problems usually include v frictional contact conditions. Bulk forming problems are characterized by large distortions, flash formation, and contact interacti simulation is presented in Chapter 7, "Quasi-Static Analysis." Materials with degradation and failure Material degradation and failure often lead to severe convergence difficulties in implicit analysis programs, but ABAQUS/Explicit the concrete cracking model, in which tensile cracking causes the material stiffness to become negative. An example of materia stiffness can degrade until it reduces to zero, at which time the elements are removed from the model entirely. Each of these types of analyses can include temperature and heat transfer effects. 3.2 Finite element method for explicit dynamics This section contains a conceptual and an algorithmic description of the ABAQUS/Explicit solver as well as a discussion on the 3.2.1 Stress wave propagation illustrated This section attempts to provide some conceptual understanding of how forces propagate through a model when using the expl propagation of a stress wave along a rod modeled with three elements, as shown in Figure 3-1 . We study the state of the rod Figure 3-1 Initial configuration of a rod with a concentrated load, , at the free end. In the first time increment node 1 has an acceleration, , as a result of the concentrated force, , applied to it. The acceleration ca rate, , in element 1. The increment of strain, , in element 1 is obtained by integrating the strain rate through the time of incremen increment in strain. In this case the initial strain is zero. Once the element strain has been calculated, the element stress, , is ob elastic material the stress is simply the elastic modulus times the total strain. This process is shown in Figure 3-2 . Nodes 2 and them. Figure 3-2 Configuration at the end of increment 1 of a rod with a concentrated load, , at the free end. In the second increment the stresses in element 1 apply internal, element forces to the nodes associated with element 1, as sho calculate dynamic equilibrium at nodes 1 and 2. Figure 3-3 Configuration of the rod at the beginning of increment 2. The process continues so that at the start of the third increment there are stresses in both elements 1 and 2, and there are force continues until the analysis reaches the desired total time. Figure 3-4 Configuration of the rod at the beginning of increment 3. 3.2.2 Time integration ABAQUS/Explicit uses a central difference rule to integrate the equations of motion explicitly through time, using the kinematic c at the next increment. At the beginning of the increment the program solves for dynamic equilibrium, which states that the nodal nodal forces (the difference between the external applied forces, , and internal element forces, ): The accelerations at the beginning of the current increment (time ) are calculated as Since the explicit procedure always uses a diagonal, or lumped, mass matrix, solving for the accelerations is trivial; there are no determined completely by its mass and the net force acting on it, making the nodal calculations very inexpensive.

The accelerations are integrated through time using the central difference rule, which calculates the change in velocity assuming added to the velocity from the middle of the previous increment to determine the velocities at the middle of the current incremen The velocities are integrated through time and added to the displacements at the beginning of the increment to determine the di Thus, satisfying dynamic equilibrium at the beginning of the increment provides the accelerations. Knowing the accelerations, th time. The term ``explicit'' refers to the fact that the state at the end of the increment is based solely on the displacements, veloc method integrates constant accelerations exactly. For the method to produce accurate results, the time increments must be quit increment. Since the time increments must be small, analyses typically require many thousands of increments. Fortunately, each equations to solve. Most of the computational expense lies in the element calculations to determine the internal forces of the ele determining element strains and applying material constitutive relationships (the element stiffness) to determine element stresses Here is a summary of the explicit dynamics algorithm: 1.Nodal calculations. a.Dynamic equilibrium. b.Integrate explicitly through time. 2.Element calculations. a.Compute element strain increments, , from the strain rate, . b.Compute stresses, , from constitutive equations. c.Assemble nodal internal forces, . 3.Set to and return to Step 1. 3.2.3 Advantages of the explicit time integration method The explicit method is especially well-suited to solving high-speed dynamic events that require many small increments to obtain the solution can be obtained efficiently. Contact conditions and other extremely discontinuous events are readily formulated in the explicit method and can be enforced accelerations can be adjusted to balance the external and internal forces during contact. The most striking feature of the explicit method is the lack of a global tangent stiffness matrix, which is required with implicit me iterations and tolerances are not required. 3.3 Automatic time incrementation and stability The stability limit dictates the maximum time increment used by the ABAQUS/Explicit solver. It is a critical factor in the performan stability limit and discuss how ABAQUS/Explicit determines this value. Issues surrounding the model design parameters that affe include the model mass, material, and mesh. 3.3.1 Conditional stability of the explicit method With the explicit method the state of the model is advanced through an increment of time, , based on the state of the model at t state can be advanced and still remain an accurate representation of the problem is typically quite short. If the time increment is to have exceeded the stability limit. A possible effect of exceeding the stability limit is a numerical instability, which may lead to the stability limit exactly, so conservative estimates are used instead. The stability limit has a great effect on reliability and accur computational efficiency ABAQUS/Explicit chooses the time increments to be as close as possible to the stability limit without ex 3.3.2 Definition of the stability limit The stability limit is defined in terms of the highest frequency in the system (). Without damping the stability limit is defined by th and with damping it is defined by the expression where is the fraction of critical damping in the mode with the highest frequency. (Critical damping defines the limit between osc vibration. ABAQUS/Explicit always introduces a small amount of damping in the form of bulk viscosity to control high-frequency always reduces the stability limit. The actual highest frequency in the system is based on a complex set of interacting factors, and it is not computationally feasib estimate that is efficient and conservative. Instead of looking at the global model, we estimate the highest frequency of each ind dilatational mode. It can be shown that the highest element frequency determined on an element-by-element basis is always gr finite element model. Based on the element-by-element estimate, the stability limit can be redefined using the element length, , and the wave speed For most element types--a distorted quadrilateral element, for example--the above equation is only an estimate of the actual e element length should be determined. As an approximation the shortest element distance can be used, but the resulting estimate smaller the stability limit. The wave speed is a property of the material. For a linear elastic material with a Poisson's ratio of zero where is Young's modulus and is the mass density. The stiffer the material, the higher the wave speed, resulting in a smaller st resulting in a larger stability limit. Our simplified stability limit definition provides some intuitive understanding. The stability limit is the transit time of a dilatational length. If we know the size of the smallest element dimension and the wave speed of the material, we can estimate the stability the dilatational wave speed is 5000 m/s, the stable time increment is on the order of 1 ?10-6 s. 3.3.3 Automatic time incrementation in ABAQUS/Explicit

ABAQUS/Explicit uses equations such as those discussed in the previous section to adjust the time increment size throughout th the model, is never exceeded. Time incrementation is automatic and requires no user intervention, not even a suggested initial t resulting from the numerical model. Since the finite element program has all of the relevant details, it can determine an efficient allow the user to override the automatic time incrementation, if desired. ``Summary,'' Section 3.9, briefly discusses manual time 3.3.4 Mass scaling to control time incrementation Since the mass density influences the stability limit, under some circumstances scaling the mass density can potentially increase complex discretization of many models, there are often regions containing very small or poorly shaped elements that control the and may exist in localized areas. By increasing the mass of only these controlling elements, the stability limit can be increased s the model may be negligible. The automatic mass scaling features in ABAQUS/Explicit can keep offending elements from hindering the stability limit. There ar scaling factor directly or defining a desired element-by-element stable time increment for the elements whose mass is to be sca ABAQUS/Explicit User's Manual, permit additional user control over the stability limit. However, use caution when employing mas change the physics of the problem. 3.3.5 Effect of material on stability limit The material model affects the stability limit through its effect on the dilatational wave speed. In a linear material the wave speed during the analysis result from changes in the smallest element dimension during the analysis. In a nonlinear material, such as a yields and the stiffness of the material changes. ABAQUS/Explicit monitors the effective wave speeds in the model throughout th for stability estimates. After yielding, the stiffness decreases, reducing the wave speed and, consequently, increasing the stabili 3.3.6 Effect of mesh on stability limit Since the stability limit is roughly proportional to the shortest element dimension, it is advantageous to keep the element size as mesh is often necessary. To obtain the highest possible stability limit while using the required level of mesh refinement, the best the stability limit is based on the smallest element dimension in the model, even a single small or poorly shaped element can red ABAQUS/Explicit provides a list in the status (.sta) file of the 10 elements in the mesh with the lowest stability limit. If the model than those of the rest of the mesh, remeshing the model more uniformly may be worthwhile. 3.3.7 Numerical instability ABAQUS/Explicit remains stable for most elements under most circumstances. It is possible, however, to define spring and dash of an analysis. Therefore, it is useful to be able to recognize a numerical instability if it occurs in your analysis. If it does occur, oscillatory solutions. The examples at the end of this chapter use simple spring and dashpot problems to illustrate instabilities. 3.4 Example: stress wave propagation in a bar This example demonstrates some of the fundamental ideas in explicit dynamics described earlier in the discussion of how stress limits and the effect of mesh refinement and materials on the solution time. The bar has the dimensions shown in Figure 3-5 . Figure 3-5 Problem description for wave propagation in a bar. To make the problem a one-dimensional strain problem, all four lateral faces are on rollers; thus, the three-dimensional model s the properties shown in Figure 3-5 . The free end of the bar is subjected to a blast load with a magnitude of 1.0 ?105 Pa, as sh 3.88 ?10-5 s. Figure 3-6 Blast amplitude versus time. Using the material properties (neglecting Poisson's ratio), we can calculate the wave speed of the material using the equations i At this speed the wave passes to the fixed end of the bar in 1.94 ?10-4 s. Since we would like to see the stress propagate along refined mesh to capture the stress wave accurately. It seems adequate to have the blast load take place over the span of 10 ele we would like the blast duration times the wave speed to equal the length of 10 elements: The length of 10 elements is 0.2 m. Since the length of the bar is 1.0 m, this means that we would like to have 50 elements alon elements in each of the transverse directions, making the mesh 50 ?10 ?10. This mesh is shown in Figure 3-7 . Figure 3-7 50 ?10 ?10 mesh. Create this mesh in your preprocessor. Use the coordinate system shown in Figure 3-7 . 3.4.1 Node and element sets You will need to have node and element sets defined to apply the loads and boundary conditions and to visualize output. Define . Figure 3-8 Node sets. Define the element sets shown in Figure 3-9 . Figure 3-9 Element sets for modeling. In addition, define an element set containing three elements in the center of the bar. Select these elements such that their faces m from the free end, as shown in Figure 3-10 . These elements will be used for postprocessing. Figure 3-10 Element sets for postprocessing. 3.4.2 Reviewing the input file--the model data

At this point we assume that you have created the basic mesh using your preprocessor. In this section you will review your input Model description The following would be a suitable description in the *HEADING option for this simulation: *HEADING Stress wave propagation in a bar -- 50x10x10 elements SI units (kg, m, s, N) Element connectivity Check to make sure that you are using the correct element type (C3D8R). It is possible that the preprocessor specified the elem should begin with the following: *ELEMENT, TYPE=C3D8R, ELSET=BAR The name given for the ELSET parameter in your model may not be BAR. If necessary, change the name to BAR. Section properties The section properties are the same for all elements. Use element set BAR to assign the material properties to the elements. *SOLID SECTION, ELSET=BAR, MATERIAL=STEEL Material properties The bar is made of steel, which we assume to be linear elastic with a Young's modulus of 207 ?109 Pa, a Poisson's ratio of 0.3, block in your model: *MATERIAL, NAME=STEEL *ELASTIC 207.0E9, 0.3 *DENSITY 7800.0, Fixed boundary conditions Fix all the translations at the built-in, right-hand end of the bar. Constrain the front, back, top, and bottom faces of the bar so t node sets that you defined previously, apply the following boundary conditions to your model: *BOUNDARY NFIX, 1, 3 NFRONT, 3, 3 NBACK, 3, 3 NTOP, 2, 2 NBOT, 2, 2 Amplitude definition The blast load is applied at its maximum value instantaneously and is held constant for 3.88 ?10-5 s. Then the load is suddenly used to define the time variation of loads and boundary conditions. On the data lines following the *AMPLITUDE option, pairs of <time>, <amplitude>, <time>, <amplitude>, etc. Up to four data pairs can be entered on each data line. ABAQUS considers the amplitude to be held constant following the last follows: *AMPLITUDE, NAME=BLAST 0., 1., 3.88E-5, 1., 3.89E-5, 0, 3.90E-5, 0. 3.4.3 Reviewing the input file--the history data We will now review the history data associated with this problem, including the step definition, loading, bulk viscosity, and outpu Step definition The step definition indicates that this is an explicit dynamics analysis with a duration of 2.0 ?10-4 s. You can also include a des *STEP Blast loading *DYNAMIC, EXPLICIT , 2.0E-4 Loading Apply the pressure load with a value of 1.0 ?105 Pa to the free face of the bar, which you previously defined to be in an element magnitude specified under the *DLOAD option times the value interpolated from the amplitude curve. To apply the load correctly element faces. For the model defined in ``Stress wave propagation in a bar,'' Section A.2, the free face is face number 3, which depends on the order in which the nodes are defined on the *ELEMENT option, as shown in Figure 3-11 . Use the amplitude na *DLOAD, AMPLITUDE=BLAST ELOAD, <P1, P2, P3, P4, P5, or P6>, 1.0E5 If you define the pressure load in your preprocessor, the correct face identifier should be determined automatically.

Figure 3-11 Face label identifier for a C3D8R element. Bulk viscosity To keep the stress wave as sharp as possible, set the quadratic bulk viscosity (discussed in ``Bulk viscosity,'' Section 3.6.1) to z *BULK VISCOSITY , 0.0 Output requests By default, many preprocessors create an ABAQUS input file that has a large number of output request options. If, when you ed created, delete them all because they will generally generate too much output. You want to have an output database file created during the analysis so that you can use ABAQUS/Viewer to postprocess the re written to the output database) are adequate to show the stress wave propagating through the mesh. Set the parameter VARIABL field data for a *DYNAMIC, EXPLICIT procedure to the output database file. In addition, request stress (S) history output in eleme *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=4 *OUTPUT, HISTORY, TIME INTERVAL=1.0E-6 *ELEMENT OUTPUT, ELSET=EOUT S, *END STEP 3.4.4 Running the analysis After storing your input in a file called wave_50x10x10.inp, run the analysis using the following command: abaqus job=wave_50x10x10 If your analysis does not complete, check the data file, wave_50x10x10.dat, and status file, wave_50x10x10.sta, for error messa trouble running your analysis, compare your input file to the one given in ``Stress wave propagation in a bar,'' Section A.2. Status file The status file, wave_50x10x10.sta, contains information about moments of inertia, followed by information concerning the initia are listed in rank order. Most critical elements : Element number Rank Time increment Increment ratio ---------------------------------------------------------101 1 1.931897E-06 1.000000E+00 105 2 1.931897E-06 1.000000E+00 107 3 1.931897E-06 1.000000E+00 111 4 1.931897E-06 1.000000E+00 115 5 1.931897E-06 1.000000E+00 119 6 1.931897E-06 1.000000E+00 401 7 1.931897E-06 1.000000E+00 405 8 1.931897E-06 1.000000E+00 407 9 1.931897E-06 1.000000E+00 The status file continues with information about the progress of the solution. STEP 1 ORIGIN 0.00000E+00 Total memory used for step 1 is approximately 2.8 megawords Global time estimation algorithm will be used. Scaling factor : 1.0000 STEP TOTAL CPU STABLE CRITICAL KINETIC INCREMENT TIME TIME TIME INCREMENT ELEMENT ENERGY 0 0.000E+00 0.000E+00 00:00:02 1.932E-06 101 0.000E+00 Results number 0 at increment zero. ODB Field Frame Number 0 of 4 requested intervals at increment zero. 6 1.159E-05 1.159E-05 00:00:05 1.932E-06 401 4.487E-05 12 2.318E-05 2.318E-05 00:00:07 1.932E-06 101 9.566E-05 16 3.332E-05 3.332E-05 00:00:09 3.104E-06 101 1.389E-04 20 4.572E-05 4.572E-05 00:00:10 3.088E-06 101 1.712E-04 22 5.189E-05 5.189E-05 00:00:11 3.082E-06 101 1.695E-04 ODB Field Frame Number 1 of 4 requested intervals at 5.188920E-05 26 6.420E-05 6.420E-05 00:00:13 3.071E-06 101 1.699E-04

. . . 3.4.5 Postprocessing Start ABAQUS/Viewer by typing abaqus viewer odb=wave_50x10x10 at the operating system prompt. Plotting the stress along a path We would like to look at how the stress distribution along the length of the bar changes with time. To do so, we will look at the s the analysis. Create a curve of the stress in the 1-direction (S11) approximately along the center of the bar for each of the first three frames along the center of the bar using a node list. To create a node list path along the center of the bar: 1.From the main menu bar, select Tools->Path->Create. The Create Path dialog box appears. 2.Name the path Center. Accept the default selection of Node list as the path type, and click Continue. The Edit Node List Path dialog box appears. 3.In the Node Labels table, enter the nodes along the center of the bar and click OK. For example, if you created your mesh usin Section A.2, the table entry is 100011:105011:100. (This input specifies a range of nodes from 100011 to 105011 in increments Tip: If you did not use the input file shown in ``Stress wave propagation in a bar,'' Section A.2, and need to identify the nodes a turn on the node labels using the undeformed plot options. From the main menu bar, select Plot->Undeformed Shape; then sele Plot Options dialog box. You can use the view manipulation tools in the toolbar to alter the default isometric view of the bar. To save X-Y plots of stress along the path at three different times: 1.From the main menu bar, select Tools->XY Data->Manager. 2.In the XY Data Manager dialog box that appears, click Create. The Create XY Data dialog box appears. 3.Choose Path as the X-Y data source, and click Continue. The Create XY Data from Path dialog box appears with the path that you created visible in the list of available paths. If the unde highlighted in the plot. 4.Accept True distance as the selection in the X Values portion of the dialog box. 5.Click Field Output in the Y Values portion of the dialog box to open the Field Output dialog box. 6.Select the S11 stress component, and click OK. The field output variable in the Create XY Data from Path dialog box changes to indicate that stress data in the 1-direction (S11 Note: ABAQUS/Viewer warns you that the field output variable will not affect the current image. Leave the plot mode As is, and 7.Click Step/Frame in the Y Values portion of the Create XY Data from Path dialog box. 8.In the Step/Frame dialog box that appears, choose the second of the five recorded frames. (The first frame listed, frame 0, is OK. The Y Values portion of the Create XY Data from Path dialog box changes to indicate that data from Step 1, frame 1 will be crea 9.To save the X-Y data, click Save As. The Save XYData As dialog box appears. 10.Name the X-Y data S11_T1, and click OK. S11_T1 appears in the XY Data Manager. 11.Repeat Steps 7 through 10 to create X-Y data for frames 2 and 3. Name the data sets S11_T2 and S11_T3, respectively. 12.To close the Create XY Data from Path dialog box, click Cancel. To plot the stress curves: 1.In the XY Data Manager dialog box, drag the cursor to highlight all three X-Y data sets. 2.Click Plot. ABAQUS/Viewer plots the stress in the 1-direction along the center of the bar for frames 1, 2, and 3, corresponding to simulatio respectively. Customize the plot by changing tick mark intervals, axis titles, and the curve line styles. To customize the X-Y plot: 1.From the main menu bar, select Options->XY Plot. The XY Plot Options dialog box appears. 2.Click the Tick Marks tab. The Tick Marks options become available.

3.Specify that the Y-axis major tick marks occur at 20 ?103 s increments. 4.For both the X-axis and Y-axis minor tick mark options, specify zero minor tick marks between each major tick mark interval. You can also customize the axis titles. 5. Click the Titles tab to make the Titles options available. 6.In the X-axis area, select a User-specified title source. In the Title text field, enter Distance along bar. 7.In the Y-axis area, specify Stress - S11 as the Y-axis title. 8.Click OK to apply the customized X-Y plot parameters and to close the XY Plot Options dialog box. To customize the appearance of the curves in the X-Y plot: 1.From the main menu bar, select Options->XY Curve. The XY Curve Options dialog box appears. 2.In the XY Data field, select S11_T2. 3.Choose the dotted line style for the S11_T2 curve. The S11_T2 curve becomes dotted. 4.Repeat Steps 2 and 3 to make the S11_T3 curve dashed. 5.Click Dismiss to close the XY Curve Options dialog box. The customized plot appears in Figure 3-12 . Figure 3-12 Stress (S11) along the bar at three different times. We can see that the length of the bar affected by the stress wave is approximately 0.2 m in each of the three curves. This distan during its time of application, which can be checked by a simple calculation. If the length of the wave front is 0.2 m and the wav 0.2 m is 3.88 ?10-5 s. As expected, this is the duration of the blast load that we applied. The stress wave is not exactly square oscillation of the stress behind the sudden changes in stress. Linear bulk viscosity, discussed later in this chapter, damps the rin Creating a history plot Another way to study the results is to view the time history of stress at three different points within the bar. To plot the stress history of the elements in set EOUT: 1.From the main menu bar, select Result->History Output. ABAQUS/Viewer displays the History Output dialog box. The Output Variables field contains a list of all the variables in the histor can plot. To see the complete description of the variable choices, increase the width of the History Output dialog box by draggin 2.Drag the cursor to highlight stress in the 1-direction (S11) for all three elements in element set EOUT. 3.At the bottom of the History Output dialog box, click Plot. ABAQUS/Viewer displays an X-Y plot of the longitudinal stress in each element versus time. 4.Click Dismiss to close the dialog box. As before, you can customize the appearance of the plot. To customize the X-Y plot: 1.From the main menu bar, select Options->XY Plot. The XY Plot Options dialog box appears. 2.Click the Titles tab. The Titles options become available. 3.In the X-axis area, specify Total time as the X-axis title. 4. Click OK to apply the customized X-Y plot options and to close the dialog box. To customize the appearance of the curves in the X-Y plot: 1.From the main menu bar, select Options->XY Curve. The XY Curve Options dialog box appears. 2. In the XY Data field, select the temporary X-Y data label that corresponds to the element closest to the free end of the bar. (O wave.) 3.Select a User-specified legend source. 4.In the Legend text field, enter S11-0.25. 5.In the XY Data field, select the temporary X-Y data label that corresponds to the element in the middle of the bar. (This is the 6.Specify S11-0.5 as the curve legend text, and change the curve style to dotted. 7.In the XY Data field, select the temporary X-Y data label that corresponds to the element closest to the fixed end of the bar. (T 8.Specify S11-0.75 as the curve legend text, and change the curve style to dashed. 9.Click Dismiss to close the dialog box. The customized plot appears in Figure 3-13 . Figure 3-13 Time history of stress (S11) at three points along the length of the bar (0.25 m, 0.5 m, and 0.75 m). In the history plot we can see that stress at a given point increases as the stress wave travels through the point. Once the stres point oscillates about zero.

3.4.6 How the mesh affects the stable time increment and CPU time In ``Automatic time incrementation and stability,'' Section 3.3, we discussed how mesh refinement affects the stability limit and t propagation problem. We began with a reasonably refined mesh of square elements with 50 elements along the length and 10 e purposes, we will now use a coarse mesh of 25 ?5 ?5 elements and observe how refining the mesh in the various directions cha Figure 3-14 Meshes from least to most refined. Table 3-1 shows how the CPU time (normalized with respect to the coarse mesh model result) changes with mesh refinement results, based on the simplified stability equations presented in this guide; the second half of the table provides the results obta workstation. Table 3-1 Mesh refinement and solution time. Mesh Simplified Theory Actual (s) Number of Elements CPU Time (s) Max (s) Number of Elements Normalized CPU Time 25 ?5 ?5 A B C 6.06e-6 625 1 50 ?5 ?5 A/2 2B 4C 3.14e-6 1250 4 50 ?10 ?5 A/2 4B 8C 3.12e-6 2500 8.33 50 ?10 ?10 A/2 8B 16C 3.11e-6 5000 16.67 For the theoretical results we choose the coarsest mesh, 25 ?5 ?5, as the base state, and we define the stable time increment, t C, respectively. As the mesh is refined, two things happen: the smallest element dimension decreases, and the number of eleme CPU time. In the first level of refinement, the 50 ?5 ?5 mesh, the smallest element dimension is cut in half and the number of ele over the previous mesh. However, further doubling the mesh to 50 ?10 ?5 does not change the smallest element dimension; it o increases by only a factor of two over the 50 ?5 ?5 mesh. Further refining the mesh so that the elements are uniform and square and the CPU time. This simplified calculation predicts quite well the trends of how mesh refinement affects the stable time increment and CPU time predicted and actual stable time increment values. First, recall that we made the approximation that the stable time increment is We then assumed that the characteristic element length, , is the smallest element dimension, whereas ABAQUS/Explicit actually

size and shape of the element. Another complication is that ABAQUS/Explicit employs a global stability estimator, which allows a difficult to predict the stable time increment accurately before running the analysis. However, since the trends follow nicely from stable time increment will change with mesh refinement. 3.4.7 How the material affects the stable time increment and CPU time The same wave propagation analysis performed on different materials would take different amounts of computer time, dependin change the material from steel to aluminum, the wave speed would change from 5.15 ?103 m/s to The change from aluminum to steel has minimal effect on the stable time increment, because the stiffness and the density differ more substantial, as the wave speed decreases to which is approximately one-fifth the wave speed of steel. The stable time increment for the lead bar would be five times the stab 3.5 Comparison of explicit and implicit procedures ABAQUS/Explicit and ABAQUS/Standard are capable of solving a wide variety of problems. The characteristics of implicit and ex given problem. For those problems that can be solved with either method, the question of which solver to use has a direct beari Understanding the characteristics of implicit and explicit procedures will help you answer this question. 3.5.1 General comparison For both the explicit and the implicit time integration procedures, equilibrium is defined in terms of the external applied forces, , where is the mass matrix. Both procedures solve for nodal accelerations and use the same element calculations to determine th procedures lies in the manner in which the nodal accelerations are computed. In the implicit procedure a set of linear equations solving this set of equations is high when compared to the relatively low cost of the nodal calculations with the explicit method. 3.5.2 Review of implicit procedure in ABAQUS/Standard ABAQUS/Standard uses automatic incrementation based on the full Newton iterative solution method. Newton's method seeks to and compute displacements at the same time. The time increment, , is relatively large compared to that used in the explicit meth nonlinear problem each increment typically requires several iterations to obtain a solution within the prescribed tolerances. Each displacements, . Each iteration requires the solution of a set of simultaneous equations, which is an expensive procedure for large models. The effective stiffness matrix, , is a linear combination of the tangent stiffnes continue until several quantities--force residual, displacement correction, etc.--are within the prescribed tolerances. For a smo convergence, as illustrated below: Iteration Relative Error 1 1 2 10-2 3 10-4 . . . . . . However, if the model contains highly discontinuous processes, such as contact and frictional sliding, quadratic convergence m Cutbacks in the time increment size may become necessary to satisfy equilibrium. In extreme cases the resulting time increment typical stable time increment for an explicit analysis, while still carrying the high solution cost of implicit iteration. In some cases Each iteration in an implicit analysis requires solving a large system of linear equations, a procedure that requires considerable c equation solver requirements are dominant over the requirements of the element and material calculations, which are similar for the equation solver requirements grow rapidly so that, in practice, the maximum size of an implicit analysis that can be solved on and memory available on the machine rather than by the required computation time. 3.5.3 Advantages of ABAQUS/Explicit For many analyses it is clear whether ABAQUS/Standard or ABAQUS/Explicit should be used. For example, earlier in this chapter propagation analysis; on the other hand, ABAQUS/Standard is more efficient for solving smooth nonlinear problems. There are, simulated well with either program. Typically, these are problems that usually would be solved with ABAQUS/Standard but may h complexities, resulting in a large number of iterations. Such analyses are expensive in ABAQUS/Standard because each iteration solved. Whereas ABAQUS/Standard must iterate to determine the solution to a nonlinear problem, ABAQUS/Explicit determines the solut from the previous increment. Even though a static analysis requires a large number of time increments using the explicit method

same analysis in ABAQUS/Standard would require many expensive iterations. Another advantage of ABAQUS/Explicit is that it requires much less disk space and memory than ABAQUS/Standard for the sam of the two programs may be comparable, the substantial disk space and memory savings of ABAQUS/Explicit make it attractive. 3.5.4 Cost of mesh refinement in explicit and implicit analyses Using the explicit method, the computational cost is proportional to the number of elements and roughly inversely proportional to increases the computational cost by increasing the number of elements and reducing the smallest element dimension. As an exa elements. If the mesh is refined by a factor of two in all three directions, the computational cost increases by a factor of 2 ?2 ?2 of 2 as a result of the decrease in the smallest element dimension. The total computational cost of the analysis increases by a fa requirements are proportional to the number of elements with no dependence on element dimensions; thus, these requirements Whereas predicting the cost increase with mesh refinement for the explicit method is rather straightforward, cost is more difficul from the problem-dependent relationship between element connectivity and solution cost, a relationship that does not exist in th that for many problems the computational cost is roughly proportional to the square of the number of degrees of freedom. Cons uniform, square elements. Refining the mesh by a factor of two in all three directions increases the number of degrees of freedo increase by a factor of roughly (23)2, or 64. The disk space and memory requirements increase in the same manner, although th The explicit method shows great cost savings over the implicit method as the model size increases, as long as the mesh is relat versus model size using the explicit and implicit methods. For this problem the number of degrees of freedom scales with the nu Figure 3-15 Cost versus model size in using the explicit and implicit methods. 3.6 Damping of dynamic oscillations There are two reasons for adding damping to a model: to limit numerical oscillations or to add physical damping to the system. damping into the analysis. 3.6.1 Bulk viscosity Bulk viscosity introduces damping associated with volumetric straining. Its purpose is to improve the modeling of high-speed dy forms of bulk viscosity. You can set bulk viscosity to nondefault values from step to step by using the *BULK VISCOSITY option, pressure is not included in the material point stresses because it is intended as a numerical effect only. As such, it is not consid Linear bulk viscosity By default, linear bulk viscosity is always included to damp "ringing" in the highest element frequency. It generates a bulk viscos the following equation: where is a damping coefficient, whose default value is 0.06, is the current material density, is the current dilatational wave spe strain rate. Quadratic bulk viscosity Quadratic bulk viscosity is included only in continuum elements (except for the plane stress element, CPS4R) and is applied only pressure is quadratic in the strain rate, according to the following equation: where is the damping coefficient, whose default value is 1.2. The quadratic bulk viscosity smears a shock front across several elements and is introduced to prevent elements from collapsing element problem in which the nodes on one side of the element are fixed and the nodes on the other side have an initial velocity The stable time increment size is precisely the transit time of a dilatational wave across the element. Therefore, if the initial noda the element collapses to zero volume in one time increment. The quadratic bulk viscosity pressure introduces a resisting pressur Figure 3-16 Element with fixed nodes and prescribed velocities. Fraction of critical damping due to bulk viscosity The bulk viscosity pressures are based on only the dilatational modes of each element. The fraction of critical damping in the hi where is the fraction of critical damping. The linear term alone represents 6% of critical damping, whereas the quadratic term is 3.6.2 Viscous pressure Viscous pressure loads are commonly used in structural problems and quasi-static problems to damp out the low-frequency dy minimal number of increments. These loads are applied as distributed loads (*DLOAD) defined by the following formula: where is the pressure applied to the body; is the viscosity, given on the data line as the magnitude of the load; is the velocity applied; and is the unit outward normal to the surface at the same point. For typical structural problems it is not desirable to ab percentage--perhaps 1 or 2 percent--of the quantity as an effective way of minimizing ongoing dynamic effects. 3.6.3 Material damping The material model itself may provide damping in the form of plastic dissipation or viscoelasticity. For many applications such da damping defined using the *DAMPING option, which is part of the *MATERIAL option block. There are two damping factors asso and for stiffness proportional damping. Mass proportional damping The factor defines a damping contribution proportional to the mass matrix for an element. The damping forces that are introduc The resulting effect can be likened to the model moving through a viscous fluid so that any motion of any point in the model trig does not reduce the stability limit significantly.

Stiffness proportional damping The factor defines damping proportional to the elastic material stiffness. A ``damping stress,'' , proportional to the total strain ra where is the strain rate. For hyperelastic and hyperfoam materials is defined as the initial elastic stiffness. For all other material added to the stress caused by the constitutive response at the integration point when the dynamic equilibrium equations are form introduced for any nonlinear analysis and provides standard Rayleigh damping for linear analyses. For a linear analysis stiffness matrix equal to times the stiffness matrix. Stiffness proportional damping must be used with caution because it may significantly 3.6.4 Discrete dashpots Yet another option is to define individual dashpot elements. Each dashpot element provides a damping force proportional to the is that it enables you to apply damping only at points where you decide it is necessary. Dashpots always should be used in para do not cause a significant reduction in the stability limit. 3.7 Energy balance Energy output is often an important part of an ABAQUS/Explicit analysis. Comparisons between various energy components can appropriate response. 3.7.1 Statement of energy balance An energy balance for the entire model can be written as where is the internal energy, is the viscous energy dissipated, is the frictional energy dissipated, is the kinetic energy, and is energy components is , which should be constant. In the numerical model is only approximately constant, generally with an erro Internal energy The internal energy is the sum of the recoverable elastic strain energy, ; the energy dissipated through inelastic processes such creep, ; and the artificial strain energy, : The artificial strain energy includes energy stored in hourglass resistances and transverse shear in shell and beam elements. Lar or other changes to the mesh are necessary. Viscous energy The viscous energy is the energy dissipated by damping mechanisms, including bulk viscosity damping and material damping. A energy is not part of the energy dissipated through viscoelasticity or inelastic processes. External work of applied forces The external work is integrated forward continuously, defined entirely by nodal forces (moments) and displacements (rotations). work. 3.7.2 Output of the energy balance Each of the energy quantities can be requested as output and can be plotted as time histories summed over the entire model, p within each element. The variable names associated with the energy quantities summed over the entire model or element sets ar Table 3-2 Whole model energy output variables. Variable Name Energy Quantity ALLIE Internal energy, :ALLIE = ALLSE + ALLPD + ALLCD + ALLAE. ALLKE Kinetic energy, . ALLVD Viscous dissipated energy, . ALLFD Frictional dissipated energy, . ALLCD Energy dissipated by viscoelasticity, . ALLWK Work of the external forces, . ALLSE Stored strain energy, . ALLPD Inelastic dissipated energy, . ALLAE Artificial strain energy, . ETOTAL Energy balance: . In addition, ABAQUS/Explicit can produce element-level energy output and energy density output, as listed in Table 3-3 .

Table 3-3 Whole element energy output variables. Variable Name Whole Element Energy Quantity ELSE Elastic strain energy. ELPD Plastic dissipated energy. ELCD Creep dissipated energy. ELVD Viscous dissipated energy. ELASE Artificial energy = drill energy + hourglass energy. EKEDEN Kinetic energy density in the element. ESEDEN Elastic strain energy density in the element. EPDDEN Plastic energy density dissipated in the element. EASEDEN Artificial strain energy density in the element. ECDDEN Creep strain energy density dissipated in the element. EVDDEN Viscous energy density dissipated in the element. 3.8 Potential instabilities with springs and dashpots Some element types fall outside the stability calculations performed by ABAQUS/Explicit. The following elements have the poten pring elements ashpot elements The following elements cannot destabilize the analysis but can only help to stabilize an analysis: ass elements otary inertia elements ydrostatic fluid elements lements that are part of a rigid body We will use simple models to illustrate instabilities in ABAQUS/Explicit. While the program has been designed to provide efficient situations involving springs and dashpots in which a solution may become unstable. It is useful to be able to recognize such ins The following example uses very simple models to illustrate stability problems that could also occur in larger, more complex mod analyses have become unstable, you can then use the same methods to determine whether your own engineering analyses have difficult to determine ahead of time whether or not the analysis will become unstable because the locations and loadings on the that might have otherwise become unstable could be subjected to a contact constraint that keeps the instability from occurring. Springs, as well as elements such as continuum elements, impose certain stability requirements on an analysis. Each element ha and mass associated with the element. Since a spring element has stiffness but no mass of its own, mass is associated with a s containing mass. If a spring element is connected to continuum elements, for example, it generally is not possible to determine not possible for ABAQUS/Explicit to calculate the stable time increment for a spring element. If the stable time increment require of the rest of the model, the spring defines the controlling stable time increment for the model, and the analysis may become un The model shown in Figure 3-17 is made up of two separate, simple bodies: a single truss element and a spring element with is fixed, while the right-hand side is loaded. Since the truss element has both stiffness and mass, ABAQUS/Explicit determines a element has stiffness but no mass, ABAQUS/Explicit does not calculate a stable time increment for the spring element. Likewise the mass element, which has mass but no stiffness. The details of the input file corresponding to this model are not discussed h with springs and dashpots,'' Section A.3. Figure 3-17 Model consisting of a truss, spring, and mass element. In this simple example, however, we can calculate the stability requirements of the spring-mass system analytically because the element. Since the spring stiffness and associated mass are constant throughout the analysis, the maximum stable time increme increment for the truss, however, changes throughout the analysis, as the length and, thus, the stiffness of the truss changes. A smaller than that required for the spring-mass system. As the truss stretches, its stable time increment increases, and eventually

the controlling factor in the analysis. When the stable time increment for the analysis as determined by the truss becomes greate becomes unstable. The stable time increment versus time is illustrated in Figure 3-18 . Figure 3-18 Stable time increment versus time for the truss element and the spring-mass system. 3.8.1 Determining the stable time increment The spring-mass system is loaded dynamically with a constant force of 1000 N, causing the mass to accelerate and the spring t will oscillate according to the applied force and the mass, and the stiffness of the spring. With a spring stiffness of 1 ?105 N/m is 0.01 m, and the maximum displacement is 0.02 m. We can calculate the stable time increment for the spring-mass system jus increment of an element is related to its frequency, according to the following relation: To calculate the undamped circular natural frequency of the spring-mass system, we use the following equation: Therefore, the stable time increment for the spring is For the truss element ABAQUS/Explicit calculates the stable time increment using the following relations: where is the wave speed of the material and is the characteristic element length, defined as At the start of the analysis , making the stable time increment The truss stretches with a constant velocity of 1.0 m/s; thus, the length of the truss element at any given time is and the stable time increment is At time 0.169 s the stable time increment for the analysis, which conforms to the stable time increment for the truss, begins to e Since ABAQUS/Explicit does not consider the spring-mass system in the stable time increment calculation, the spring-mass sys truss exceeds that for the spring-mass system. 3.8.2 Identifying the instability Fetch the input file stability.inp, and run the analysis. Examine the contents of the status (.sta) file, and plot the displacement hi to determine the source of the instability. Status file The output from the analysis provides us with several indications that the analysis has become unstable. The status file is particu . . . ------------------------------------------------------------------------------STABLE TIME INCREMENT INFORMATION -------------------------------------------------------------------------------

The stable time increment estimate for each element is based on linearization about the initial state.

Initial time increment = 3.13065E-03 Statistics for all elements: Mean = 3.13065E-03 Standard deviation = 0.00000E+00 Most critical elements : Element number Rank Time increment Increment ratio ---------------------------------------------------------3 1 3.130655E-03 1.000000E+00

The single precision ABAQUS/Explicit executable will be used in this analysis.

------------------------------------------------------------------------------SOLUTION PROGRESS ------------------------------------------------------------------------------STEP 1 ORIGIN 0.00000E+00

Total memory used for step 1 is approximately 28.4 kilowords Global time estimation algorithm will be used. Scaling factor : 1.0000 STEP TOTAL CPU STABLE CRITICAL KINETIC INCREMENT TIME TIME TIME INCREMENT ELEMENT ENERGY MONITOR 0 0.000E+00 0.000E+00 00:00:00 3.131E-03 3 2.500E-01 0.000E+00 Results number 0 at increment zero. ODB Field Frame Number 0 of 1 requested intervals at increment zero. 20 1.005E-01 1.005E-01 00:00:00 9.438E-03 3 6.271E+00 1.561E-02 27 2.015E-01 2.015E-01 00:00:00 2.552E-02 3 2.765E+03-6.320E-01 ***WARNING: Large rotation detected for SPRINGA element 1. The analysis may go unstable. This message is printed during the first applicable increment, but will not be printed during subsequent increments for the remainder of the step. ***WARNING: Large rotation is detected in 1 1D element (truss, spring, or dashpot). The analysis may go unstable. 31 34 37 40 3.099E-01 4.025E-01 5.049E-01 6.172E-01 3.099E-01 4.025E-01 5.049E-01 6.172E-01 00:00:00 00:00:00 00:00:00 00:00:00 2.979E-02 3.305E-02 3.632E-02 3.960E-02 3 3 3 3 5.797E+11-1.290E+04 9.504E+18 5.903E+07 6.254E+26-5.333E+11 1.414E+35 8.834E+15

. . . For this analysis we suspected in advance that the spring-mass system would become unstable; therefore, we selected the righ monitor column are, therefore, the displacement of the mass element. The first few increments show that the mass has a displac increment 27 at a time of 0.2015 s the displacement of node 2 has become m, well outside the correct range. We anticipated th time exceeded 0.169 s, and increment 27 is the first increment in the status file beyond a time of 0.169 s. Between increments 2 first warning messages regarding analysis stability appear. In this case the warning indicates that the SPRINGA element has und degree of freedom (which spring elements do not possess) but rather a rigid rotation of the spring that has occured because the effectively turned inside out). Beyond increment 31 the displacement of the mass increases rapidly to a value of 8.834 ?1015 m Displacement history The displacement history of the spring's free end is shown graphically in Figure 3-19 . Notice that the distance increases betwe indicating that the time increment used by ABAQUS/Explicit increases throughout the analysis. It is clear from the unrealistic disp Figure 3-19 Displacement history of the free end of the spring (node 2). Using energy to determine instability In a more complicated analysis you may not know ahead of time which parts of the model are likely to become unstable. You mu solution has become unstable besides monitoring the displacements at a single node. The kinetic energy shown for each increm stability. The advantage of checking the kinetic energy over monitoring a particular degree of freedom is that the kinetic energy over the entire model. An unrealistic growth in the kinetic energy may indicate that the analysis has become unstable but does n starts to grow unrealistically at the same time that the stretch in the spring becomes unrealistic, indicating that the two effects h The energies are useful indications of the solution stability, and the most general way to view the energies is to create history pl energy balance (ETOTAL), the kinetic and internal energies (ALLKE, ALLIE), as well as the external work (ALLWK). Figure 3-20 Plots of total model energies from the selected results file. Up to approximately 0.16 s the energy balance remains nearly constant at zero. A constant energy balance is an indication that balance strongly suggests that the solution is unstable. The energy plots show that when the time reaches 0.16 s, the energy ba 3.8.3 Removing the instability Three methods of removing the instability are discussed in the following sections. Adding mass The preferred method of removing an instability caused by springs is to increase the mass associated with the springs. Usually, complex response of the associated continuum. Springs can model the stiffness accurately, but they do not account for the mas

as the mass associated with the spring increases; correspondingly, the stable time increment of the spring-mass system increa spring, the stable time increment of the spring-mass system can be increased such that it will always be greater than the stable will have the desired effect on the behavior of the structure without controlling the stability of the analysis. We can calculate the mass necessary to add to the right-hand node of the spring element so that the spring-mass system rema shown previously, by the end of the step the stable time increment for the truss has increased to We can then determine the natural frequency of the spring-mass system necessary to have a factor of safety of four over the tru time increment equal to four times the truss's maximum stable time increment, we can solve for the necessary mass at the free e Solving for the mass, , with 0.3043 s gives a mass of 37050 kg, which means adding 37045 kg. For this simple spring-mass sy drastically. However, for a larger and more massive model the additional mass would likely have little effect on the structural res numerical solution. The displacement of the free end of the spring is shown in Figure 3-21 . Figure 3-21 Spring-mass system response with a 37050 kg mass at the right-hand node. In a real analysis springs can be used not only to connect a structure to ground (a fixed boundary condition), as shown previous such a model both ends of the spring are free to move, as shown in Figure 3-22 . Figure 3-22 Spring-mass system with both ends free. Both ends may require additional mass for numerical stability. The natural frequency for this case is where is the mass associated with one node of the spring, and is the mass associated with the other node. If the same mass is assoc simplify to and If we desired a specific stability limit for the spring-mass system, we could again solve the equation for the stable time incremen mass to add to each of the nodes. If we include a factor of safety of four, the equation becomes The effect of damping Dashpot elements are often used in conjunction with spring elements to provide damping at discrete points within a model. Usin effects of damping on stability. As with springs, dashpots affect the stable time increment of the analysis, yet ABAQUS/Explicit d increment. In fact, a dashpot in parallel with a spring always lowers the actual stable time increment of the spring. On the other damping on the stable time increment. With no material damping whereas with material damping where is the ratio of the applied damping to the critical damping. The stable time increment for case is always less than it is for case . Generally, it is not possible to calculate because you do n determine ahead of time at what value of an analysis will become unstable. Table 3-4 summarizes the mass that should be added to spring and spring-dashpot systems to ensure stability up to the desir solved for instead of . Table 3-4 Added mass to ensure stability. Case Added Mass (with 4?factor of safety) i) ii) iii) iv) Controlling the time incrementation If adding mass to the spring nodes is not physically appropriate, you can retain stability by controlling the time incrementation. T The FIXED TIME INCREMENTATION parameter causes ABAQUS/Explicit to persist with the stable time increment that was calcula TIME INCREMENTATION parameter eliminates the instability from the solution because the stable time increment is not permitted additional factor of safety, you can set the SCALE FACTOR parameter to the desired value. 3.9 Summary BAQUS/Explicit uses a central difference rule to integrate the kinematics explicitly through time. he explicit method requires many small time increments. Since there are no simultaneous equations to solve, each increment i he explicit method has great cost savings over the implicit method as the model size increases. he stability limit is the maximum time increment that can be used to advance the kinematic state and still remain accurate. BAQUS/Explicit automatically controls the time increment size throughout the analysis to maintain stability. s the material stiffness increases, the stability limit decreases; as the material density increases, the stability limit increases. or a mesh with a single material, the stability limit is roughly proportional to the smallest element dimension. n some situations an ABAQUS/Explicit analysis may become unstable. The example problems in this chapter describe how to 4. Finite Elements and Rigid Bodies

Finite elements and rigid bodies are the fundamental components of an ABAQUS/Explicit model. Finite elements are deformable shape. While users of "finite element" analysis programs tend to have some understanding of what finite elements are, the gene somewhat new. For computational efficiency ABAQUS/Explicit has a general rigid body capability. Any body or part of a body composed of any rigid bodies over deformable bodies is that the motion of a rigid body is described completely by no more than six degrees of fr have many degrees of freedom and require expensive element calculations to determine the deformations. When such deformat rigid body produces significant computational savings without affecting the overall results. 4.1 Finite elements The wide range of elements that are available in ABAQUS may appear intimidating at first. However, the extensive element library problems. This section introduces the five aspects of an element that influence how it behaves. 4.1.1 Characterizing elements Each element is characterized by the following: amily egrees of freedom (directly related to the element family) umber of nodes ormulation ntegration Each element in ABAQUS has a unique name, such as T2D2, S4R, or C3D8R. The element name, as you saw in the overhead ho of the TYPE parameter on the *ELEMENT option in the input file. The element name identifies each of the five aspects of an elem Family Figure 4-1 shows the element families most commonly used in a stress analysis. One of the major distinctions between differe assumes. Figure 4-1 Commonly used element families. The element families that you will use in this guide--continuum, shell, beam, truss, rigid, and special-purpose elements--are d as infinite, membrane, and hydrostatic fluid, are not covered in this guide; if you are interested in using them in your models, rea The first letter or letters of an element's name indicate to which family the element belongs. For example, S4R is a shell element element. Degrees of freedom The degrees of freedom are the fundamental variables calculated during the analysis. For a stress/displacement simulation the d elements, the rotations at each node as well. Coupled temperature-displacement elements also have temperature degrees of fre The following numbering convention is used for the degrees of freedom in two- and three-dimensional models: 1 Translation in direction 1 2 Translation in direction 2 3 Translation in direction 3 4 Rotation about the 1-axis 5 Rotation about the 2-axis 6 Rotation about the 3-axis 11 Temperature Directions 1, 2, and 3 correspond to the global 1-, 2-, and 3-directions, respectively, unless a local coordinate system has bee When using axisymmetric elements, the displacement and rotation degrees of freedom are as follows: 1 Translation in the -direction 2 Translation in the -direction 6 Rotation in the - plane Directions (radial) and (axial) correspond to the global 1- and 2-directions, respectively, unless a local coordinate system has Order of interpolation

Displacements and rotations are calculated only at the nodes of the element. At any other point in the element, the displacemen Elements that have nodes only at their corners, such as the 8-node brick shown in Figure 4-2 , use linear interpolation in each elements. Elements with midside nodes are often called quadratic or second-order elements. In ABAQUS/Explicit all elements a which use a modified second-order interpolation. Figure 4-2 Hexahedral or brick element. Typically, the number of nodes in an element is clearly identified in the element name. The 8-node brick element, as you have s called S4R. The beam element family uses a different convention, wherein the order of interpolation is identified in the name. Th Formulation An element's formulation refers to the mathematical theory used to define the element's behavior. In the absence of adaptive me based on the Lagrangian or material description of behavior: the element deforms with the material. In the alternative Eulerian, o flows through them. Adaptive meshing in ABAQUS/Explicit combines the features of pure Lagrangian and Eulerian analyses and material; it is not discussed in this guide. Integration ABAQUS uses numerical techniques to integrate various quantities over the volume of each element. For most elements ABAQUS each integration point in each element. ABAQUS uses the letter "R" at the end of the element name to indicate reduced-integration elements. For example, CAX4R is th element. 4.1.2 Continuum elements Between the different element families, continuum or solid elements can be used to model the widest variety of components. Co material in a component. Since they can be connected to other elements on any of their faces, continuum elements, like bricks nearly any shape, subjected to nearly any loading. ABAQUS/Explicit has both stress/displacement and coupled temperature-dis stress/displacement elements in this guide. Continuum elements in ABAQUS have names that begin with the letter "C." The next two letters indicate the dimensionality and t indicate a three-dimensional element; "AX," an axisymmetric element; "PE," a plane strain element; and "PS," a plane stress elem Two-dimensional continuum element library ABAQUS has several classes of two-dimensional continuum elements that differ from each other in their out-of-plane behavior. and CAX4R) or triangular (CPE3, CPE6M, CPS3, CPS6M, CAX3, and CAX6M). Figure 4-3 shows the three classes that are used Figure 4-3 Plane strain, plane stress, and axisymmetric elements. Plane strain elements assume that the out-of-plane strain, , is zero; they can be used to model thick structures. Plane stress elements assume that the out-of-plane stress, , is zero; they are suitable for modeling thin structures. Axisymmetric elements, the "CAX" class of elements (CAX3, CAX4R, and CAX6M), model a 360?ring; they are suitable for analyz axisymmetric loading. Two-dimensional solid elements must be defined in the 1-2 plane so that the node ordering is counterclockwise around the elem Figure 4-4 Correct nodal connectivity for two-dimensional elements. When using a preprocessor to generate the mesh, ensure that the element normals all point in the same direction as the positive connectivity will cause ABAQUS to issue an error message stating that elements have negative volume. Three-dimensional continuum element library Three-dimensional continuum elements can be hexahedra (C3D8R), wedges (C3D6), or tetrahedra (C3D4 and C3D10M). Whenever possible hexahedral elements (C3D8R) or second-order tetrahedral elements (C3D10M) should be used. First-order t very fine meshes are required for an accurate solution. Degrees of freedom All of the continuum elements have translational degrees of freedom at each node. Correspondingly, degrees of freedom 1, 2, a of freedom 1 and 2 are active in plane strain elements, plane stress elements, and axisymmetric elements. Element properties The *SOLID SECTION option defines the material and any additional geometric data associated with a set of continuum element geometric information is required: the nodal coordinates completely define the element geometry. For plane stress and plane str the data line. For example, if the elements are 0.2 m thick, the element property definition would be the following: *SOLID SECTION, ELSET=<element set name>, MATERIAL=<material name> 0.2, Integration Each first-order continuum element in ABAQUS/Explicit has a single integration point. For first-order triangular, tetrahedral, and full integration, while for quadrilateral and hexahedral (brick) elements having one integration point is considered to be reduced i either three or four integration points, depending on whether they are two- or three-dimensional; this is considered to be full int Element output variables By default, element output variables such as stress and strain reference the global Cartesian coordinate system. Thus, the com

acts in the global 1-direction. Even if the element rotates during a simulation, as shown in Figure 4-5 (b), the default is still to u element variables. Figure 4-5 Default material directions for continuum elements. However, ABAQUS allows you to define a local coordinate system for element variables, using the *ORIENTATION option. This lo simulations. A local coordinate system can be very useful if the object being modeled has some natural material orientation, suc 4.1.3 Shell elements Shell elements are used to model structures in which one dimension (the thickness) is significantly smaller than the other dimens Shell element names in ABAQUS begin with the letter "S." The axisymmetric shell begins with the letters "SAX." The first number element, except for the case of axisymmetric shells, for which the first number indicates the order of interpolation. All shell elem arbitrarily large rotations with the following exceptions: if the element name ends with the letter "S," the element uses a small-str name ends with the letters "SW," the element uses a small-strain formulation but considers warping. Shell element library Triangular and quadrilateral elements are available with linear interpolation and your choice of large-strain and small-strain form For most analyses the standard large-strain shell elements (S4R, S3R, and SAX1) are appropriate. If, however, the analysis invo small-strain shell elements (S4RS, S3RS, and S4RSW) are more computationally efficient. Degrees of freedom The three-dimensional shell elements (S3R, S4R, S3RS, S4RS, and S4RSW) have six degrees of freedom at each node (three tra The axisymmetric shell (SAX1) has three degrees of freedom at each node: 1 Translation in the -direction 2 Translation in the -direction 6 Rotation in the - plane Element properties Use either the *SHELL GENERAL SECTION or the *SHELL SECTION option to define the thickness and material properties for a s *SHELL SECTION, ELSET=<element set name>, MATERIAL=<material name> <thickness>, <number of section points> or *SHELL GENERAL SECTION, ELSET=<element set name>, MATERIAL=<material name> <thickness>, If you specify the *SHELL SECTION option, ABAQUS uses numerical integration to calculate the behavior at selected points (call Figure 4-6 . The MATERIAL parameter references a material property definition, which may be linear or nonlinear. You can spec Figure 4-6 Section points through the thickness of a shell element. The *SHELL GENERAL SECTION option allows you to define the cross-section behavior in a number of general ways to model li cross-section behavior directly in terms of section engineering quantities (area, moments of inertia, etc.) with this option, there element cross-section. Therefore, *SHELL GENERAL SECTION is less expensive computationally than *SHELL SECTION. The res stresses and strains are calculated only when they are requested for output. Reference surface offsets The reference surface of the shell is defined by the shell element's nodes and normal definitions. When modeling with shell elem midsurface. However, many situations arise in which it is more convenient to define the reference surface as offset from the she usually represent either the top or bottom surface of the shell body. In this case it may be easier to define the reference surface the shell's midsurface. Shell offsets can also be used to define a more precise surface geometry for contact problems where shell thickness is importan contact constraints in ABAQUS/Explicit. The effect of offset and thickness in contact can be suppressed, if required. Shell offsets can also be useful when modeling a shell with continuously varying thickness. In this case defining the nodes at the the other is rough, as in some aircraft structures, it is easiest to use shell offsets to define the nodes at the smooth surface. Offsets can be introduced by using the OFFSET parameter on the *SHELL SECTION and *SHELL GENERAL SECTION options. Th measured from the shell's midsurface to the shell's reference surface. The degrees of freedom for the shell are associated with the reference surface. All kinematic quantities, including the element's may lead to a surface integration error, affecting the stiffness, mass, and rotary inertia for the shell section. For stability purpose used for shell elements on the order of the offset squared, which may result in errors in the dynamics for large offsets. When lar point constraints or rigid body constraints instead. Element output variables

The element output variables for shells are defined in terms of local material directions that lie on the surface of each shell elem 4.1.4 Beam elements Beam elements are used to model components in which one dimension (the length) is significantly greater than the other two di the beam is significant. Beam element names in ABAQUS begin with the letter "B." The next character indicates the dimensionality of the element: "2" fo The third character indicates the interpolation used: "1" for linear interpolation. Linear beams (B21 and B31) are available in two Degrees of freedom Three-dimensional beams have six degrees of freedom at each node: three translational degrees of freedom (1-3) and three ro Two-dimensional beams have three degrees of freedom at each node: two translational degrees of freedom (1 and 2) and one r the model. Element properties Use the *BEAM SECTION option to define the geometry of the beam cross-section; the nodal coordinates define only the length MATERIAL parameter refers to a material property definition. ABAQUS calculates the cross-section behavior of the beam by num nonlinear material behavior. Formulation and integration The beams are shear deformable and account for finite axial strains; therefore, they are suitable for modeling both slender and n Element output variables The stress components in three-dimensional beam elements are the axial stress () and the shear stress due to torsion (). The sh Corresponding strain measures are also available, as well as estimates of transverse shear forces on the section. The two-dimensional beam uses only axial stress and strain. The axial force, bending moments, and curvatures about the local beam axes can also be requested for output. For details of w ABAQUS/Explicit User's Manual. 4.1.5 Truss elements Truss elements are rods that can carry only tensile or compressive loads. They have no resistance to bending; therefore, they ar elements can be used as an approximation for cables or strings (for example, in a tennis racket). Trusses are also sometimes us The truss element names begin with the letter "T." The next two characters indicate the dimensionality of the element--"2D" for The final character represents the number of nodes in the element. Linear truss elements (T2D2 and T3D2) are available in two a Degrees of freedom Truss elements have only translational degrees of freedom at each node. Three-dimensional truss elements have degrees of fre degrees of freedom 1 and 2. Element properties The *SOLID SECTION option is used to specify the name of the material property definition associated with the given set of truss *SOLID SECTION, ELSET=<element set>, MATERIAL=<material> <cross-sectional area> Element output variables Axial stress and strain are available as output. 4.1.6 Mass and rotary inertia elements These elements (MASS and ROTARYI) define mass and rotary inertia at a discrete point. Degrees of freedom Mass elements have three translational degrees of freedom at each node. Rotary inertia elements have three rotational degrees o Element properties Use the *MASS option to define the element property for a mass element: *MASS, ELSET=<element set name> <mass magnitude>, Use the *ROTARY INERTIA option to define the element properties for a rotary inertia element: *ROTARY INERTIA, ELSET=<element set name> <I11>, <I22>, <I33>,<I12>, <I13>, <I23> Element output variables No output is available for these elements. 4.1.7 Spring and dashpot elements Spring and dashpot elements (SPRINGA and DASHPOTA) are used to model the effective stiffness or damping between two nod remains between the two nodes throughout the deformation. Degrees of freedom Spring and dashpot elements have three translational degrees of freedom at each node. Element properties

Use the *SPRING option to define the linear or nonlinear element properties for a spring element. To be consistent with SPRINGA line following *SPRING and *DASHPOT is blank. If defining a linear spring, use the following: *SPRING, ELSET=<element set name> <spring stiffness>, If defining a nonlinear spring, use the following, repeating the second data line as often as necessary: *SPRING, ELSET=<element set name>, NONLINEAR <force>, <relative displacement> Use the *DASHPOT option to define the linear or nonlinear element properties for a dashpot element. If defining a linear dashpo *DASHPOT, ELSET=<element set name> <dashpot coefficient>, If defining a nonlinear dashpot, use the following, repeating the second data line as often as necessary: *DASHPOT, ELSET=<element set name>, NONLINEAR <force>, <relative velocity> Element output variables For spring elements S11 is the force in the spring, and E11 is the relative displacement across the spring. For dashpot elements displacement across the dashpot. 4.2 Rigid bodies In ABAQUS/Explicit a rigid body is a collection of nodes and elements whose motion is governed by the motion of a single node 7 . The shape of the rigid body is defined either as an analytical rigid surface obtained by revolving or extruding a two-dimensio meshing the body with nodes and elements. The shape of the rigid body does not change during a simulation but can undergo l body can be calculated based on the contributions from its elements, or they can be assigned specifically. Figure 4-7 Elements forming a rigid body. The motion of a rigid body can be prescribed by applying boundary conditions at the rigid body reference node. Loads on a rigi and distributed loads applied to elements that are part of the rigid body or from loads applied to the rigid body reference node. connections to deformable elements and through contact with deformable elements. 4.2.1 Determining when to use a rigid body In dynamic analyses rigid bodies can be used to model very stiff components that are either fixed or undergoing large dynamic m deformable components, and they provide a convenient method of specifying certain contact interactions. When ABAQUS/Explic ideally suited for modeling tooling (such as punch, die, drawbead, blank holder, roller, etc.) and may also be effective as a meth It may be useful to make parts of a model rigid for verification purposes. For example, in complex models where all potential con the impact region could be included as part of a rigid body, resulting in faster run times while developing a model. When the use body definitions can be removed and an accurate deformable finite element representation can be incorporated throughout. The principal advantage to representing portions of a model with rigid bodies rather than deformable finite elements is computat elements that are part of a discrete rigid body. Although some computational effort is required to update the motion of the node distributed loads, the motion of the rigid body is determined completely by a maximum of six degrees of freedom at the rigid bo Rigid bodies are particularly effective for modeling relatively stiff parts of a model for which tracking waves and stress distributio the stiff region can result in a very small global time increment. Since rigid bodies and elements that are part of a rigid body do a deformable finite element representation in a stiff region can result in a much larger global time increment, without significantly Rigid bodies defined with analytical rigid surfaces are slightly cheaper in terms of computational cost than discrete rigid bodies. contact with discrete rigid bodies because analytical rigid surfaces can be smooth, whereas discrete rigid bodies are inherently rigid surfaces are limited. 4.2.2 Creating a rigid body To create a discrete rigid body, use the *RIGID BODY option as the property reference for the elements forming the rigid body. U node to the rigid body. A rigid body reference node has both translational and rotational degrees of freedom and must be define *RIGID BODY, REF NODE=<node>, ELSET=<element set name>, PIN NSET=<node set name>, TIE NSET=<node set name> <thickness>, In addition to the rigid body reference node, discrete rigid bodies consist of a collection of nodes that are generated by assignin connection to other elements. Nodes that are part of a rigid body are one of two types: in nodes, which have only translational degrees of freedom. ie nodes, which have both translational and rotational degrees of freedom. The rigid body node type is determined by the type of elements on the rigid body to which the node is attached. The node type to a rigid body. For pin nodes only the translational degrees of freedom are part of the rigid body, and the motion of these degre reference node. For tie nodes both the translational and rotational degrees of freedom are part of the rigid body and are constra 4.2.3 Rigid elements

The rigid body capability in ABAQUS/Explicit allows any elements--not just rigid elements--to be part of a rigid body. For exam same effect if a *RIGID BODY option refers to the element set that contains the elements forming the rigid body. The rules gove are applied, pertain to all element types that form the rigid body, including rigid elements. The names of all rigid elements begin with the letter "R." The next characters indicate the dimensionality of the element. For exa element is axisymmetric. The final character represents the number of nodes in the element. Rigid element library The three-dimensional quadrilateral (R3D4) and triangular (R3D3) rigid elements can be used to model the two-dimensional surf Two-node rigid elements are available for plane strain and plane stress (R2D2) as well as axisymmetric models (RAX2). Physical properties All rigid elements must reference a *RIGID BODY option. For the planar and beam elements the cross-sectional area can be def elements the thickness can be defined on the data line. The default thickness is zero. Alternatively, the NODAL THICKNESS para at the nodes. These data are required when applying body forces or when the thickness is needed for the contact definition. 4.3 Example: hourglassing in a rubber block Under certain loading conditions linear reduced-integration elements can experience a pattern of nonphysical deformation called element modeling a small piece of material subjected to pure bending, as shown in Figure 4-8 . Dotted visualization lines are s Figure 4-8 Deformation of a first-order element with reduced integration subjected to a bending moment. As the element deforms, neither of the dotted visualization lines changes in length, and the angle between them also does not c element's single integration point are zero. This deformation mode is, therefore, a zero-energy mode because no strain energy i is unable to resist this type of deformation since it has no stiffness in this mode. In coarse meshes this zero-energy mode can p ABAQUS/Explicit has first-order, reduced-integration quadrilateral and hexahedral elements that have hourglass modes. Hourgla ABAQUS/Explicit includes sophisticated controls to prevent hourglassing from being a problem in most real analyses. However, take a few increments to control hourglassing. In some severe cases hourglassing can propagate through the mesh before the h illustrates such a modeling situation. In this example you will consider a thick rubber block as it is slowly compressed along its diagonal by a rigid surface, as shown hourglassing is a problem and, if it is a problem, how to correct the model to prevent the problem from occurring. Figure 4-9 Rubber block being compressed diagonally by a rigid surface. 4.3.1 Node and element sets Use your preprocessor to create a two-dimensional model of the rubber block with a 10 ?10 mesh of two-dimensional plane str ``Hourglassing in a rubber block,'' Section A.4. The node sets shown in Figure 4-10 are necessary to apply the loads and boundary conditions and to visualize output. Put the element in an element set called ERIGID. You could define the rigid body using either deformable elements, discrete rigid eleme we assume that you are using a single discrete rigid element. The reference node, which you should place in node set NREF, ca close to its rigid body. Figure 4-10 Node sets NBOT, NRHS, and NREF. 4.3.2 10 x 10 uniform mesh: reviewing the input file--the model data We now review the model data associated with this problem, including the model description, block definition, rigid body definiti and surface definitions. Model description Since you will study several different meshes in this example, choose a model heading that describes the analysis as well as the *HEADING Hourglassing example 10 X 10 regular mesh SI Units (kg, m, s, N) Defining the block Ensure that the element type is CPE4R. Use the following section properties to give the block a thickness of 0.10 m and refer to *SOLID SECTION, ELSET=EALL, MATERIAL=RUBBER .10, Defining the rigid body Use a single rigid element (element type R2D2) to define the rigid body that compresses the block at a 45?angle. Make the surfa The rigid element nodes must be separate from the nodes defining the block. Define the rigid element so that its positive norma elements the positive normal direction is defined by a 90?counterclockwise rotation from the direction going from the first node node that will control the motion of the rigid body. Use the following section properties to define the rigid body: *RIGID BODY, ELSET=<element set name>, REF NODE=<node number> Material properties The block in this example is made of rubber and, thus, should be modeled using the hyperelastic material model. The hyperelast

rubber. Refer to Chapter 5, "Materials," for more information about defining hyperelastic materials. *MATERIAL, NAME=RUBBER ** hyperelastic constants in Pa *HYPERELASTIC, POLYNOMIAL, N=1 3.2E6, .8E6 ** density is 1500 kg/m^3 *DENSITY 1500., Fixed boundary conditions Fixed boundary conditions can be defined in either the model or the history part of the input file. Define roller boundary conditio body's rotational degree of freedom as well. *BOUNDARY NBOT, 2 NRHS, 1 NREF, 6 Amplitude definition To gradually compress the rubber block by prescribing a nodal displacement at the rigid body reference node, you must first de displacements in the history definition will then refer to this loading amplitude. Define a smooth linear loading amplitude that has 1.0 at the end of the step (time 0.1 s), such as the following: *AMPLITUDE, NAME=CRUSH, DEFINITION=SMOOTH STEP 0., 0., .1, 1. Since the parameter DEFINITION was set equal to SMOOTH STEP in the above option, the first and second derivatives of the am interval. This promotes a smooth transition from one amplitude level to another. This parameter is commonly used to promote qu amplitude curves,'' Section 7.2.1. Surface definitions Define surfaces on the side of the rigid element that will contact the block and on the exterior of the block. Since the normal of occur on the rigid element's positive (SPOS) face. *SURFACE, NAME=RIGID ERIGID, SPOS *SURFACE, NAME=SOLID EALL, 4.3.3 10 x 10 uniform mesh: reviewing the input file--the history data The history data required for this simulation are discussed next, including the step definition, contact definitions, loading, and ou Step definition The following would be an appropriate heading for this step: *STEP Compress the rubber block The goal is to compress the block slowly so that the dynamic effects are not dominant. Since this model is quite small, compute amount of time to compress the block in a nearly static manner. (Quasi-static analyses will be discussed in detail in Chapter 7, for the step. *DYNAMIC, EXPLICIT , 0.1 Contact definitions Contact is discussed in detail in Chapter 6, "Contact." Define the contact between the rigid body and the block using the followi *CONTACT PAIR, INTERACTION=INTER SOLID, RIGID *SURFACE INTERACTION, NAME=INTER Loading The loading for this analysis is a displacement boundary condition imposed on the rigid body reference node. As the rigid body distance of 0.017 m at a 45?angle, impose equal displacement boundary conditions of 0.012 m in both the 1-direction and the *BOUNDARY, TYPE=DISPLACEMENT, AMPLITUDE=CRUSH NREF, 1, 1, .012 NREF, 2, 2, -.012 Output requests You want to have an output database file created during the analysis so you can use ABAQUS/Viewer to postprocess the results

history data to the output database file. Add the following output requests to your input file: *OUTPUT, FIELD, NUMBER INTERVAL=20, VARIABLE=PRESELECT *OUTPUT, HISTORY, FREQUENCY=1 *ENERGY OUTPUT ALLAE, ALLKE, ALLSE, ALLVD, ALLFD, ALLWK, ETOTAL Setting the FREQUENCY parameter on the *OUTPUT, HISTORY option equal to 1 causes history data to be written to the output postprocessing. However, if many variables are chosen, the price of such detail can be a large output database file. Indicate the end of a step with the option *END STEP Make sure that this input option is the last option in your model. 4.3.4 Running the analysis After storing your input in a file called hourglass_square.inp, run the analysis using the following command: abaqus job=hourglass_square If the analysis does not complete, check the files hourglass_square.dat and hourglass_square.sta for error messages. Modify yo running the analysis, compare your input file to the one provided in ``Hourglassing in a rubber block,'' Section A.4. 4.3.5 Postprocessing Start ABAQUS/Viewer by typing abaqus viewer odb=hourglass_square at the operating system prompt. Plotting the deformed model shape To begin this exercise, plot the deformed model shape. To plot the deformed model shape: 1.From the main menu bar, select Plot->Deformed Shape; or use the tool in the toolbox. The deformed model plot appears in the current viewport as shown in Figure 4-11 . Figure 4-11 Final deformed shape of mesh 1. Looking closely at the deformation, you will notice that much of the mesh has a pattern of alternating trapezoids, which indicate is most pronounced near the corner that is being pushed in. In this example the hourglassing pattern is severe enough that we c Usually, hourglassing is not a severe problem if it is not readily visible in the deformed mesh. A more quantitative approach is to the energy dissipated to control hourglassing deformation. If the artificial strain energy is excessive, too much strain energy may Studying the artificial strain energy To determine what is an excessive value of artificial strain energy, the most useful approach is to compare the artificial strain en elastic; therefore, a comparison with the elastic strain energy is appropriate. In ABAQUS/Explicit the variable ALLAE is the total energy dissipated as artificial strain energy and the variable ALLSE is the elast and elastic terms; however, since the viscous term is usually predominant, most of the energy that goes into artificial strain ener energy and the elastic strain energy in ABAQUS/Viewer to determine whether the amount of artificial strain energy is excessive in To create history plots of the artificial and elastic strain energies: 1.From the main menu bar, select Result->History Output. 2.Select the variable ALLAE from the list of available Output Variables in the History Output dialog box. 3.Click Plot. ABAQUS/Viewer plots the artificial strain energy history as shown in Figure 4-12 . Figure 4-12 History plot of artificial strain energy (ALLAE) in mesh 1. 4.Select the variable ALLSE from the list of available Output Variables in the History Output dialog box. 5.Click Plot. ABAQUS/Viewer plots the elastic strain energy history as shown in Figure 4-13 . Figure 4-13 History plot of elastic strain energy (ALLSE) in mesh 1. 6.Click Dismiss to close the History Output dialog box. ABAQUS/Viewer allows you to create new X-Y data objects by performing operations on previously saved X-Y data objects. You of the artificial strain energy to the elastic strain energy versus time. Then, you will plot the ratio of artificial strain energy to elast To plot the ratio of artificial strain energy to elastic strain energy versus time: 1.From the main menu bar, select Tools->XY Data->Manager. The XY Data Manager dialog box appears. 2.From this dialog box, click Create. 3.From the Create XY Data dialog box that appears, select ODB history output and click Continue. The History Output dialog box appears.

4.From the list of available Output Variables in the History Output dialog box, select ALLAE and click Save As. The Save XYData As dialog box appears. 5.Name the X-Y data object ALLAE, and click OK. 6.Use a similar technique to save a data object containing the elastic strain energy (ALLSE). Name this data object ALLSE. 7.Dismiss the History Output dialog box. You are now ready to operate on the X-Y data objects. 8.In the XY Data Manager dialog box, click Create. 9.From the Create XY Data dialog box that appears, select Operate on XY data and click Continue. The Operate on XY Data dialog box appears. 10.From the XY Data field, select ALLAE. "ALLAE" appears in the text field at the top of the dialog box. 11.From the Operators field, select the division symbol (/). A division symbol appears following "ALLAE" in the text field at the top of the dialog box. 12.From the XY Data field, select ALLSE. "ALLSE" is appended to the expression in the text field at the top of the dialog box. 13.Click Save As. The Save XYData As dialog box appears. 14.Name the data object AESE, and click OK. A dialog box may appear warning that ABAQUS/Viewer detected an attempt to divide by zero. In this case the elastic strain ener ABAQUS/Viewer sets the corresponding value to zero. 15.Dismiss the Operate on XY Data dialog box. You are now ready to plot the ratio of artificial strain energy to elastic strain energy. 16.From the XY Data Manager dialog box, select AESE and click Plot. ABAQUS/Viewer plots the ratio of artificial strain energy to elastic strain energy versus time. By default, there is no title on the ve 17.From the main menu bar, select Options->XY Plot. The XY Plot Options dialog box appears. 18.Click the Titles tab. From the Y-Axis options, change the title source to User-specified; then enter AESE for Whole Model for 19.Click OK. ABAQUS/Viewer creates the plot shown in Figure 4-14 . Figure 4-14 History of the artificial strain energy to elastic strain energy ratio. This plot shows that the ratio of artificial to elastic strain energy reaches a maximum of 15.7% early in the analysis. After this po continues to increase, indicating that the hourglassing problem is not worsening. At the conclusion of the analysis, the ratio drop analysis the ratio of energy dissipated as artificial strain energy to actual strain energy is well over 10%, while a general rule is th 10% we need to think about what is possibly causing the excessive artificial energy and how we can decrease the ratio to impro Understanding what is causing the mesh to hourglass so severely To help understand what is causing the mesh to hourglass so severely, consider the deformed model shape near the time when this occurs. To determine when hourglassing is most severe: 1.From the main menu bar, select Tools->Query. 2.In the Query dialog box that appears, click OK. The Probe Values dialog box appears. 3.Move the cursor over the curve, and locate the peak value on the curve. As you move the mouse, notice that ABAQUS/Viewer displays the X-Y coordinates of the point currently under the cursor in the peak value of the ratio of artificial strain energy to elastic strain energy occurs at 0.044 s. 4.Click Cancel to close the Probe Values dialog box. Plot the deformed model shape in the vicinity of 0.044 s into the simulation to gain further insight into the source of the hourglas To plot the deformed model shape near the time when hourglassing is most severe: 1.From the main menu bar, select Result->Step/Frame. The Step/Frame dialog box appears. 2.In the Frame field of the Step/Frame dialog box, select the increment with a step time nearest to 0.044 s. 3.Click OK. Subsequent deformed model plots will correspond to the increment you just selected. 4.From the main menu bar, select Plot->Deformed Shape; or use the tool in the toolbox. ABAQUS/Viewer displays the deformed shape of the model. Turn on the display of node symbols. 5.From the main menu bar, select Options->Deformed Shape.

The Deformed Shape Plot Options dialog box appears. 6.Click the Labels tab, and toggle on Show node symbols. 7.Click OK. The deformed model plot is updated to reflect the selected deformed plot options. 8.Zoom in on the upper left corner of the mesh. From the toolbar, click the tool and drag the mouse to select a rectangular region that encloses the upper left corner of the mes ABAQUS/Viewer zooms in on the selected region of the mesh as shown in Figure 4-15 . Figure 4-15 Hourglassing in the corner region of mesh 1. 9.Click mouse button 2 to exit the box zoom mode. The displaced mesh shows the classic hourglassing pattern, in which adjacent elements are alternating trapezoids. Several obse contacts the rigid body at a single node, is deformed excessively. Second, the hourglassing pattern is most severe in elements appears to be more severe in the elements with free edges than in interior elements. Since these observations are based only on It appears that the hourglassing is seeded at the corner element, where the contact at a single node is similar to a concentrated the unconstrained boundaries of the block and along the diagonal. As the analysis progresses, the free boundaries become furt the hourglassing decreases. Table 4-1 lists three common causes of hourglassing and remedies for the problem. Table 4-1 Common causes and remedies of excessive hourglassing. Cause Remedy Concentrated force at a single node Distribute the force among several nodes or apply a distributed load. Boundary condition at a single node Distribute the boundary constraint among several nodes. Contact at a single node Distribute the contact constraint among several nodes. In summary, loads, boundary conditions, or contact constraints at a single node tend to promote hourglassing, as shown in Figu over two or more nodes greatly reduces hourglassing problems, as shown in Figure 4-17 . Figure 4-16 Concentrated force or constraint seeding hourglassing deformation. Figure 4-17 Improved deformation pattern with distributed force or constraint. There are two specific changes we can make to the rubber block mesh to reduce hourglassing. 1.Refine the mesh, as mesh refinement always decreases hourglassing; or 2.Round the corner so that contact never occurs at only a single node. Distributing the contact constraint over several nodes rem 4.3.6 Effects of changing the mesh To illustrate the effects of refining the mesh and distributing the contact constraint over several nodes, we have created the thre Figure 4-18 From left to right: a coarse, flat-cornered mesh (mesh 2); a fine, square mesh (mesh 3); a fine, flat-cornered mes Meshes 2 and 4 incorporate the simplest type of rounded corner: a single triangular element replaces the square element at the a single corner node; instead, contact initially occurs at two nodes, as shown in Figure 4-19 , thereby reducing hourglassing. C some of the material at the corner. With mesh refinement, however, the change to the model becomes less significant. Figure 4-19 Initial contact between the rigid body and the fine, flat-cornered mesh (mesh 4). We study the quality of the results from these new meshes in the same manner as before, using X-Y plots of energy and deform energy history using the original mesh and using the three new meshes. Since the block is compressed monotonically, it follows the plot confirms. The maximum strain energy is lower in the models with the flat corner than in the models with the sharp corne corner and compressing less material stores less strain energy. With mesh refinement the differences decrease between the flat Figure 4-20 Elastic strain energy history for all 4 meshes. Figure 4-21 shows the histories of artificial strain energy for the four meshes, for which the trends are clear. Figure 4-21 Artificial strain energy history of all four meshes. Mesh 1 (the coarse, square mesh) has by far the highest artificial strain energy; and mesh 3 (the fine, square mesh) has the nex than that of the coarse mesh. For each of these cases the artificial strain energy initially dissipates rapidly but eventually reaches occurs while there is contact at only a single node; after two nodes come into contact, the highest artificial strain energy increas sooner two or more nodes are in contact, the sooner the hourglassing will diminish. From the beginning the flat-cornered meshe plateau in the artificial strain energy. Even the coarse, flat-cornered mesh (mesh 2) has a lower artificial strain energy than the f between two nodes does more to decrease hourglassing than refining the mesh. Still lower in artificial strain energy is the fine, f include mesh refinement and distributed contact forces. Figure 4-22 shows how the ratio of the dissipated artificial strain energy to elastic strain energy changes as the analysis progre Figure 4-22 History of artificial strain energy to elastic strain energy for all four meshes. The trends are the same as for the artificial strain energy plots alone. The advantage of energy ratios is that they allow us to use

hourglassing is a problem. For both the coarse and fine square meshes the ratio, which is between 10% and 15% throughout the becomes highly compressed does the ratio drop below 5%. For the flat-cornered meshes the ratio remains below 2% throughou For this example we showed the most basic way to distribute the corner contact forces by flattening the corner of the rubber blo hourglassing, a more general approach would be to round the corner with a smooth fillet instead of replacing the corner elemen filleted mesh are shown in Figure 4-23 . Though their behavior is slightly more complicated than that of their flat-cornered coun same manner. Once more than one node is in contact at the corner, the hourglassing problem diminishes. The rounded meshes because they allow a better representation of the physical problem. Figure 4-23 Left: coarse, filleted mesh (mesh 5); right: fine, filleted mesh (mesh 6). 4.4 Summary BAQUS has an extensive library of elements that can be used for a wide range of structural applications. Element type has imp your simulation. he degrees of freedom active at a node depend on the element types attached to the node. he element name completely identifies the element's family, formulation, number of nodes, and type of integration. ll elements must refer to an element property option. The element property option provides any additional data required to def material property definition. or continuum elements ABAQUS defines the element output variables, such as stress and strain, with respect to the global Ca system by using the *ORIENTATION option. or three-dimensional shell elements ABAQUS defines the element output variables with respect to a coordinate system based system by using the *ORIENTATION option. or computational efficiency any part of a model can be defined as a rigid body, which has degrees of freedom only at its refer s a method of constraint, rigid bodies are computationally more efficient than multi-point constraints. ourglassing can be caused by concentrated forces, boundary conditions, or contact acting at a single node. Refining the mes nodes often can prevent the problem. 5. Materials The material library in ABAQUS/Explicit allows most engineering materials to be modeled, including metals, plastics, rubbers, foa concrete. This guide discusses four of the most commonly used material models: linear elasticity, metal plasticity, rubber elastic in detail in the ABAQUS/Explicit User's Manual. 5.1 Defining materials in ABAQUS You can define any number of materials within an ABAQUS model. Each material definition begins with a *MATERIAL option. The assign the material definition to specific elements in the model. The material definition is one of the few situations in which the position of option blocks in the ABAQUS input file is important. A behavior, such as its elastic modulus or density, must follow the *MATERIAL option directly. Furthermore, the material option blo interrupted by other nonmaterial options. ABAQUS issues an error message if it cannot associate a material behavior option bloc For example, consider the material description of an elastic-plastic metal, which requires several material behavior options to su property option blocks, ABAQUS/Explicit always requires a density. Thus, the complete material description would be: A non-material option block between the *PLASTIC and *DENSITY options, as shown below, would cause ABAQUS to terminate 5.2 Plasticity in ductile metals Many metals exhibit approximately linear elastic behavior at low strain levels, as shown in Figure 5-1 . Figure 5-1 Stress-strain behavior for a linear elastic material, such as steel, at small strains. A constant stiffness, called the Young's modulus or the elastic modulus, characterizes the material behavior. At higher strain lev 5-2 . Figure 5-2 Nominal stress-strain behavior of an elastic-plastic material in a tensile test. 5.2.1 Characteristics of plasticity in ductile metals The plastic behavior of a material is described by its yield point and its post-yield hardening. Figure 5-2 shows a stress-strain a point on the stress-strain curve known as the elastic limit or the yield point, the behavior changes from elastic to plastic. In mo 0.05 to 0.1% of the material's elastic modulus. The deformation of the metal prior to reaching the yield point creates only elastic strains, which are recovered fully if the applied the yield stress, permanent or plastic deformation begins to occur. Strains associated with plastic deformation are called plastic increase according to the original elastic modulus, so that any additional straining contains both elastic and plastic components Once the metal yields, the stiffness for continued loading decreases dramatically, while the Young's modulus still defines the sti unloading, its stiffness is equal to the Young's modulus until its stress-strain curve on reloading once again intersects the harde loading along the hardening curve. Often plastic deformation increases a material's yield stress upon subsequent loadings, a be A metal deforming plastically under a tensile load may experience highly localized deformation, called necking, after reaching its unit original area) drops well below the ultimate strength, as the nominal strain (length change per unit original length) continues the test specimen, the nature of the test itself, and the stress and strain measures used. For example, testing the same material

have a necking region because the specimen does not thin as it deforms in compression. A mathematical model describing the in the compressive and tensile behavior independent of the structure's geometry or the nature of the applied loads. Replacing no strain measures allows us to account for the change in area during the finite deformations. 5.2.2 Stress and strain measures for finite deformations Strains in compression and tension are the same only if considered in the limit as ; i.e., and where is the current length, is the original length, and is the true strain or logarithmic strain. The stress measure that is the conjugate to the true strain is called the true stress and is defined as where is the force in the material and is the current area. A ductile metal subjected to finite deformations will have identical stre plotted against true strain. 5.2.3 Defining plasticity in ABAQUS When defining plasticity data in ABAQUS, you must use true stress and true strain. ABAQUS requires these values to interpret the data are supplied using values of nominal stress and strain. In such situations you must convert the plastic material data from no The relationship between true strain and nominal strain is established by expressing the nominal strain as Adding unity to both sides of this expression and taking the natural log of both sides provides the relationship between the true The relationship between true stress and nominal stress is formed by considering the incompressible nature of the plastic deform negligible, so The expression relating the current area to the original area is Substituting this definition of into the definition of true stress gives where can also be written as Making this final substitution provides the relationship between true stress and nominal stress and strain: Use the *PLASTIC option in ABAQUS to define the post-yield behavior for most metals. The data pairs on the *PLASTIC option d data pair defines the initial yield stress and the corresponding initial plastic strain, which must have a value of zero. ABAQUS co segments to form a continuous, piecewise-linear plasticity curve. You can use any number of data pairs to approximate the actu approximation to the actual material behavior. The strains provided in material test data used to define the plastic behavior are not likely to be the plastic strains in the material You must decompose these total strain values into the elastic and plastic strain components. The plastic strain is obtained by su divided by the Young's modulus, from the value of total strain (see Figure 5-3 ). This relationship is written where is true plastic strain, is true total strain, is true elastic strain, is true stress, and is Young's modulus. Figure 5-3 Decomposition of the total strain into elastic and plastic components. Example of converting material test data to ABAQUS input The stress-strain curve in Figure 5-4 will be used as an example of how to convert the test data defining a material's plastic b points shown on the nominal stress-strain curve will be used as the data for the *PLASTIC option. Figure 5-4 Elastic-plastic material behavior and corresponding ABAQUS input data. The first step is to use the equations relating the true stress to the nominal stress and strain and the true strain to the nominal st and true strain. Once these values are known, the equation relating the plastic strain to the total and elastic strains (shown earlie each yield stress value. The converted data are shown in Table 5-1 . Table 5-1 Converting nominal stress and strain to true stress and strain. Nominal Stress Nominal Strain True Stress True Strain Plastic Strain 200E6 0.00095 200.2E6 0.00095 0.0 240E6

0.025 246E6 0.0247 0.0235 280E6 0.050 294E6 0.0488 0.0474 340E6 0.100 374E6 0.0953 0.0935 380E6 0.150 437E6 0.1398 0.1377 400E6 0.200 480E6 0.1823 0.1800 While there are only slight differences between the nominal and true values at small strains, there are very significant differences stress-strain data to ABAQUS if the strains in the simulation will be large. The format of the input data defining this material beh Regularization of user-defined data When performing an analysis, ABAQUS/Explicit may not use exactly the material data defined by the user; for efficiency, all mate regularized. Material data can be functions of temperature, external fields, and internal state variables, such as plastic strain. Fo determined by interpolation, and, for efficiency, ABAQUS/Explicit fits the user-defined curves with curves composed of equally data used during the analysis. It is important to understand the differences that might exist between the regularized material curv input file. To illustrate the implications of using regularized material data, consider the following two cases. Figure 5-5 shows a case in w example ABAQUS/Explicit generates the six regular data points shown, and the user's data are reproduced exactly. Figure 5-6 regularize exactly. In this example it is assumed that ABAQUS/Explicit has regularized the data by dividing the range into 10 inte Figure 5-5 Example of user data that can be regularized exactly. Figure 5-6 Example of user data that are difficult to regularize. ABAQUS/Explicit attempts to use enough intervals such that the maximum error between the regularized data and the user-defin using the RTOL parameter on the *MATERIAL option. If more than 200 intervals are required to obtain an acceptable regularized message. In general, the regularization is more difficult if the smallest interval defined by the user is small compared to the rang strain of 1.0 makes the range of strain values large compared to the small intervals defined at low strain levels. Removing this la Interpolation between data points ABAQUS/Explicit interpolates linearly between the regularized data points to obtain the material's response and assumes that the data. Thus, the stress in the material shown in Figure 5-5 and Figure 5-6 will never exceed 300 MPa; when the stress in the m assumed to be zero. 5.3 Example: blast loading on a stiffened plate In this example you will assess the response of a stiffened square plate subjected to a blast loading. The plate is firmly clamped to it. The plate is constructed of 25 mm thick steel and is 2 m square. The stiffeners are made from 12.5 mm thick plate and hav material properties in more detail. Figure 5-7 Problem description for blast load on a flat plate. The purpose of this example is to determine the response of the plate and to see how it changes as the sophistication of the ma standard elastic-plastic material model. Subsequently, we study the effects of including material damping and rate-dependent m 5.3.1 Coordinate system Use the default rectangular coordinate system with the plate lying in the 1-3 plane. Since the plate thickness is significantly sma of type S4R for the model.

5.3.2 Mesh design Create your mesh based on the design shown in Figure 5-8 , which is a relatively coarse mesh of 20 ?20 elements in the plate corresponds to the input file shown in ``Blast loading on a stiffened plate,'' Section A.5. It provides moderate accuracy while kee element normals for the plate all point in the positive 1-direction. Doing so ensures that the stiffeners lie on the SPOS face of th properties and shell offsets later. Figure 5-8 Mesh design for the stiffened plate. 5.3.3 Node and element sets Figure 5-9 shows all the sets necessary to apply the element properties, loads, and boundary conditions. Figure 5-9 Node and element sets. Include all the nodes on the perimeter of the plate in a node set called EDGE. These nodes will have a completely fixed boundar containing the node at the center of the plate. Put the plate elements in an element set called PLATE, and put the stiffener eleme set for output purposes called STIFFMAX, which contains the four center elements on the central stiffener. These elements will b 5.3.4 Reviewing the input file--the model data We now review the model data for this problem, including the model description, node and element definitions, element properti and amplitude definition for the blast load. Model description The *HEADING option is used to include a title and model description in the input file. The heading is useful for future reference the evolution of complex models. It can be several lines long, but only the first line will be printed as a title on the output pages. analysis. *HEADING Blast load on a flat plate with stiffeners S4R elements (20x20 mesh) Normal stiffeners (20x2) SI units (kg, m, s, N) Nodal coordinates and element connectivity Use your preprocessor to generate the mesh shown in Figure 5-8 and the sets shown in Figure 5-9 . Element properties and shell offset Give each element set the section properties shown below. Include the appropriate MATERIAL parameter on each *SHELL SECT definition. *SHELL SECTION, MATERIAL=STEEL, ELSET=PLATE, OFFSET=SPOS 0.025, *SHELL SECTION, MATERIAL=STEEL, ELSET=STIFF 0.0125, The material named STEEL will be defined in the next section. Setting OFFSET to SPOS offsets the midsurface of the plate one h make the PLATE nodes lie on the SPOS shell face instead of on the shell midsurface. The purpose of the shell offset in this case overlapping any material with the plate. Figure 5-10 shows the cross-section of the joint between the stiffener and panel using Figure 5-10 Stiffener joint in which the plate's midsurface is offset from its nodes. If the stiffener and base plate elements are joined at common nodes at their midsurfaces, an area of material overlaps, as show Figure 5-11 Overlapping material if OFFSET were not used. If the thicknesses of the plate and stiffener is small in comparison to the overall dimensions of the structure, this overlapping ma analysis results. However, if the stiffener is short in comparison to its width or to the thickness of the base plate, the additional s the whole structure. Material properties Assume that both the plate and stiffeners are made of steel (Young's modulus of 210.0 GPa and Poisson's ratio of 0.3). At this deformation, but we know the value of the yield stress and details of the post-yield behavior for this steel. We will add this inform yield stress is 300 MPa, and the yield stress increases to 400 MPa at a plastic strain of 35%. The plasticity data are shown below *MATERIAL, NAME=STEEL *ELASTIC 210.0E9, .3 *PLASTIC 300.0E6, 0.000 350.0E6, 0.025 375.0E6, 0.100 394.0E6, 0.200 400.0E6, 0.350

*DENSITY 7800.0, Figure 5-12 Yield stress versus log plastic strain. During the analysis ABAQUS calculates values of yield stress from the current values of plastic strain. As discussed earlier, the p are regular--when the stress-strain data are at equally spaced values of plastic strain. To avoid having the user input regular da case the data are regularized by ABAQUS/Explicit by expanding to 15 equally spaced points with increments of 0.025. To illustrate the error message that is produced when ABAQUS/Explicit cannot regularize the material data, try setting the regula pair, as shown below: The combination of the low tolerance value (RTOL=0.001) and the small interval in the user-defined data leads to difficulty in reg produced in the status (.sta) file: ***ERROR: Failed to regularize material data. Try regularizing the independent variable intervals. Boundary conditions Fully constrain the edges of the plate using the node set EDGE defined previously. *BOUNDARY EDGE, ENCASTRE Alternatively, you could specify the degrees of freedom by number. *BOUNDARY EDGE, 1, 6 Amplitude definition for blast load Since the plate will be subjected to a load that varies with time, you must define an appropriate amplitude curve to describe the follows: *AMPLITUDE, NAME=BLAST 0.0, 0.0, 1.0E-3, 7.0E5, 10E-3, 7.0E5, 20E-3, 0.0 50E-3, 0.0 Figure 5-13 Pressure load as a function of time. The pressure increases rapidly from zero at the start of the analysis to its maximum of 7.0 ?105 N in 1 ms, at which point it rema ms. It then remains at zero for the remainder of the analysis. 5.3.5 Reviewing the input file--the history data The history data begin with the *STEP option, which is followed immediately by a title for the step. After the title, specify the *DY *STEP Apply blast loading ** Explicit analysis with a time duration of 50 ms *DYNAMIC, EXPLICIT , 50E-03 Applying the blast load Use the *DLOAD option to apply the blast load to the plate. It is important to ensure that the pressure load is being applied in th direction of the positive shell normal. For shell elements the positive normal direction is obtained using the right-hand rule abou magnitude of the load has been defined in the BLAST amplitude definition, you need to apply only a unit pressure under *DLOAD (where the stiffeners are on the bottom of the plate). Such a pressure load will place the outer fibers of the stiffeners in tension. *DLOAD, AMPLITUDE=BLAST PLATE, P, 1.0

Figure 5-14 Definition of positive pressure load. Output requests To check on the progress of the solution, use the *MONITOR option to monitor the deflection at the center node of the plate dur center node by adding the following command to your input file: *MONITOR, NODE=<center node number>, DOF=1 Set the number of intervals during the step at which preselected field data are written to the output database file (ODB) to 25. Th since the total time for the step is 50 ms. In general, you should try to limit the number of frames written during the analysis to k analysis saving information every 2 ms should provide sufficient detail to study the response of the structure visually. *OUTPUT, FIELD, NUMBER INTERVAL=25, VARIABLE=PRESELECT A more detailed set of output can be saved for selected parts of the model by using the *OUTPUT, HISTORY option. Set the TIM data at 500 points during the analysis. Write von Mises stress (MISES), equivalent plastic strain (PEEQ), and volumetric strain rat nodes that will undergo the maximum displacements are at the center of the plate, use node set NOUT to output displacement a

save the following energy variables: kinetic energy (ALLKE), recoverable strain energy (ALLSE), work done (ALLWK), energy lost energy lost in viscous dissipation (ALLVD), artificial energy (ALLAE), and the energy balance (ETOTAL). *OUTPUT, HISTORY, TIME INTERVAL=1.0E-4 *ELEMENT OUTPUT, ELSET=STIFFMAX PEEQ, MISES *NODE OUTPUT, NSET=NOUT U, V *ENERGY OUTPUT ALLKE, ALLSE, ALLWK, ALLPD, ALLIE, ALLVD, ALLAE, ETOTAL *END STEP Save your input in a file called blast_base.inp since these results will serve as a base state from which to compare subsequent abaqus job=blast_base 5.3.6 Output We now examine the output information contained in the status (.sta) file. Status file Information concerning model information, such as total mass and center of mass, and the initial stable time increment can be f (i.e., those resulting in the smallest time increments) in rank order are also shown here. If your model contains a few elements th the small elements will be the most critical elements and will control the stable time increment. The stable time increment inform affecting the stable time increment, allowing you to change the mesh to improve the situation, if necessary. It is ideal to have a mesh is uniform; thus, the 10 most critical elements share the same minimum time increment. The beginning of the status file is ------------------------------------------------------------------------------MODEL INFORMATION (IN GLOBAL X-Y COORDINATES) ------------------------------------------------------------------------------Total mass in model = 838.50 Center of mass of model = ( 3.488360E-03, 9.999977E-01, 9.999990E-01) Moments of Inertia : About Center of Mass I(XX) 5.557794E+02 I(YY) 2.750539E+02 I(ZZ) 2.849024E+02 I(XY) 9.059906E-06 I(YZ) 4.272461E-04 I(ZX) 3.337860E-06

About Origin 2.232779E+03 1.113565E+03 1.123411E+03 2.925000E+00 8.385000E+02 2.924998E+00

------------------------------------------------------------------------------STABLE TIME INCREMENT INFORMATION -------------------------------------------------------------------------------

The stable time increment estimate for each element is based on linearization about the initial state.

Initial time increment = 6.99621E-06 Statistics for all elements: Mean = 1.01237E-05 Standard deviation = 1.71465E-06 Most critical elements : Element number Rank Time increment Increment ratio ---------------------------------------------------------1033 1 6.996214E-06 1.000000E+00

1038 2 6.996214E-06 1.000000E+00 2033 3 6.996214E-06 1.000000E+00 2038 4 6.996214E-06 1.000000E+00 3033 5 6.996214E-06 1.000000E+00 3038 6 6.996214E-06 1.000000E+00 1013 7 6.996215E-06 9.999999E-01 1018 8 6.996215E-06 9.999999E-01 2013 9 6.996215E-06 9.999999E-01 2018 10 6.996215E-06 9.999999E-01 During the analysis the status file can be viewed to monitor the progress of the analysis. Shown below is the beginning of the so increments have been carried out than you would expect from an ABAQUS/Standard analysis and that the output database file is -------------------------------------------------------------------------------SOLUTION PROGRESS -------------------------------------------------------------------------------STEP 1 ORIGIN 0.00000E+00 Total memory used for step 1 is approximately 755.2 kilowords Global time estimation algorithm will be used. Scaling factor : 1.0000

STEP TOTAL CPU STABLE CRITICAL KINETIC INCREMENT TIME TIME TIME INCREMENT ELEMENT ENERGY MONITOR 0 0.000E+00 0.000E+00 00:00:00 6.996E-06 1033 0.000E+00 0.000E+00 Results number 0 at increment zero. ODB Field Frame Number 0 of 25 requested intervals at increment zero. 235 2.000E-03 2.000E-03 00:00:12 8.573E-06 2026 4.468E+03 4.192E-03 ODB Field Frame Number 1 of 25 requested intervals at 2.000078E-03 469 4.006E-03 4.006E-03 00:00:25 8.572E-06 2018 1.119E+04 2.519E-02 ODB Field Frame Number 2 of 25 requested intervals at 4.005884E-03 702 6.003E-03 6.003E-03 00:00:38 8.572E-06 2031 6.167E+03 4.584E-02 ODB Field Frame Number 3 of 25 requested intervals at 6.003135E-03 935 8.000E-03 8.000E-03 00:00:50 8.556E-06 2031 1.766E+02 5.033E-02 ODB Field Frame Number 4 of 25 requested intervals at 8.000187E-03 1170 1.001E-02 1.001E-02 00:01:03 8.545E-06 2030 2.308E+03 4.535E-02 ODB Field Frame Number 5 of 25 requested intervals at 1.000848E-02 Output for the node referenced on the *MONITOR option is also included in this file. 5.3.7 Postprocessing Run ABAQUS/Viewer by entering the command abaqus viewer odb=blast_base at the operating system prompt. Plotting the undeformed model shape You will begin this exercise by plotting the undeformed model shape of the plate. To plot the undeformed model shape: 1.From the main menu bar, select Plot->Undeformed Shape. The undeformed model shape is displayed in the viewport. 2.To change the undeformed plot options, select Options->Undeformed Shape from the main menu bar or click Undeformed Sh The Undeformed Shape Plot Options dialog box appears; by default, the Basic tab is selected. 3.Choose the Filled render style. 4.Click OK. The plot changes to display the current plot settings. Changing the view The default view is isometric, which does not provide a particularly clear view of the plate. To improve the viewpoint, rotate the v toolbar. To rotate the view:

1.From the main menu bar, select View->Specify. The Specify View dialog box appears. 2.From the Specify View dialog box, select the Viewpoint method. Using the Viewpoint method, you enter three values representing the X-, Y-, and Z-position of an observer. You can also speci points upward. 3.Enter the X-, Y-, and Z-coordinates of the viewpoint vector as 0.5,1,1 and the coordinates of the up vector as 1,0,0. 4.Click OK. ABAQUS displays your model in the specified view. You can also specify a view by entering values for rotation angles, zooming, Viewport annotations The default viewport annotations display the triad, title block, and state block in the viewport. You can suppress their display by To change the viewport annotations: 1.From the main menu bar, select Canvas->Viewport Annotation Options. The Viewport Annotation Options dialog box appears; by default, the Triad tab is selected. 2.Toggle off Show Triad. 3.Click the Title Block tab, and toggle off Show Title Block. 4.Click OK. To fit the undeformed model shape in the viewport, select View->Auto-fit. Animation of results Animating your results will provide a general understanding of the dynamic response of the plate under the blast loading. You wi history animation of the deformed shape. To plot the deformed model shape: 1.From the main menu bar, select Plot->Deformed Shape. The deformed model shape is displayed in the viewport. Only feature edges are visible by default. 2.To change the deformed plot options, click Deformed Shape Plot Options in the prompt area. The Deformed Shape Plot Options dialog box appears; by default, the Basic tab is selected. 3.Choose All visible edges and the Filled render style. 4.Click OK. To create the time-history animation: 1.From the main menu bar, select Animate->Time History. ABAQUS/Viewer displays the deformed model shape at the beginning of the simulation and cycles through each available frame throughout the animation. By default, the animation loops through all available frames repeatedly. This can be changed so that t ABAQUS/Viewer displays the movie player controls on the left side of the prompt area. From left to right, these controls perform the following functions: play, stop, first image, previous image, next image, last image 2.To change the animation control options, click Animation Options in the prompt area. The Animation Options dialog box appears; by default, the Player tab is selected. 3.From the Mode option, choose Play once. 4.Click OK. You will see from the animation that as the blast loading is applied, the plate begins to deflect. Over the duration of the load the load has dropped to zero. The maximum displacement occurs at approximately 8 ms, and a displaced plot of that state is shown Figure 5-15 Displaced shape at 8 ms. The animation images can be saved to a file for playback at a later time. To save the animation: 1.From the main menu bar, select Animate->Save As. The Save Image Animation dialog box appears. 2.In the Settings field, enter the file name blast_base. The format of the animation can be specified as either QuickTime or AVI. 3.Choose the QuickTime format, and click OK. The animation is saved as blast_base.mov in your current directory. You can play the animation file using the ABAQUS movie pla To play an animation file using the movie player: 1.From the main menu bar, select Animate->Image Animation. The Open Image Animation dialog box appears. 2.Select the file blast_base.mov. 3.Click OK. 4.The ABAQUS movie player utility appears, and the animation automatically starts to play. To change any of the default options, click Animation Options in the prompt area.

5.After the animation has finished, click Dismiss in the prompt area to exit the ABAQUS movie player utility. History output Since it is not particularly easy to see the deformation of the plate from the deformed plot, it may be better to view the deflection graph. The displacement of the node in the center of the plate is of particular interest since the largest deflection occurs at this To generate a history plot of the central node displacement: Use the following commands to display the displacement history of the central node, as shown in Figure 5-16 (with displacem 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Plot the spatial displacement, U1, at node 411. 3.To save the current X-Y data, click Save As. Name the data DISP. 4.Click Dismiss. The units of the displacements in this plot are meters. Modify the data to create a plot of displacement (in millimeters) versus tim 5.Select Tools->XY Data->Manager. The DISP data are listed in the XY Data Manager dialog box. To create the plot with the displacement values in millimeters, you need to multiply the DISP X-Y data by 1000. 6.Click Create in the XY Data Manager dialog box; then select Operate on XY Data in the Create XY Data dialog box. Click Conti The Operate on XY Data dialog box opens. 7.Click DISP in the XY Data field; then click * in the Operators field. 8.In the text box at the top of the dialog box, enter 1000 following the * symbol so that the expression appears as: "DISP" * 1000 9.Click Plot Expression to see the modified X-Y data. Save the data as U_BASE2. 10.Click Dismiss. 11.To change the plot titles, click XY Plot Options in the prompt area. The XY Plot Options dialog box appears. Click the Titles tab. 12.Click Title Source for the Y-axis and select User-specified. Change the title to Displacement (mm). 13.Click OK. The resulting plot is shown in Figure 5-16 . Figure 5-16 Central displacement as a function of time. The plot shows that the displacement reaches a maximum of 50.1 mm at 7.4 ms and then oscillates after the blast load is remov The other quantities saved as history output in the output database are the total energies of the model. The energy histories can highlight significant physical effects. To generate history plots of the model energies: Use the following commands to display the histories of five different energy output variables--ALLKE, ALLSE, ALLPD, ALLIE, an 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Select the output variable Artificial strain energy: ALLAE for Whole Model. Click Save As. Name the curve ALLAE. 3.Click OK. 4.Similarly, save the history results for the ALLKE, ALLSE, ALLPD, and ALLIE output variables as X-Y data. 5.Click Dismiss. 6.Select Tools->XY Data->Manager. The ALLAE, ALLKE, ALLSE, ALLPD, and ALLIE X-Y data objects are listed in the XY Data Manager. 7.In the XY Data Manager dialog box, select ALLAE, ALLKE, ALLSE, ALLPD, and ALLIE using [Ctrl]+Click; and click Plot to plot t 8.Click Dismiss. To more clearly distinguish between the different curves in the plot, change their line styles. 9.Click XY Curve Options in the prompt area. The XY Curve Options dialog box appears. 10.For the curve ALLSE, select a dashed line style. For the curve ALLPD, select a dotted line style. For the curve ALLAE, select a chain dashed line style. For the curve ALLIE, select the second thinnest line type. 11.Click Dismiss. Customize the appearance of the plot further by changing the plot titles and the position of the legend. 12.To change the plot titles, click XY Plot Options in the prompt area. The XY Plot Options dialog box appears. Click the Titles tab. 13.Change the title for the Y-axis to Energy. 14.Click OK.

15.To change the position of the legend, select Canvas->Viewport Annotation Options. 16.Click the Legend tab. The position of the upper left corner of the legend is determined by the values specified in the % Viewport X and % Viewport Y b 17.Change the % Viewport X value to 65, and the % Viewport Y value to 50. This will position the legend so that the upper left corner is 65% of the way across the viewport from left to right and 50% of the 18.Click OK. The resulting plot is shown in Figure 5-17 . Figure 5-17 Energy terms as a function of time. We can see that once the load has been removed and the plate vibrates freely, the kinetic energy increases as the strain energy therefore, has its maximum strain energy, it is almost entirely at rest, causing the kinetic energy to be at a minimum. Note that the plastic strain energy rises to a plateau and then rises again. From the plot of kinetic energy we can see that the se rebounded from its maximum displacement and is moving back in the opposite direction. We are, therefore, seeing plastic defor Even though there is no indication that hourglassing is a problem in this analysis, study the artificial strain energy to make sure. artificial strain energy is the energy used to control hourglass deformation, and the output variable ALLAE is the accumulated art deformation as the plate deforms, the total internal energy is much greater than the elastic strain energy alone. Therefore, it is m energy to an energy quantity that includes the dissipated energy as well as the elastic strain energy. Such a variable is the total i quantities. The artificial strain energy is approximately 1% of the total internal energy, indicating that hourglassing is not a proble One thing we can notice from the deformed shape is that the central stiffener is subject to almost pure in-plane bending. Using depth of the stiffener is not sufficient to model in-plane bending behavior. While the solution from this coarse mesh appears to we will study how the solution changes when we refine the mesh of the stiffener. Remember that caution must be taken when re time by increasing the number of elements and decreasing the element size. An input file for a model with a refined stiffener mesh is included in ``Blast loading on a stiffened plate,'' Section A.5 (blast_refin stiffener instead of two. This increase in the number of elements increases the solution time by approximately 20%. In addition, two as a result of the reduction of the smallest element dimension in the stiffeners. Since the total increase in solution time is a mesh increases by approximately a factor of 1.2 ?2, or 2.4, over that of the original mesh. Figure 5-18 shows the histories of artificial energy for both the original mesh and the mesh with the refined stiffeners. As antic important question, however, is whether or not the results changed significantly from the original to the refined mesh. Figure 5-1 almost identical in both cases, indicating that the original mesh is capturing the overall response adequately. One advantage of of stress and plastic strain through the stiffeners. Figure 5-18 Artificial energy in the original and refined models. Figure 5-19 Central displacement history in original and refined meshes. Contour plots Use the contour plotting capability of ABAQUS/Viewer to display the von Mises stress and equivalent plastic strain distributions i create the plots: from the main menu bar, select File->Open and choose the file blast_refined.odb. To generate contour plots of the von Mises stress and equivalent plastic strain: 1.From the main menu bar, select Result->Field Output. The Field Output dialog box appears. 2.Select the stress output variable (S) from the Output Variable area. The stress invariants are displayed in the Invariant area. 3.Select the Mises stress invariant. 4.Click Section Points to select a section point. The Section Points dialog box appears. 5.Select shell under Category and the SPOS section point from the list of available section points. 6.Click OK. The Select Plot Mode dialog box appears. 7.Toggle on Contour, and click OK. ABAQUS/Viewer displays a contour plot of the von Mises stress. The view that you set earlier for the animation exercise should be changed so that the stress distribution is clearer. By default, o edges are visible. You will also need to reposition the legend in the plot. 8.From the main menu bar, select View->Views Toolbox to bring up the Views dialog box. Select the isometric view. 9.To show all the visible edges, click Contour Options in the prompt area. The Contour Plot Options dialog box appears. 10.Choose Exterior visible edges. 11.Click OK. 12.To change the legend position back to the default position, select Canvas->Viewport Annotation Options from the main menu 13.Click the Legend tab in the Viewport Annotation Options dialog box. Click Defaults to reset the legend position to its default l 14.Click OK.

Figure 5-20 shows a contour plot of the von Mises stress at the SPOS section point (section point 5). Now contour the equivalent plastic strain. 15.From the main menu bar, select Result->Field Output. The Field Output dialog box appears. 16.Select the equivalent plastic strain output variable (PEEQ) from the Output Variable area. 17.Click OK. Figure 5-21 shows a contour plot of the equivalent plastic strain at the SPOS section point (section point 5). Figure 5-20 Contour plot of von Mises stress. Figure 5-21 Contour plot of equivalent plastic strain. 5.3.8 Reviewing the analysis The objective of this analysis is to study the deformation of the plate and the stress in various parts of the structure when it is su you will need to consider the assumptions and approximations made and identify some of the limitations of the model. Damping Undamped structures continue to vibrate with constant amplitude. Over the 50 ms of this simulation the frequency of the oscillat amplitude vibration is not the response that would be expected in practice since the vibrations in this type of structure would ten oscillations. The energy loss typically occurs by a variety of mechanisms including frictional effects at the supports and damping Consequently, we need to consider the presence of damping in the analysis to model this energy loss. The energy dissipated by already some damping present in the analysis. By default, a bulk viscosity damping (discussed in Chapter 3, "Overview of Explic modeling of high-speed events. In this shell model only linear damping is present. With the default value the oscillations would eventually die away, but it would t Material damping should be used to introduce a more realistic structural response. Modify the material data block to include dam *DAMPING, ALPHA=50.0, BETA=0.0 BETA is the parameter that controls stiffness proportional damping, and at this stage we will leave it set to zero. The time period for the oscillation of the plate is approximately 10 ms, so we need to increase the analysis period to allow enoug the analysis period to 150 ms. The results of the damped analysis clearly show the effect of mass proportional damping. Figure 5-22 shows the displacemen analyses. (We have extended the analysis time to 150 ms for the undamped model to compare the data more effectively.) The p damped analysis the oscillation has decayed to a nearly static condition. Figure 5-22 Damped and undamped displacement histories. Rate dependence Some materials, such as mild steel, show an increase in the yield stress with increasing strain rate. In this example the loading r The *RATE DEPENDENT option is used with the *PLASTIC option to introduce strain-rate dependence. Add the following to the material option block: *RATE DEPENDENT 40.0, 5.0 With this definition of rate-dependent behavior, the ratio of the dynamic yield stress to the static yield stress () is given for an e and are material constants (40.0 and 5.0 in this case). When the *RATE DEPENDENT option is included, the yield stress effectively increases as the strain rate increases. Therefore, be expect a stiffer response in the analysis with rate dependence. Both the displacement history of the central portion of the plate s Figure 5-24 confirm that the response is indeed stiffer when rate dependence is included. Figure 5-23 Displacement of the central node with and without rate dependence. The results are, of course, sensitive to the material data. In this case the values of and are typical of a mild steel, but more pre Figure 5-24 Plastic strain energy with and without rate dependence. 5.4 Hyperelasticity The stress-strain behavior of typical rubber materials, which is shown in Figure 5-25 , is elastic but highly nonlinear. Figure 5-25 Typical stress-strain curve for rubber. This type of material behavior is called hyperelasticity. The deformation of hyperelastic materials, such as rubber, remains elastic ABAQUS/Explicit makes the following assumptions when modeling a hyperelastic material: he material behavior is elastic. he material behavior is isotropic. he material is nearly incompressible (Poisson's ratio is 0.475 by default.) Elastomeric foams are another class of highly nonlinear, elastic materials. They differ from rubber materials in that they have a h loads. They are modeled using the hyperfoam model in ABAQUS/Explicit. The hyperfoam model will not be discussed in this guid 5.4.1 Strain energy potential ABAQUS uses a strain energy potential (U), rather than a Young's modulus and Poisson's ratio, to relate stresses to strains in hy

available: a polynomial model, the Ogden model, the Arruda-Boyce model, and the van der Waals model. Simpler forms of the p neo-Hookean, reduced polynomial, and Yeoh models. The polynomial form of the strain energy potential is the one that is most commonly used. Its form is where is the strain energy potential; is the elastic volume ratio; and are measures of the distortion in the material; and , , and The parameters describe the shear behavior of the material, and the parameters introduce compressibility. If the material were all the values of would be zero, and the second part of the equation shown above could be ignored. If the number of terms, , is and The other hyperelastic material models are similar in concept and are described in the ABAQUS/Explicit User's Manual. You must provide ABAQUS with the relevant material parameters to use a hyperelastic material. For the polynomial form these ar parameters when modeling hyperelastic materials; however, more likely you will be given test data for the materials that you mus material parameters for you (using a least squares fit). 5.4.2 Defining hyperelastic behavior using test data A convenient way of defining a hyperelastic material is to supply ABAQUS with experimental test data. ABAQUS then calculates t tests for which ABAQUS can fit data are: niaxial tension and compression. quibiaxial tension and compression. lanar tension and compression (pure shear). olumetric tension and compression. The deformation modes seen in these tests and the ABAQUS input options used to define the data for each are shown in Figure materials must be given to ABAQUS as nominal stress and nominal strain values. Figure 5-26 Deformation modes and ABAQUS input options for the various experimental tests for defining hyperelastic material Achieving the best material model from your data The quality of the results from a simulation using hyperelastic materials strongly depends on the material test data that you provi ABAQUS calculate the best possible material parameters. Wherever possible, try to obtain experimental test data from more than one deformation state to help ABAQUS form a much mor Figure 5-26 produce equivalent deformation modes for incompressible materials. The following are equivalent tests for incomp niaxial tension ?Equibiaxial compression. niaxial compression ?Equibiaxial tension. lanar tension ?Planar compression. You do not need to include data from a particular test if you already have data from another test that models the same deformat In addition, the following may improve your hyperelastic material model: btain test data for the deformation modes that are likely to occur in your simulation. For example, if your component is loaded compressive, rather than tensile, loading. oth tension and compression data are allowed, with compressive stresses and strains entered as negative values. If possible, since the fit of a single material model to both tensile and compressive data will normally be less accurate for each individual tes rovide more data at the strain magnitudes that you expect the material will be subjected to during the simulation. For example, do not provide much, if any, test data at high strain values (over 100%). erform one-element simulations of the experimental tests, and compare the results ABAQUS calculates to the experimental da deformation mode that is important to you, try to obtain more experimental data for that deformation mode. These one-element simulations are very easy to perform in ABAQUS/CAE. Please consult the ABAQUS/CAE User's Manual for d Stability of the material model It is possible for the hyperelastic material model determined from the test data to be unstable at certain strain magnitudes. ABAQ where unstable behavior will occur and prints a warning message in the data (.dat) file. You should check this information carefu model experiences strains beyond the stability limits. See ``Hyperelastic behavior,'' Section 9.3.1 of the ABAQUS/Explicit User's Manual, for suggestions on improving the accuracy a Compressibility Most solid rubber materials have very little compressibility compared to their shear flexibility. This behavior is not a problem with problem when using other elements, such as plane strain, axisymmetric, and three-dimensional solid elements. For example, in quite satisfactory to assume that the material is fully incompressible: the volume of the material cannot change except for therm as an O-ring used as a seal), the compressibility must be modeled correctly to obtain accurate results. Except for plane stress cases, it is not possible to assume that the material is fully incompressible in ABAQUS/Explicit because each material calculation point. An incompressible material also has an infinite wave speed, resulting in a time step of zero. The that, in many cases, the actual material behavior provides too little compressibility for the algorithms to work efficiently. Thus, ex compressibility for an efficient solution, knowing that this makes the bulk behavior of the model softer than that of the actual ma not the solution is sufficiently accurate or whether the problem can be modeled at all with ABAQUS/Explicit because of this num

material by the ratio of its initial bulk modulus, , to its initial shear modulus, . Poisson's ratio, , also provides a measure of comp Table 5-2 provides some representative values: Table 5-2 Relationship between compressibility and Poisson's ratio. Poisson's ratio 10 0.452 20 0.475 50 0.490 100 0.495 1,000 0.4995 10,000 0.49995 If no value is given for the material compressibility in the hyperelastic option, by default ABAQUS/Explicit assumes , correspondi have ratios in the range of 1,000 to 10,000 ( 0.4995 to 0.49995) and filled elastomers have ratios in the range of 50 to 200 ( 0 than is available in most elastomers. However, if the elastomer is relatively unconfined, this softer modeling of the material's bul cases where the material is highly confined--such as when it is in contact with stiff, metal parts and has a very small amount of -it may not be feasible to obtain accurate results with ABAQUS/Explicit. If you are defining the compressibility rather than accepting the default value, an upper limit of 100 is suggested for the ratio of solution and require the use of excessively small time increments. 5.5 Material damping An ABAQUS/Explicit model often includes energy dissipation mechanisms--dashpots, inelastic material behavior, and the like-need to introduce additional "structural" or general damping, as such damping would be unimportant compared to these other d energy dissipation sources, and some models require additional energy dissipation beyond that provided by other parts of the m damping is also discussed in ``Material damping,'' Section 3.6.3. Material damping in ABAQUS/Explicit is Rayleigh damping, which is defined using the *DAMPING option, a suboption of the *MA with Rayleigh damping: for mass proportional damping and for stiffness proportional damping. Generally, mass proportional da stiffness proportional damping is used to damp out the high-frequency response (to smooth a noisy solution). Mass proportional damping The factor defines a damping contribution proportional to the mass matrix for an element. The damping forces that are introduc The resulting effect can be likened to the model moving through a viscous ``ether'' so that any motion of any point in the model Stiffness proportional damping The factor defines damping proportional to the elastic material stiffness. A ``damping stress,'' , proportional to the total strain ra where is the strain rate. For hyperelastic and hyperfoam materials is defined as the initial elastic stiffness. For all other material added to the stress caused by the constitutive response at the integration point when the dynamic equilibrium equations are form introduced for any nonlinear case and provides standard Rayleigh damping for linear cases. For a linear case stiffness proportio equal to times the stiffness matrix. To avoid a dramatic drop in the stable time increment, the stiffness proportional damping fa the initial stable time increment without damping. 5.6 Summary BAQUS contains an extensive library of material models, including models for metal plasticity and rubber elasticity. he stress-strain data for the metal plasticity model must be defined in terms of true stress and true plastic strain. Nominal stre data. he metal plasticity model assumes nearly incompressible behavior once the material has yielded. or efficiency ABAQUS/Explicit regularizes user-defined material curves by fitting them with curves composed of equally space olynomial, Ogden, Arruda-Boyce, van der Waals, Mooney-Rivlin, neo-Hookean, reduced polynomial, and Yeoh strain energy yperelasticity models allow the material coefficients to be determined directly from experimental test data. The test data must tability warnings may indicate that a hyperelastic material model is unsuitable for the strain ranges you wish to analyze. he default Poisson's ratio for hyperelastic materials in ABAQUS/Explicit is 0.475. Some analyses may require increasing Poisso enerally, mass proportional damping is used to damp low-frequency oscillations, and stiffness proportional damping is used t o avoid a dramatic drop in the stable time increment, the stiffness proportional damping factor, , should be less than or of the without damping. 6. Contact

Many analyses in ABAQUS/Explicit involve contact between separate components. As bodies come into contact, they impose no goal of a contact analysis is to determine the areas of the model that are in contact and the stresses on the bodies due to the c Contact is different from most constraints because it is discontinuous, meaning that a contact constraint must be applied only w to determine whether or not the bodies are in contact and apply the appropriate constraints. If contact is lost, the contact const 6.1 Overview of contact capabilities Contact modeling in ABAQUS/Explicit is based on the concept of surfaces coming into contact with one another. The user must define interactions between the surfaces. The contact capability is very general. Contact can be defined between surfaces of alm deformable surface, a deformable surface can contact a rigid surface, a deformable surface can contact itself, or a set of node Once bodies are in contact, the surface interaction properties, such as the coefficient of friction, are defined. Another option is eventually fail, causing surface debonding. 6.2 Defining contact The contact definition in ABAQUS/Explicit has two parts: the definition of the surfaces that may come into contact and the defin ABAQUS/Standard, the definition of surfaces is part of the model data. In ABAQUS/Explicit contact is applied as a constraint, so and removed at will during the history of the simulation. 6.2.1 Defining surfaces Surfaces are created using the *SURFACE option. The TYPE parameter on this option dictates the type of surface created: he parameter TYPE=ELEMENT (default) on the *SURFACE option creates a surface by specifying the underlying element faces he parameter TYPE=NODE on the *SURFACE option creates a pure slave surface composed of the specified set of nodes. No several reasons, including more accurate frictional forces and the ability to use the balanced master-slave approach, it is better when possible. The concepts of master and slave surfaces are discussed later in this chapter. he TYPE parameter on the *SURFACE option can also be used to specify the specific form of an analytical rigid surface. The d discussed later in this chapter. Surfaces on continuum elements A two-dimensional, first-order continuum element, such as CPE4R, has four faces consisting of the segments defined by nodes The face identifiers consist of the letter ``S'' followed by the face number. A surface is defined by specifying the element numbe Figure 6-1 Face numbers on a two-dimensional, first-order (4-node) element. For example, use the following option block to include face 2 of the element shown in Figure 6-1 in a surface called FLANGE1 *SURFACE, NAME=FLANGE1 5, S2 As is the case for many options in ABAQUS, both element numbers and element sets can be used; the use of element sets can specify both element sets and individual elements in the same *SURFACE option block. For example, the surface TOPSURF con follows: *ELSET, ELSET=TOP, GENERATE 5, 8 *SURFACE, NAME=TOPSURF TOP, S3 5, S4 8, S2 Figure 6-2 Face numbers and elements that form the surface TOPSURF. ABAQUS can determine the free faces of two- and three-dimensional continuum elements automatically and use them to create elements whose free faces compose the surface on the data lines of the *SURFACE option. Element sets as well as individual e that do not contain free faces. For example, the surface shown in Figure 6-2 could also be defined as follows: *SURFACE, NAME=TOPSURF TOP, While generally not necessary, surfaces on continuum elements also can be defined by a set of nodes. To define a node-based following the *SURFACE, TYPE=NODE option. Surfaces on shell, membrane, and rigid elements There are three ways to define surfaces on shell, membrane, and rigid elements: using single-sided surfaces, double-sided sur Using single-sided contact surfaces, you select one of the two available faces as a contact surface. The face whose normal po while the face whose normal points in the direction of the negative element normal is called SNEG, as shown in Figure 6-3 . Figure 6-3 Surfaces on two- and three-dimensional shell, membrane, and rigid elements. Using the right-hand rule, the nodal ordering of an element defines the positive element normal. By default, the contact surface current thickness of the element. When using shell elements, you can instead specify the shell reference surface (the surface de THICK parameter on the *SURFACE option. The following option block defines surface SURF1 as the surface composed of all th

*SURFACE, NAME=SURF1 SHELLS, SPOS Single-sided contact surfaces account for contact only on the face and not on the edge. Using a node-based surface in conjun face and the edge. The following option block defines a node-based surface named EDGE, composed of the edge nodes in a n *SURFACE, TYPE=NODE, NAME=EDGE NEDGE, Figure 6-4 Node-based surface for contact on a shell edge. Double-sided contact surfaces are more general because both the SPOS and SNEG faces and all free edges are included autom either face or on the edges of the elements forming the double-sided surface. For example, a slave node can start on one side the other side during the course of an analysis. No face identifier is specified when defining double-sided contact because both surfaces are available only for three-dimensional elements. The following option block defines surface SURF2 as the surface co *SURFACE, NAME=SURF2 SHELLS, Rigid surfaces Rigid surfaces are the surfaces of rigid bodies. They can be defined as an analytical shape, or they can be based on the underly Analytical rigid surfaces are created by defining a series of connected lines, arcs, and parabolas. The parameter ANALYTICAL S surface (defined with the TYPE parameter on the *SURFACE option) with a rigid body. The *RIGID BODY option must be defined option defines the dimensionality of the surface, and it has three possible values: se TYPE=SEGMENTS to define a two-dimensional analytical rigid surface. se TYPE=CYLINDRICAL to define a three-dimensional analytical rigid surface that is extruded infinitely in the out-of-plane dire se TYPE=REVOLUTION to define a three-dimensional analytical rigid surface of revolution. The following is an example input for the two-dimensional analytical rigid surface named SRIGID shown in Figure 6-5 : *SURFACE, TYPE=SEGMENTS, NAME=SRIGID START, 5.0, 0.0 LINE, 10.0, 0.0 CIRCL, 15.0, 5.0, 10.0, 5.0 where the rigid body is defined previously by *RIGID BODY, ANALYTICAL SURFACE=SRIGID, REF NODE=10000 Figure 6-5 Analytical rigid surface. Discretized rigid surfaces are based on the underlying elements that make up a rigid body; thus, they can be more geometrically are defined using the *SURFACE option in exactly the same manner as surfaces on deformable bodies. Surfaces on one-dimensional, deformable elements Contact surfaces on one-dimensional, deformable elements, such as beams and trusses, must be node-based surfaces. To de numbers and node sets on the data line following the *SURFACE, TYPE=NODE option, as follows: *SURFACE, TYPE=NODE, NAME=SNODE 101, 102, NSET1, NSET2 6.2.2 Defining the interaction between surfaces To define an interaction between two surfaces, you must specify the surfaces as a contact pair and the type of interaction betw Contact pairs Define possible contact between two surfaces in an ABAQUS simulation by specifying the surface names on the *CONTACT PAI When defining a contact pair, you must decide whether to use the small-sliding formulation or the more general finite-sliding fo two surfaces is less than a small proportion of the characteristic length of an element face. The small-sliding formulation is sele *CONTACT PAIR option. The small-sliding contact search algorithm is much less expensive than the finite-sliding algorithm. Each contact pair must refer to a surface interaction definition, in much the same way that each element must refer to an eleme *CONTACT PAIR option to refer to a *SURFACE INTERACTION option, through which the different surface interaction models, su The following example specifies that surfaces FLANGE1 and FLANGE2 can interact with each other and that the amount of relati *CONTACT PAIR, INTERACTION=FRIC, SMALL SLIDING FLANGE1, FLANGE2 A surface may contact itself if only one surface is given on the *CONTACT PAIR option; for example, *CONTACT PAIR, INTERACTION=FRIC TUBE, Single-surface contact is much more computationally expensive than two-surface contact. However, single-surface contact is analysis. In such cases it is not possible to identify prior to the analysis which parts of the surface will interact. Specifying the interaction model Each contact pair must refer to a *SURFACE INTERACTION option block, which can specify a model for the interaction between

which defines the frictional properties of the surfaces, and *BOND, which defines a spot weld that can degrade and fail at the s is frictionless contact with no bonding. The friction model is discussed in ``Surface interaction models,'' Section 6.3.3. The follo named FRIC with a friction coefficient of 0.1: *SURFACE INTERACTION, NAME=FRIC *FRICTION 0.1, 6.3 Contact algorithms The contact search algorithm, the contact constraint, and the surface interaction models available in ABAQUS/Explicit are discus 6.3.1 Contact search algorithm Determining the parts of the contact surfaces that are in contact at any given time is the most expensive part of the contact ana Two-surface contact Because it is possible for a node on one contact surface to contact any of the facets on the opposite contact surface, ABAQUS motions of the contact surfaces. While the contact search algorithm is transparent to the user and is rarely a concern, some situ methods. While the balanced master-slave algorithm is used in most cases, for discussion we consider a pure master-slave alg balanced approach performs the same contact search twice, reversing the master and slave surfaces the second time.) At the beginning of each step an exhaustive, global search is conducted to determine the closest master surface facet for each considers the possibility of contact between every slave node and every master facet, a global search is very expensive and is p shows the global search to determine which of all of the facets on the master surface is the closest facet to slave node 50. The element 10. Node 100 is determined to be the node on that master facet that is closest to slave node 50; therefore, it is designa search is to determine the closest master facet and a tracked master surface node for each slave node. Figure 6-6 Two-dimensional global contact search. Since the cost of each global search is so high, a less expensive local search is performed in most increments. In a local search the previous tracked master surface node to determine the closest facet. In Figure 6-7 the slave surface of the model shown in Since the previous tracked master surface node was 100, the nearest master surface facet of those attached to node 100 (facet Figure 6-7 Two-dimensional local contact search. In this case facet 10 is the closest to node 50. The next step is to determine the current tracked master surface node from the n on facet 10 to slave node 50. The local search continues until the tracked master surface node remains the same from one itera surface node changed from 100 to 101, so the local search continues. Again, the closest master facet is determined from the m Facet 11 is determined to be the closest facet, and node 102 is determined to be the new tracked master surface node. Since n iteration does not change the tracked master surface node, and the local search ends. Since the time increments are very short, for most situations the contacting bodies move a very small amount from one increme the motion of the contact surfaces. However, there are certain situations that may cause the local contact search to fail. One su containing a hole. Figure 6-8 Example in which local contact search may fail. The shaded element face has been identified as the closest master facet to the slave node belonging to a separate, contacting master facet and its neighbors for contact in the next increment. If the slave node later displaces across the hole and reaches th search algorithm will still be checking only the shaded facet and its neighbors. Potential contact between the slave node and ma contact search. To overcome the problem, force ABAQUS/Explicit to perform a global contact search more often because a glob global contact search more often than the default (every 100 increments), set the GLOBTRKINC parameter on the *CONTACT CO between global contact searches. Use caution when using the GLOBTRKINC parameter because frequent global contact searche Single-surface contact Another option is to allow a single surface to contact itself. For example, the inside of a tube could be defined as a single surfac generality and complexity of single-surface contact, a global contact search is performed every few increments, making singlecontact. 6.3.2 Applying the normal contact constraint By default, ABAQUS/Explicit uses a kinematic contact formulation that achieves precise compliance with the contact conditions proceeds under the assumption that contact does not occur. If at the end of the increment there is an overclosure, iterations are with the contact. The depth of each node's penetration, its mass, and the time increment are used to calculate the resisting forc occurring. The normal contact constraint can optionally be enforced with a penalty contact algorithm. The penalty contact algorithm results kinematic algorithm as it allows small residual penetrations of the slave nodes into the master surface. The penalty stiffness is c penetration is not significant. The penalty stiffness is chosen automatically by ABAQUS/Explicit but can be overridden by specify The penalty contact algorithm can model some types of contact that the kinematic contact algorithm cannot. For example, the p surfaces (except when both surfaces are analytical rigid surfaces). To select the penalty algorithm, set the MECHANICAL CONST Pure master-slave approach

In the pure master-slave approach one of the surfaces is the master surface and the other is the slave surface. As the two bodie distributed to the nodes on the master surface. When the kinematic contact formulation is used, these forces and the masses of the master surface nodes. The acceleration corrections for the slave nodes are then determined using the predicted penetration corrections for the master surface nodes. These corrections are used to obtain a corrected configuration in which the contact c is possible for the master surface to penetrate the slave surface, while the slave nodes are strictly constrained against penetratin used, equal and opposite contact forces (proportional to the penetration distance) are applied to the master and slave nodes at by default for shell-to-solid contact, in which the solid is the pure master surface, and for deformable-to-rigid contact, in whic slave constraint when kinematic compliance is used is illustrated in Figure 6-9 . Figure 6-9 Pure master-slave contact constraint with kinematic compliance. Balanced master-slave approach Balanced master-slave contact is more expensive computationally than pure master-slave contact. However, it minimizes the pe accurate results in most cases. The balanced approach simply applies the pure master-slave algorithm twice, reversing the surf obtained with surface 1 as the slave, and another set of corrections is obtained with surface 2 as the slave. The final acceleratio two acceleration corrections, where the weight of the average is specified by the user using the WEIGHT parameter on the *CON 0.5 so that the same weight is used for each of the acceleration corrections. Setting WEIGHT to 1.0 specifies a pure master-sla Conversely, a weight of zero means that the second surface is the master surface. The balanced master-slave contact constrain . Figure 6-10 Balanced master-slave contact constraint with kinematic compliance. 6.3.3 Surface interaction models There are several options for surface interaction models to use once contact is established. Friction model If the *FRICTION option is used, ABAQUS/Explicit uses a Coulomb friction model to resist relative tangential motion between two default friction coefficient is zero. The tangential motion is zero until the surface traction reaches a critical shear stress value, wh following equation: where is the coefficient of friction and is the normal contact stress. Figure 6-11 shows the variation of shear stress with slip, Figure 6-11 Coulomb friction model. Optionally, a friction stress limit can be specified using the TAUMAX parameter on the *FRICTION option. (TAUMAX cannot be us as a node-based surface) The *FRICTION option is a suboption of the *SURFACE INTERACTION option. In the following example, a coefficient of friction o *SURFACE INTERACTION, NAME=INTER *FRICTION, TAUMAX=80.0E6 0.1, Often the friction coefficient at the initiation of slipping from a sticking condition is different from the friction coefficient when sli coefficient, and the latter is referred to as the kinetic friction coefficient. In ABAQUS/Explicit an exponential decay law is availab Figure 6-12 ). Figure 6-12 Exponential decay friction model. These friction coefficients are defined when the EXPONENTIAL DECAY parameter is used on the *FRICTION option. *FRICTION, EXPONENTIAL DECAY ,, Frictional constraints are imposed using a predictor/corrector algorithm much like that used to apply the normal contact constra opposite surface in the predicted configuration is calculated using the mass associated with the node, the distance the node ha calculated using this force is greater than , the surfaces are slipping, and the force corresponding to is applied. In either case t surface at the slave node and the nodes of the master surface facet that it contacts. Tied contact Tied contact bonds together the tangential and normal motions of surfaces initially in contact throughout the analysis. Rotationa specified only in the first step by using the TIED parameter on the *CONTACT PAIR option. If the surfaces specified under *CON contact will not be imposed, and the two surfaces will never interact throughout the analysis. To force the surfaces to be exactly parameter on the *CONTACT PAIR option. ADJUST forces any initially overclosed nodes, as well as any open nodes within the s have to be adjusted by distances that are a large fraction of the length of the side of the element (to which the node is attached adjustments if possible. Tied contact is particularly useful for rapid mesh refinement between dissimilar meshes. Spot welds and debonding Nodes can be ``spot welded'' to the nearest point on the master surface using the *BOND option, which is a suboption of *SURF master surface can yield and degrade until it eventually fails, according to the specified failure criterion. Once the bond has faile nonbonded slave node in a pure master-slave contact pair. Since a node set is specified under *BOND, the option can be asso

set is the slave. Use the WEIGHT parameter on the *CONTACT PAIR option to force the pure master-slave relationship. The resu except that in this case the bond is imperfect and can yield and fail. If the slave nodes are not exactly in contact with the maste in a strain-free manner so that they are in contact. The bond carries a normal force and a shear force, and the yield criterion uses a combination of the two forces. The post-yield or a damaged failure model. The time-to-failure model gradually shrinks the yield surface to zero over a specified time period, w the basis of the total energy dissipated in the Mode I and Mode II response. For further information, see the ABAQUS/Explicit Us Other contact interaction options Other contact interaction options include adhesive contact behavior, softened contact behavior, and viscous contact damping. T can be found in the ABAQUS/Explicit User's Manual. 6.4 Modeling considerations We now discuss the following modeling considerations: correct definition of surfaces, possibility of overconstraining the model, 6.4.1 Correct surface definitions Certain rules must be followed when defining contact surfaces. All surfaces must be continuous and simply connected, while a v requirement has several implications for what constitutes a valid or invalid surface definition. In two dimensions the surface must be either a simple, nonintersecting curve with two terminal ends or a closed loop. Figure 6surfaces. Figure 6-13 Valid and invalid two-dimensional surfaces. In three dimensions an edge of an element face belonging to a valid surface may be either on the perimeter of the surface or sh surface cannot be joined just at a shared node; they must be joined across a common element edge. An element edge cannot b illustrates valid and invalid three-dimensional surfaces. In addition, it is possible to define two- or three-dimensional double-sided surfaces. In this case both sides of a shell, membra as shown in Figure 6-15 . Extending surfaces ABAQUS/Explicit does not extend surfaces automatically beyond the perimeter of the surface defined by the user. Figure 6-14 Valid and invalid three-dimensional surfaces. Figure 6-15 Valid double-sided surface. If a node from one surface is in contact with another surface and it slides along the surface until it reaches an edge, it may ``fal node may later reenter from the back side of the surface, thereby violating the kinematic constraint and causing large jumps in a to extend surfaces somewhat beyond the regions that will actually contact. Figure 6-16 shows two simple box-like bodies constructed of brick elements. The upper box has a contact surface defined on definition in ABAQUS/Explicit, the lack of extensions beyond the ``raw edge'' could be problematic. In the lower box the surface beyond the flat, upper surface. If contact is to occur only at the top face of the box, this extended surface definition minimizes c behind the contact surface. Figure 6-16 Surface perimeters. Mesh seams Defining two nodes with the same coordinates (double noding) can generate a seam or crack in a valid surface that appears to the surface can fall through this crack and slide behind the contact surface. The contact algorithm will cause a large, nonphysic seams can be detected in ABAQUS/Viewer by drawing the free edges of the model. Any seams that are not part of the desired p Figure 6-17 Example of a double-noded mesh. Complete surface definition Figure 6-18 illustrates a two-dimensional model of a simple connection between two parts. The contact definition shown in the surfaces do not represent a complete description of the geometry of the bodies. At the beginning of the analysis some of the no Figure 6-19 shows an adequate surface definition for this connection. The surfaces are continuous and describe the entire ge Figure 6-18 Example of an incorrect surface definition. Figure 6-19 Correct surface definition. Consistent surface normals Normal directions must be consistent for surfaces with underlying shell, membrane, or rigid elements. The face identifier SPOS i normal, and the face identifier SNEG indicates the reverse. The user must define surfaces on elements of these types so that the shows a mesh of SAX1 elements whose normals are not continuous from one element to the next. If a surface was defined usin issue a warning message stating that the surface is not valid. Figure 6-20 Inconsistent surface normals. A valid surface could be defined with this mesh if the surface definition shown in the figure, which uses both SPOS and SNEG fa is used. Highly warped surfaces When a surface is highly warped from one facet to the next, a more expensive search algorithm must be used than the algorithm

solution as efficient as possible, ABAQUS monitors the warpage of the surfaces and issues a warning if surfaces become too w than 20? ABAQUS issues a warning message. Once a surface is deemed to be highly warped, ABAQUS switches from the more algorithm to account for the difficulties posed by the highly warped surface. For the sake of efficiency ABAQUS does not check for highly warped surfaces every increment. Rigid surfaces are checked for h not change shape during the analysis. Deformable surfaces are checked for high warpage every 20 increments by default. Some severity quite suddenly, making the default 20 increment frequency check inadequate. The user can change the frequency of the on the *CONTACT CONTROLS option to the desired number of increments. Some analyses in which the surface warping is less algorithm associated with highly warped surfaces. Use the WARP CUT OFF parameter on the *CONTACT CONTROLS option to re Rigid element discretization Complex rigid surface geometries can be modeled using rigid elements. Rigid elements in ABAQUS/Explicit are not smoothed; th of unsmoothed surfaces is that the surface defined by the user is exactly the same as the surface used by ABAQUS; the disadva refinement to define smooth bodies accurately. In general, using a large number of rigid elements to define a rigid surface does number of rigid elements does increase the memory overhead significantly. The user must ensure that the discretization of any curved geometry on rigid bodies is adequate. If the rigid body discretization ``snag,'' leading to erroneous results, as illustrated in Figure 6-21 . Figure 6-21 Potential effect of coarse rigid body discretization. A node that is snagged on a sharp corner may be trapped from further sliding along the rigid surface for some time. Once enoug will snap dynamically before contacting the adjacent facet. Such motions cause noisy solutions. The more refined the rigid surfa Analytical rigid surfaces are smooth and are computationally efficient. They should normally be used for rigid bodies whose shap 6.4.2 Overconstraining the model Just as multiple conflicting boundary conditions should not be defined at a given node, multi-point constraints and contact con they may generate conflicting kinematic constraints. Unless the constraints are entirely orthogonal to one another, the model wil ABAQUS/Explicit tries to satisfy the conflicting constraints. 6.4.3 Mesh refinement For contact as well as all other types of analyses, the solution improves as the mesh is refined. For contact analyses using a pu surface is refined adequately so that the master surface facets do not overly penetrate the slave surface. By default, deformable which does not require high mesh refinement on the slave surface to have adequate contact compliance. Mesh refinement is ge bodies; in this case the deformable body is always the pure slave surface and, thus, must be refined enough to interact with any penetration that can occur if the discretization of the slave surface is poor compared to the dimensions of the features on the m penetrations of the rigid surface would be much less severe. Figure 6-22 Example of inadequate slave surface discretization. Tied contact for mesh refinement Including the TIED parameter on the *CONTACT PAIR option prevents surfaces initially in contact from penetrating, separating, o easy means of mesh refinement. Since any gaps that exist between the two contact surfaces, however small, will result in nodes the ADJUST parameter to ensure that the two surfaces are exactly in contact at the start of the analysis. As the tied contact formulation constrains only translational degrees of freedom, any rotational degrees of freedom are left unco stiffener made of shell elements perpendicularly to a plate of shell elements, the stiffener will be free to pivot about its contact e you must ensure that the unconstrained rotations will not cause problems. 6.4.4 Initial contact overclosure ABAQUS/Explicit does not allow any initial overclosure of contact surfaces. The undeformed coordinates of nodes on contact su When using the balanced master-slave approach, both surfaces are adjusted; when using the pure master-slave approach, only adjusting the surface to remove overclosures do not cause any initial strain or stress for contact pairs defined in the first step of contact will not always resolve initial overclosures exactly, resulting in stresses even in the first increment. Switching to the pure In subsequent steps any nodal adjustments to remove initial overclosures cause strains that often cause severe mesh distortions brief increment. For example, if the node is overclosed by 1.0 ?10-3 m and the increment time is 1.0 ?10-7 s, the acceleration Such a large acceleration applied to a single node typically will cause warnings about deformation speed exceeding the wave sp few increments later, once the large acceleration has deformed the associated elements significantly. Even a very slight initial ov the small time increments. Consequently, it is important that in Step 2 and beyond any new contact surfaces that you define are Figure 6-23 shows a common case of initial overclosure of two contact surfaces. All of the nodes on the contact surfaces lie e inner surface is finer than that of the outer surface and since the element edges are linear, some nodes on the finer, inner surfac Figure 6-23 Original overclosure of two contact surfaces. Assuming that the pure master-slave approach is used, Figure 6-24 shows the initial, strain-free displacements applied to the external loads this geometry is stress free. If the default, balanced master-slave approach is used, a different initial set of displa free. Figure 6-24 Corrected contact surfaces.

6.5 Example: circuit board drop test In this example you will investigate the behavior of a circuit board in protective crushable foam packaging dropped at an angle o packaging is adequate to prevent circuit board damage when the board is dropped from a height of 1 meter. Figure 6-25 show millimeters and the material properties. Figure 6-25 Dimensions in millimeters and material properties. 6.5.1 Coordinate system While the circuit board will be dropped at an angle, it is easiest to use the *SYSTEM option to define the mesh aligned with a loc *SYSTEM option transforms nodal coordinates from the local coordinate system to the global coordinate system. This option allo coordinate system, which is rotated by the desired angle relative to the global coordinate system. The *SYSTEM option defines a new coordinate system by specifying three points: a local origin, a point on the local x-axis, and circuit board, use the following option to tilt the mesh so that it lands on its corner: *SYSTEM 0., 0., 0., .5, .707, .25 -.5, .707, -.5 All subsequent nodal definitions will be in this local coordinate system. To reset the coordinate system to the default, use anoth 6.5.2 Mesh design The overall mesh for this problem is shown in Figure 6-26 . Figure 6-26 Mesh of the circuit board and foam packaging. Define the circuit board so that the shell normals are in the direction indicated. Defining the bottom corner of the foam packagin the circuit board and packaging. Since the ground onto which the board will be dropped is effectively rigid, use a single R3D4 e dimensional solid structure that should be modeled using C3D8R elements. The circuit board itself can be considered as a thin, circuit board with S4R elements, and model the chips with MASS elements. Since you will be using shell elements for the circuit board, ABAQUS/Explicit will, by default, use the current shell element thickn the foam packaging are both the same thickness (2 mm) so that there is a snug fit between the two bodies. In this example the packaging is a mesh of 6 ?7 ?15 elements, as shown in Figure 6-27 . Figure 6-27 Packaging mesh detail. The MASS elements are positioned as shown in Figure 6-28 . The mesh for the packaging is too coarse near the impacting cor adequate for a low-cost preliminary study. 6.5.3 Node and element sets Figure 6-28 and Figure 6-29 show all of the sets necessary to apply the element properties, loads, initial conditions, and bou postprocessing. Figure 6-28 Position of mass elements on circuit board. Numbers in parentheses are (x, y) coordinates in millimeters based on Figure 6-29 Necessary node and element sets. Include the circuit board elements in an element set called BOARD, and include the corresponding circuit board nodes in a node elements in an element set called PACK, and include the nodes in a node set called PACK. The element sets will be used to refe node sets will be used to apply initial conditions. Include the bottom row of elements in the circuit board in an element set called set called FLOOR containing the floor's rigid element and a node set called REF containing the reference node for the rigid surfa chips in an element set called CHIPS. 6.5.4 Simulating free fall Two methods could be used to simulate the circuit board being dropped from a height of 1 meter. You could model the circuit b allow ABAQUS/Explicit to calculate the motion under the influence of gravity; however, this method is clearly impractical becaus ``free-fall'' part of the simulation. The most efficient method is to model the circuit board and packaging in an initial position ver meter drop (4.439 m/s). 6.5.5 Reviewing the input file--the model data We now review the model data required for this simulation, including the model description, the node and element definitions, el and surface definitions. Model description Using the *HEADING option, provide a suitable heading for your model, such as the one shown below. You should use SI units i *HEADING Circuit board drop test 1.0 meter drop SI units (kg, m, s, N) Nodal coordinates and element connectivity Use your preprocessor to generate the mesh in the local coordinate system. Precede the nodal definitions with the *SYSTEM op described previously. Your nodal definitions for the foam packaging and circuit board will look something like the following:

*SYSTEM 0., 0., 0., .5, .707, .25 -.5, .707, -.5 *NODE 1, 0.005, -0.010, 0.012 11, 0.005, -0.010, 0.162 . . . ** ** Reset coordinate system ** *SYSTEM When you have finished defining the nodes in the rotated, local coordinate system, use the *SYSTEM option again without any d global coordinate system. Define the nodes for the rigid surface so that it is large enough to keep the deformable bodies from fa the bottom corner of the foam packaging to ensure that there is no initial overclosure of the contact surfaces. Element properties Give each element set appropriate section properties. Include the appropriate MATERIAL parameter on each section option so th named the foam packaging material FOAM, and we will define it in the next section. *SOLID SECTION, ELSET=PACK, MATERIAL=FOAM For the circuit board it is most meaningful to output stress results in the longitudinal and lateral directions, aligned with the edge directions for the circuit board mesh. We can use the same local coordinate system that we previously defined using the *SYSTE the *ORIENTATION option with the DEFINITION=COORDINATES parameter. On the first data line specify the x-, y-, and z-coord coordinate system. On the second data line specify an additional rotation of zero degrees about the local 3- (or z-) axis. The n SECTION option. *SHELL SECTION, ELSET=BOARD, MATERIAL=PCB, ORIENTATION=OR1 0.002, *ORIENTATION, NAME=OR1, SYSTEM=RECTANGULAR, DEFINITION=COORDINATES 0.5, 0.707, 0.25, -0.5, 0.707, -0.5 3, 0.0 Define the mass of each of the chips on the circuit board to be 0.005 kg using the *MASS option. *MASS, ELSET=CHIPS 0.005, Define the rigid body by referring to the element set FLOOR and the rigid body reference node on the *RIGID BODY option. You not the node set name. *RIGID BODY, ELSET=FLOOR, REF NODE=<reference node number> Material properties You now need to define the material properties for the circuit board and the foam packaging. For the circuit board use a PCB pl of 0.3, and a density of 500 kg/m3. *MATERIAL, NAME=PCB *ELASTIC 45.E9, 0.3 *DENSITY 500.0, Model the foam packaging material using the crushable foam plasticity model. Use the *ELASTIC option to define the Young's m density is 100.0 kg/m3. *MATERIAL, NAME=FOAM *ELASTIC 3.E6, 0.0 *DENSITY 100.0, The yield surface of a crushable foam in the p-q (pressure stress-Mises equivalent stress) plane is illustrated in Figure 6-30 . Figure 6-30 Crushable foam model: yield surface in the p-q plane. The *CRUSHABLE FOAM, HARDENING=VOLUMETRIC option uses two data items to define the initial yield behavior.

*CRUSHABLE FOAM, HARDENING=VOLUMETRIC 1.1, 0.1 The first data item is the the ratio of initial yield stress in uniaxial compression to initial yield stress in hydrostatic compression, ; yield stress in hydrostatic tension to initial yield stress in hydrostatic compression, . This data item is given as a positive value; i Figure 6-31 Foam hardening material data. Include hardening effects with the *CRUSHABLE FOAM HARDENING option. The first data item on each line is the yield stress in item on each line is the absolute value of the corresponding plastic strain. The crushable foam hardening model follows the curv *CRUSHABLE FOAM HARDENING 0.22000E6, 0.0 0.24651E6, 0.1 0.27294E6, 0.2 0.29902E6, 0.3 0.32455E6, 0.4 0.34935E6, 0.5 0.37326E6, 0.6 0.39617E6, 0.7 0.41801E6, 0.8 0.43872E6, 0.9 0.45827E6, 1.0 0.49384E6, 1.2 0.52484E6, 1.4 0.55153E6, 1.6 0.57431E6, 1.8 0.59359E6, 2.0 0.62936E6, 2.5 0.65199E6, 3.0 0.68334E6, 5.0 0.68833E6, 10.0 Boundary conditions Fully constrain the rigid surface representing the floor by applying a fixed boundary condition to the reference node, which you p *BOUNDARY REF, ENCASTRE Initial conditions Give the circuit board and foam packaging an initial velocity of -4.43 m/s in the global 3-direction, corresponding to the velocit *INITIAL CONDITIONS, TYPE=VELOCITY BOARD, 3, -4.43 PACK, 3, -4.43 Surface definitions Define the surfaces that will experience contact during the analysis. You should determine not only which surfaces are in initial c during the analysis. The simplest and most general approach is to define the following surfaces: he bottom portion of the circuit board. he floor. he foam packaging. Both sides of the circuit board can be included in the same double-sided surface definition by simply omitting the face identifie packaging can be included in a surface definition by simply specifying the element set PACK and letting ABAQUS determine the its normal points toward the foam packaging. Depending on how you defined the rigid element for the floor, either SPOS or SNE *SURFACE, NAME=LOWBOARD LOWBOARD, *SURFACE, NAME=PACK PACK, *SURFACE, NAME=FLOOR FLOOR, <SPOS or SNEG> 6.5.6 Reviewing the input file--the history data Use the *DYNAMIC, EXPLICIT option to select a dynamic stress/displacement analysis using explicit integration, and define the *STEP

*DYNAMIC, EXPLICIT , 0.02 Creating the contact pairs Use the *CONTACT PAIR option to indicate which contact surfaces will interact during the analysis. Each contact pair must refer INTERACTION parameter. Specify a coefficient of friction of 0.3 between each of the contacting surfaces. By default, ABAQUS/E shell elements and continuum elements. For this analysis, however, it is desirable to use the balanced master-slave approach in option to 0.5. *CONTACT PAIR, INTERACTION=FRIC, WEIGHT=0.5 PACK, LOWBOARD *CONTACT PAIR, INTERACTION=FRIC PACK, FLOOR *SURFACE INTERACTION, NAME=FRIC *FRICTION 0.3, Output requests Write the preselected field data to the output database file at 20 evenly spaced intervals during the analysis by including the follo *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=20 Write values of nodal displacement (U), velocity (V), and acceleration (A) for each of the attached chips as history data to the ou strain (LE), stress (S), principal stress (SP), and logarithmic principal strain (LEP) at the surfaces (section points 1 and 5) of the most chip is attached. Select an output interval of 0.1 ms. *OUTPUT, HISTORY, TIME INTERVAL=0.1E-3 *NODE OUTPUT, NSET=CHIPS U, V, A *ELEMENT OUTPUT, ELSET=BOTPART 1, 5 LE, S, SP, LEP Write energy values summed over the entire model. Specifically, write values for kinetic energy (ALLKE), internal energy (ALLIE), energy dissipated by plastic deformation (ALLPD). *ENERGY OUTPUT ALLIE, ALLKE, ALLPD, ALLAE, ALLSE Indicate the end of the step with the *END STEP option. 6.5.7 Running the analysis Save your input in a file called circuit.inp. Run the analysis using the following command: abaqus job=circuit analysis This analysis is somewhat more complicated than the previous analyses in this guide, and it may take 45 minutes or more to run Status file Information concerning the initial stable time increment can be found at the top of the status file. The 10 most critical elements ( shown in rank order. ------------------------------------------------------------------------------MODEL INFORMATION (IN GLOBAL X-Y COORDINATES) ------------------------------------------------------------------------------Total mass in model = 3.49594E-02 Center of mass of model = (-1.076768E-02, 4.948696E-02, 8.492264E-02) Moments of Inertia : About Center of Mass I(XX) 6.655609E-05 I(YY) 9.977436E-05 I(ZZ) 6.921255E-05 I(XY) -1.344123E-05 I(YZ) -5.240334E-06 I(ZX) 3.958669E-05

About Origin 4.042923E-04 3.559498E-04 1.588800E-04 5.187227E-06 -1.521594E-04 7.155425E-05

-------------------------------------------------------------------------------

STABLE TIME INCREMENT INFORMATION -------------------------------------------------------------------------------

The stable time increment estimate for each element is based on linearization about the initial state.

Initial time increment = 7.11932E-07 Statistics for all elements: Mean = 1.04549E-05 Standard deviation = 4.05159E-06 Most critical elements : Element number Rank Time increment Increment ratio ---------------------------------------------------------42 1 7.119323E-07 1.000000E+00 87 2 7.119323E-07 1.000000E+00 64 3 7.119323E-07 9.999999E-01 89 4 7.119323E-07 9.999999E-01 5 5 7.119324E-07 9.999998E-01 21 6 7.119324E-07 9.999998E-01 30 7 7.119324E-07 9.999998E-01 39 8 7.119324E-07 9.999998E-01 47 9 7.119324E-07 9.999998E-01 62 10 7.119324E-07 9.999998E-01

***WARNING: At least one pair of adjacent slave nodes (2035,2034) on slave surface PACK are on the opposite sides of double-sided master surface LOWBOARD.

Maximum overclosure between surface PACK and surface LOWBOARD is 1.00000E-03.

***WARNING: In contact pair 1, node 452 on the slave surface LOWBOARD has 143.02 times more mass than the mass of node 16064 on the master surface PACK. Significant contact noise may result with the kinematic contact algorithm when the slave nodes have significantly greater mass than the master nodes. Suggested workarounds include setting the WEIGHT parameter so that surface PACK is a pure slave surface in the contact pair, using mass-scaling to adjust the ratio of nodal masses, or using the penalty contact algorithm.

------------------------------------------------------------------------------SOLUTION PROGRESS ------------------------------------------------------------------------------STEP 1 ORIGIN 0.00000E+00

Total memory used for step 1 is approximately 750.7 kilowords Global time estimation algorithm will be used. Scaling factor : 1.0000 STEP TOTAL CPU STABLE CRITICAL KINETIC INCREMENT TIME TIME TIME INCREMENT ELEMENT ENERGY 0 0.000E+00 0.000E+00 00:00:01 7.119E-07 87 3.430E-01 Results number 0 at increment zero. ODB Field Frame Number 0 of 20 requested intervals at increment zero.

***WARNING: Master surface PACK of contact pair # 1 contains facets with out-of-plane warping of at least 20.000 degrees in increment 280. Large warping that develops during an analysis often corresponds to severe distortion of the underlying elements. It may be appropriate to rerun the analysis with a refined mesh.

1181 1.000E-03 1.000E-03 00:01:28 8.418E-07 39 3.126E-01 ODB Field Frame Number 1 of 20 requested intervals at 1.000320E-03 2351 2.000E-03 2.000E-03 00:02:55 8.756E-07 50 2.269E-01 : : : The status file indicates several warnings were generated during the analysis. As a general rule you should always investigate th due to modeling errors or whether they can simply be ignored. The first warning states that adjacent slave nodes (on the foam p master surface (the lower part of the circuit board). This is to be expected since only one element spans the slit at the base of t positioned in the center of this slit. The location of the offending nodes is visualized easily in ABAQUS/Viewer and confirms our The second warning states that there is a large mass mismatch between a slave node on the circuit board and a master node on solution when kinematic contact is used. You are not required to eliminate the source of this warning; however, if you choose to reasonable. In this example we choose to ignore the warning. The final warning concerns out-of-plane warping. This warning is typically issued during impact and postbuckling analyses sinc problems. If contact difficulties develop as a result, it may be necessary to refine your mesh. 6.5.8 Postprocessing Start ABAQUS/Viewer by typing the following: abaqus viewer odb=circuit at the operating system prompt. Checking material directions The material directions obtained from this orientation definition can be checked with ABAQUS/Viewer. To plot the material orientation: 1.First, change the view to a more convenient setting. From the toolbar, click . From the Views dialog box that appears, select the 1-3 setting. 2.From the main menu bar, select Plot->Material Orientations. The orientation of the material directions of the circuit board at the end of the simulation are shown. The material directions are material 2-direction is yellow, and the 3-direction, if it is present, is red. 3.To view the initial material orientation, select Result->Step/Frame. In the Step/Frame dialog box that appears, select Incremen ABAQUS/Viewer displays the initial material directions. 4.To restore the display to the results at the end of the analysis, select the last increment available in the Step/Frame dialog box Animation of results You will create a time-history animation of the deformation to help you visualize the motion and deformation of the circuit board To create a time-history animation: 1.From the main menu bar, select Plot->Deformed Shape; or click in the toolbox. The deformed model shape at the end of the analysis appears. The plot shows only the feature edges of the model. Change this 2.Click Deformed Shape Plot Options in the prompt area. The Deformed Shape Plot Options dialog box appears. 3.In this dialog box, choose the Hidden render style. Click OK.

4.From the main menu bar, select Animate->Time History. The animation controls appear in the prompt area, and the animation of the deformed model shape begins. 5.In the prompt area, click to stop the animation after a full cycle has been completed. Change the animation options so that ABAQUS/Viewer swings through the animation at a faster rate. 6.From the main menu bar, select Options->Animation; or click Animation Options in the prompt area to open the Animation Op 7.Do the following: a.Choose Swing. b.Drag the frame rate slider to Fast. 8.Click OK, and replay the animation with these settings. Because you chose the Swing repetition mode, the animation steps backward through each frame when it reaches the end of th analysis. ABAQUS/Viewer steps through the animation at a faster rate, because the frame rate has been increased. The deforme is shown in Figure 6-32 . Figure 6-32 Deformed mesh at 4 ms. Plotting model energy histories Plot graphs of various energy variables versus time. Since accelerations and reaction forces can change significantly from one variables with enough data points to provide an accurate representation of the results. To plot energy histories: 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Select the ALLAE output variable, and save the data as Artificial Energy. 3.Select the ALLIE output variable, and save the data as Internal Energy. 4.Select the ALLKE output variable, and save the data as Kinetic Energy. 5.Select the ALLPD output variable, and save the data as Plastic Dissipation. 6.Select the ALLSE output variable, and save the data as Strain Energy. 7.Click Dismiss. 8.From the main menu bar, select Tools->XY Data->Manager. The XY Data Manager dialog box appears. 9.In this dialog box, select all five curves and click Plot to view the X-Y plot. The selected curves are plotted in the viewport. 10.Click Dismiss to close the dialog box. Customize the appearance of the plot. Change the line styles of the curves, change the number of decimal places displayed on change the titles of the X- and Y-axes. 11.Click XY Curve Options in the prompt area. The XY Curve Options dialog box appears. 12.In this dialog box, select different line styles and thicknesses for the curves displayed in the viewport; then click Dismiss. 13.Click XY Plot Options in the prompt area. The XY Plot Options dialog box appears. 14.In this dialog box: a.Click the Axes tab, and set the number of decimal places to be displayed on the X-axis to zero. b.Click the Tick Marks tab, and set the number of minor tick marks to be displayed on both axes equal to zero. c.Click the Titles tab, and select user-specified title sources for both the X- and Y-axes. Enter an X-axis title of Time and a Y-a 15.Click OK to apply the settings. Now reposition the legend so that it appears inside the plot. 16.From the main menu bar, select Canvas->Viewport Annotation Options. The Viewport Annotation Options dialog box appears. 17.Click the Legend tab, and enter a relative position of (50,75) for the upper left corner of the legend box in the viewport. The energy histories appear as shown in Figure 6-33 . Figure 6-33 Energy results versus time. First, consider the kinetic energy history. At the beginning of the simulation the components are in free fall, and the kinetic energ reducing the kinetic energy. The components then rotate about the impacting corner until the side of the foam packaging impac energy. The bodies remain in contact through most of the remainder of the simulation. The deformation of the foam packaging during impact transfers energy from kinetic energy to internal energy in the foam packag internal energy increases as the kinetic energy decreases. In fact, the internal energy is composed of elastic energy and plastica 33 . Elastic energy rises to a peak and then falls as the elastic deformation recovers, but the plastically dissipated energy contin Another important energy output variable is the artificial energy, which is a substantial fraction (approximately 15%) of the intern

of the solution would improve if the artificial energy could be decreased to a smaller fraction of the total internal energy. What causes high artificial strain energy in this problem? In the example in Chapter 4, "Finite Elements and Rigid Bodies," we saw that contact at a single node--such as the corner impa coarse mesh. That example posed two methods of reducing the artificial strain energy: refining the mesh or rounding the impact with the original mesh, realizing that improving the mesh would lead to an improved solution. Stresses and strains in the circuit board The next result we wish to examine is the stress and strain in the circuit board near the location of the chips. If the stress or stra securing the chips to the board will fail. Study the lateral and longitudinal stress histories at the top face (SPOS) of the element s most chip is attached. To plot stress histories: 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Select the S11 stress component on the SPOS surface of the element in element set BOTPART, and save the data as Lateral. 3.Select the S22 stress component on the SPOS surface of the element in element set BOTPART, and save the data as Longitud 4.Click Dismiss. 5.From the main menu bar, select Tools->XY Data->Manager. The XY Data Manager dialog box appears. 6.In this dialog box, select the two curves defined above; and click Plot to view the X-Y plot. 7. Click Dismiss to close the dialog box. As before, customize the plot appearance to obtain a plot similar to Figure 6-34 . Figure 6-34 Stress on the top surface of the circuit board. The stress plot shows that the longitudinal stress under the chip varies most of the time between -2 MPa and 2 MPa, with some trend with less severe peaks. Both the lateral and the longitudinal stresses have peaks at approximately 7 ms. This point has bee as the instant at which the side of the packaging hits the floor. We wish to identify the peak strain in any direction. Therefore, the maximum and minimum principal logarithmic strains are of int To plot logarithmic strain histories: 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Select the principal logarithmic strain LEP1 on the SPOS surface of the element in element set BOTPART, and save the data as 3.Select the principal logarithmic strain LEP2 on the SPOS surface of the element in element set BOTPART, and save the data as 4.Click Dismiss to close the dialog box. 5.From the main menu bar, select Tools->XY Data->Manager. The XY Data Manager dialog box appears. 6.In the dialog box, select the two curves defined above; and click Plot to view the X-Y plot. The settings of the previous plot remain in effect and are inappropriate for this plot. Change the settings to make this X-Y plot m Figure 6-35 Principal logarithmic strain values versus time. The strain curves show peak and decay behavior similar to that of the stresses. The peak values are approximately 0.06% compr Acceleration and velocity histories at the chips Another result that may assist us in determining the desirability of the foam packaging is the acceleration and velocity of the chip impact may damage the chips, even though they may remain attached to the circuit board. Therefore, we need to plot the accel accelerations to be greatest in the 3-direction, we plot the variable A3. To plot acceleration histories: 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Select the acceleration A3 of the nodes in node set Chips, and save the three X-Y data objects as Chip 6001, Chip 6002, and test,'' Section A.6, these correspond to nodes 60, 357, and 403, respectively. 3.Click Dismiss. 4.From the main menu bar, select Tools->XY Data->Manager. The XY Data Manager dialog box appears. 5.In this dialog box, select the three curves created above; and click Plot. The X-Y plot appears in the viewport. As before, customize the plot appearance to obtain a plot similar to Figure 6-36 . Figure 6-36 Acceleration of the three chips in the z-direction. The acceleration of chip 6001, which is at the top of the board, is smooth compared to the accelerations of the other chips earl However, for chip 6002 the peak acceleration is 2000 m/s2 at approximately 10 ms, while for chip 6003 the peak acceleration is the packaging impacts the floor. This result may be accentuated by the board's rotation following the initial impact.

The final plot, Figure 6-37 , shows how the three chips change velocity during impact. Figure 6-37 Velocity of the three chips in the z-direction. At the beginning of the simulation, the chips all have the same downward velocity after the 1 meter drop. The chips all decelerat profiles diverge rapidly. Only the component near the base of the board, chip 6003, reaches a zero velocity at the end of the an To plot velocity histories: 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Select the velocity V3 of the nodes in node set Chips, and save the X-Y data objects as V3 Chip 6001, V3 Chip 6002, and V3 test,'' Section A.6, these correspond to nodes 60, 357, and 403, respectively. 3.Click Dismiss to close the dialog box. 4.From the main menu bar, select Tools->XY Data->Manager. The XY Data Manager dialog box appears. 5.In this dialog box, select the three curves defined above; and click Plot. The X-Y plot is shown in the viewport. As before, customize the plot appearance to obtain a plot similar to Figure 6-37 . 6.6 Example: crushing of a tube You have been asked to analyze the crushing of a square, steel box tube between two rigid plates, as shown in Figure 6-38 . T a 500 kg mass at its other end. Both the tube and the rigid plate with mass have an initial velocity of 8.9408 m/s (20 mph) just b During impact, the tube dissipates a large amount of the initial kinetic energy into plastic deformation, and your primary interest important to model the self-contact in the tube correctly as it folds upon itself while buckling. An application for such an analysi automobile frame. Figure 6-38 Problem description: square, steel box tube crushed between two plates. 6.6.1 Mesh design Figure 6-39 shows the suggested mesh for this simulation. Figure 6-39 Box tube mesh. Shell elements are appropriate because the thickness (0.001 m) is 1/100th of the smallest planar dimension (0.1 m). While the in the entire structure because it is not known whether the lowest buckling modes are symmetric or unsymmetric. Because the tube mesh with the lowest buckling modes of the tube to obtain a smooth postbuckling response. You will obtain the first 10 buckling using ABAQUS/Standard and storing these modes in the results (.fil) file. You will then use the *IMPERFECTION option in ABAQU to perturb the nodal coordinates. 6.6.2 Node and element sets All of the necessary node and element sets are shown in Figure 6-40 . Figure 6-40 Necessary node and element sets. Place the shell elements in an element set called TUBE. Place the top and bottom rigid elements into their own element sets: RIG called MASS. Place the nodes on the top end of the tube (largest 1-coordinate) in a node set called TOP, and place the nodes set called BOT. You will use these node sets later when defining interactions with the rigid plates. Create a node set called ALL c when defining the initial velocity of the tube. 6.6.3 Preprocessing--creating the model Use your preprocessor to generate the mesh shown in Figure 6-39 , which is 8 elements wide and 32 elements long. Use S4R e plates. Each of the shell elements is 0.0125 m ?0.0125 m, and the rigid elements are 0.2 m ?0.2 m. Define the shell elements su a mass of 500 kg at the rigid body reference node of the bottom rigid body. In the ABAQUS/Explicit analysis you will give the ma cause the tube to crush against the top rigid plate. This mass element is not necessary for the ABAQUS/Standard analysis, whic input files. 6.6.4 Reviewing the input file for the ABAQUS/Standard buckling analysis--the model data Before creating the input file for the ABAQUS/Explicit analysis, you must create the input file for the ABAQUS/Standard buckling possible the same mesh and boundary conditions as the crushing analysis. The following would be a suitable description in the *HEADING Tube crush (x-direction) -- buckling analysis Tube dimensions: 0.4m x 0.1m x 0.1m SI units (kg, m, s, N) Section properties All of the shell elements in the model have the same material and the same 0.001 m (1 mm) thickness. For efficiency use the de integration points through the shell thickness. While more integration points increases the accuracy of the solution, it also increa elements: *SHELL SECTION, ELSET=TUBE, MATERIAL=STEEL 0.001, 3

The rigid elements also require section properties. Use the following section properties for the two rigid bodies: *RIGID BODY, ELSET=RIGTOP, REF NODE=<top reference node number> *RIGID BODY, ELSET=RIGBOT, REF NODE=<bottom reference node number> The mass element also requires section properties. Use the following section property to define a 500 kg mass attached to the b *MASS, ELSET=MASS 500., Material properties The tube is made of steel, which will be modeled as an isotropic, elastic-plastic material. Plasticity data and density are not nec material properties. However, include the data anyway for consistency with the ABAQUS/Explicit input file. ABAQUS/Standard ig material option block to define the material behavior: *MATERIAL, NAME=STEEL *ELASTIC 207.E9, .3 *PLASTIC 1.587E8, 0.0 1.631E8, 0.015 1.863E8, 0.033 1.932E8, 0.044 2.020E8, 0.062 2.070E8, 1.500 *DENSITY 7800., Initial conditions There are no initial conditions because you do not need to include the initial velocity in the buckling analysis. Boundary conditions Constrain the top rigid body completely by constraining its reference node. *BOUNDARY REFTOP, 1, 6 Constrain the bottom rigid body in all but the 1-degree of freedom by constraining its reference node. REFBOT, 2, 6 Contact definitions Define surfaces on the top and bottom rigid elements. The surfaces must be on the side of the elements facing the tube, which defined the rigid elements. *SURFACE, NAME=TOPSU RIGTOP, <SPOS or SNEG> *SURFACE, NAME=BOTSU RIGBOT, <SPOS or SNEG> Create a node-based surface using the nodes at the top of the tube (contained in node set TOP). *SURFACE, TYPE=NODE, NAME=TOP TOP, Define contact between the nodes at the top of the tube and the top rigid surface. In a linear perturbation step, such as *BUCKL unchanged throughout the step. Therefore, if you ensure that the top nodes of the tube are in contact with the top rigid surface throughout the buckling analysis. To ensure the closed contact condition, use the ADJUST parameter, which moves nodes within surface. *CONTACT PAIR, ADJUST=0.01, INTERACTION=TUBE TOP, TOPSU A contact pair must refer to a *SURFACE INTERACTION option, which may contain additional information regarding the contact friction between the tube and the top rigid surface. The *SURFACE INTERACTION option block contains only the following keywo *SURFACE INTERACTION, NAME=TUBE Create a node-based surface containing the nodes at the bottom of the tube. *SURFACE, TYPE=NODE, NAME=BOT BOT, Tie constraints Use the *TIE option to ``glue'' the slave nodes on the bottom of the tube to the bottom rigid surface. To ensure that the slave n parameter equal to YES and the POSITION TOLERANCE parameter to 0.01. Slave nodes of the tube that are within this tolerance

*TIE, NAME=FIXEDBOTTOM, POSITION TOLERANCE=0.01, ADJUST=YES BOT, BOTSU 6.6.5 Reviewing the input file for the ABAQUS/Standard buckling analysis--the history data Step definition To obtain the first 10 buckling modes, use the following step definition: *STEP Buckling analysis of the tube *BUCKLE, EIGENSOLVER=SUBSPACE 10, The subspace iteration eigensolver is used since the Lanczos eigensolver is not suitable for solving a buckling problem with con Loading Load the tube in a manner similar to the actual loading that the tube will experience during the crushing analysis. The actual load the buckling loads as a fraction of the applied load. Apply a concentrated force of 500 N in the positive 1-direction to the bottom *CLOAD REFBOT, 1, 500.0 Output requests You must request results (.fil) file output for the nodal displacements so that you can use the buckling analysis results to seed th preselected field variable output to the output database (.odb) file and prevent unnecessary results from being printed in the dat *NODE FILE U, *OUTPUT, FIELD, VARIABLE=PRESELECT, FREQUENCY=1 *NODE PRINT, FREQUENCY=0 *EL PRINT, FREQUENCY=0 *END STEP 6.6.6 Running the ABAQUS/Standard buckling analysis After storing your input in a file called tube_buckle.inp, run the analysis using the following command: abaqus job=tube_buckle If your analysis does not complete, check tube_buckle.dat for error messages. Modify your input to remove any errors. 6.6.7 Results of the buckling analysis After running the buckling analysis to completion, look at the eigenvalue output in the data file. The output shows 10 eigenmode buckling load, multiply the eigenvalue by the applied load; e.g., the buckling load for the first mode is 35.377 ?500.0 N=17689 N EIGENVALUE OUTPUT BUCKLING LOAD ESTIMATE = ("DEAD" LOADS) + EIGENVALUE * ("LIVE" LOADS). "DEAD" LOADS = TOTAL LOAD BEFORE *BUCKLE STEP. "LIVE" LOADS = INCREMENTAL LOAD IN *BUCKLE STEP MODE NO EIGENVALUE

1 35.377 2 47.316 3 47.316 4 60.433 5 64.596 6 66.441 7 76.665 8 78.725 9 87.067 10 87.069 Postprocessing with ABAQUS/Viewer Start ABAQUS/Viewer by entering the command abaqus viewer odb=tube_buckle at the operating system prompt. View the deformed shape of the eigenmodes, and determine whether the deformations in the buckling analysis will provide good ABAQUS/Explicit.

To plot the eigenmodes: 1.From the main menu bar, select Tools->Display Group->Create. 2.In the dialog box that appears, select element set PART-1-1.TUBE and click Replace. 3.Click Dismiss. Observe that the tube alone is now displayed. 4.From the main menu bar, select Result->Step/Frame. 5.To select the first eigenmode, choose Mode 1 in the dialog box that appears, and click OK. 6.Change the plot mode to deformed by selecting Plot->Deformed Shape or by clicking the tool in the toolbox. The first eigenmode is displayed in the viewport. 7.To change the view, click the rotate tool and drag the cursor to rotate the tube in the viewport. 8.Click Deformed Shape Plot Options in the prompt area. The Deformed Shape Plot Options dialog box appears. 9.In this dialog box, choose the Hidden render style. Click OK. The eigenmode is displayed with the changed deformed plot options. 10.Repeat Step 5 for the second eigenmode. Eigenmodes 1 and 2, which are displayed in Figure 6-41 , show that the eigenmodes are similar to the expected deformation d modes shows that they are relevant as well. Figure 6-41 Eigenmode 1 (left) and eigenmode 2 (right). Now that you have obtained the buckling modes using ABAQUS/Standard, you are ready to create the ABAQUS/Explicit input file 6.6.8 Reviewing the input file for the ABAQUS/Explicit crushing analysis--the model data Most of the model data are the same for the ABAQUS/Standard buckling analysis and the ABAQUS/Explicit crushing analysis. Sp are all the same. Also, the tie constraint is the same. One significant difference in the contact definitions is that for ABAQUS/Sta whereas in ABAQUS/Explicit the contact pair definition is part of the history data. In addition, the ABAQUS/Explicit input file con as well as an option to read the imperfections from the buckling analysis and to alter the mesh accordingly. The ABAQUS/Explic with the top rigid surface. Copy the buckling input file, tube_buckle.inp, to a file called tube_crush.inp, and then modify tube_cru Use the buckling eigenmodes to perturb the mesh When choosing perturbation magnitudes, the goal is to seed the mesh with a deformation pattern that will allow the postbuckling the perturbations used for the different eigenmodes are a few percent of the relevant structural dimension, such as shell thickne analysis, the magnitude of the perturbation associated with those modes should be greatest. Increasing the imperfection factors hand, imperfections that are too large are unrealistic. The postbuckling response of a structure with many closely spaced eigenvalues may be highly sensitive to the mesh imperfectio imperfections can cause significant changes in postbuckling behavior, requiring a sensitivity study to determine realistic mesh im significantly lower than the second, so we can safely assume that the first eigenmode will dominate. Only modes 2 and 3 have c the lowest eigenvalue that they will not dominate the postbuckling response. A structure such as a short, thin-walled cylinder ma a structure it is not clear whether the lowest eigenmode will dominate. Introduce mesh imperfections that are a maximum of 2% of the shell thickness. Since ABAQUS/Standard scales the eigenvalue o deformation in each eigenmode is 1.0 m, choosing an imperfection factor of 1.0 would perturb the mesh by exactly the displace mesh imperfections for each eigenmode that are a maximum of 2% of the shell thickness (0.001 m ?0.02 = 2.0 ?10-5 m) for the number increases, use the following imperfection factors: *IMPERFECTION, FILE=tube_buckle, STEP=1 1, 2.0E-5 2, 0.8E-5 3, 0.4E-5 4, 0.18E-5 5, 0.16E-5 6, 0.10E-5 7, 0.10E-5 8, 0.08E-5 9, 0.02E-5 10, 0.02E-5 Initial conditions The top rigid body is fixed. The tube and bottom rigid body, including the mass, have an initial velocity of 20 mph, or 8.9408 m/ *INITIAL CONDITIONS, TYPE=VELOCITY ** 1 mph = 0.44701 m/s ** 20 mph = 8.9408 m/s

REFBOT, 1, 8.9408 ALL, 1, 8.9408 Boundary conditions The boundary conditions are exactly the same as those used for the buckling analysis. Output node set Define a node set to use later when requesting output. *NSET, NSET=PRIN TOP, REFTOP, BOT, REFBOT Surface definitions We will define contact somewhat differently for the crushing analysis than for the buckling analysis, using the self-contact and d crushes, it buckles repeatedly so that many regions on the inside and on the outside of the tube contact each other. Since we d with each other, we must allow contact to occur in a very general manner so that any regions can contact any other regions, bo allows the generality of a single body, such as the tube, contacting itself in any regions of the contact surface. Delete the surface TOP, the contact pair between surfaces TOP and TOPSU, and the surface interaction TUBE. Define the entire sided contact surface by defining a shell surface on the element set, TUBE, without specifying a face identifier. *SURFACE, NAME=TUBE TUBE,

6.6.9 Reviewing the input file for the ABAQUS/Explicit crushing analysis--the history data We now discuss the history data required for this problem, including the step definition, contact definitions, and output requests Step definition Run the explicit dynamics analysis for a total of 0.03 seconds. *STEP Impact of square tube with free deceleration Initial velocity=8.9408m/s (20mph) *DYNAMIC, EXPLICIT , 0.03 Contact definitions To specify self-contact for the inside and outside of the tube, define a contact pair with the double-sided surface, TUBE, as the *CONTACT PAIR, INTERACTION=SELF TUBE, In addition, define the surface interaction, SELF, with a friction coefficient of 0.1. *SURFACE INTERACTION, NAME=SELF *FRICTION 0.1, Initially, the contact will occur only at the nodes along the top edge of the tube; but as the tube crushes, more of the tube surfa situation by specifying contact between the double-sided surface, TUBE, and the top rigid plate surface, TOPSU. Since a doubl definition, the contact pair is defined as follows: *CONTACT PAIR, INTERACTION=IMPACT TUBE, TOPSU In addition, define the surface interaction, IMPACT, with zero friction. *SURFACE INTERACTION, NAME=IMPACT Output requests Since the primary goal of this analysis is to determine the amount of energy absorbed by the tube as it is crushed, the most imp history output to the output database: *OUTPUT, HISTORY, TIME INTERVAL=0.00008 *ENERGY OUTPUT ALLWK, ALLKE, ETOTAL, ALLPD, ALLIE, ALLVD, ALLAE, ALLSE Nodal histories and a monitor node, as well as output of the preselected field variables to the output database, are also useful in *NODE OUTPUT, NSET=PRIN U, V, A, RF1, RF2, RF3 *MONITOR, NODE=<bottom rigid reference node number>, DOF=1 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=20 *END STEP 6.6.10 Running the ABAQUS/Explicit crushing analysis

After storing your input in a file called tube_crush.inp, run the analysis using the following command: abaqus job=tube_crush analysis As with the circuit board drop test, this analysis will take a while to complete. If your analysis does not complete, check the data messages. Modify your input to remove any errors. 6.6.11 Results of the crushing analysis After running the simulation successfully, visually display the deformation history of the tube with an animation using the same p animation will show the deformed mesh for each of the 20 states stored in the output database file. Viewing the deformation his be deforming correctly as it is crushed. The final deformed mesh is shown in Figure 6-42 . As you view the displaced mesh, no the contact calculations. The regions that are in contact appear with a slight gap between contacting regions. Figure 6-42 Final deformed mesh. To underscore the importance of perturbing the mesh prior to performing a postbuckling analysis, Figure 6-43 shows the final imperfections. Figure 6-43 Deformed mesh when no initial imperfections are imposed. In contrast to the smoothly buckled shape of the perturbed mesh, this unperturbed mesh deforms into sharp folds; the shape is introduced in the perturbed mesh are enough to make the postbuckling behavior proceed smoothly. Status file The status file issues several warnings regarding high out-of-plane warping in some surfaces. Such warnings are not surprising The nature of this crushing analysis is such that a highly refined mesh would be required to eliminate the large warping of some can simply accept the results as an approximate solution. STEP 1 ORIGIN 0.00000E+00 Total memory used for step 1 is approximately 1.6 megawords Global time estimation algorithm will be used. Scaling factor : 1.0000 STEP TOTAL CPU STABLE CRITICAL KINETIC INCREMENT TIME TIME TIME INCREMENT ELEMENT ENERGY MONITOR 0 0.000E+00 0.000E+00 00:00:02 1.392E-06 101 2.003E+04 0.000E+00 Results number 0 at increment zero. ODB Field Frame Number 0 of 20 requested intervals at increment zero.

***WARNING: Master surface TUBE of contact pair # 1 contains facets with out-of-plane warping of at least 20.000 degrees in increment 320. Large warping that develops during an analysis often corresponds to severe distortion of the underlying elements. It may be appropriate to rerun the analysis with a refined mesh. Postprocessing with ABAQUS/Viewer Start ABAQUS/Viewer by entering the command abaqus viewer odb=tube_crush at the operating system prompt. One way to determine the energy absorption of the tube is to look at a history plot of the pertinent energies. Since most of the e history of plastic dissipation, ALLPD. To show how the total kinetic energy of the model changes as the tube deforms, also plot energy, ALLAE, as an indication of the mesh quality. To plot the energy histories: 1.From the main menu bar, select Result->History Output. The History Output dialog box appears. 2.Select the artificial energy, ALLAE, and save the data as ALLAE. 3.Select the kinetic energy, ALLKE, and save the data as ALLKE. 4.Select the plastic dissipation, ALLPD, and save the data as ALLPD. 5.Click Dismiss to close the dialog box. 6.From the main menu bar, select Tools->XY Data->Manager. The XY Data Manager dialog box appears. 7.In this dialog box, select the three curves defined above and click Plot. Customize the appearance of the plot by changing the curve styles and the axis titles, as shown in Figure 6-44 .

Figure 6-44 History plot of ALLPD and ALLKE. By the end of the analysis 3600 J of energy has been dissipated as plastic deformation, and there is a corresponding decrease strain energy at the end of the analysis is 800 J, or 18% of the plastic dissipation. Ideally the artificial strain energy should be a m dissipation. (In previous examples we introduced 5% as a general rule.) The high percentage in this analysis indicates that the m deformations by the end of the analysis are so severe, a highly refined mesh would be required to minimize the effect of hourgla can accept the results as an approximate solution. 6.7 Summary ontact is defined in the history definition part of the input file. he contact definition in ABAQUS/Explicit has two parts: definition of the surface that may come into contact and definition of urfaces are defined using the *SURFACE option. Individual nodes can be included in a contact pair by using the TYPE=NODE assigned to a rigid body by using the ANALYTICAL SURFACE parameter on the *RIGID BODY option. The parameters TYPE=SEG specify the type of analytical rigid surface. ontact can be defined between two surfaces or between a single surface and itself. ll surfaces must be continuous and simply connected. A shell or membrane surface may span both sides of the element. urface normals for shells, membranes, or rigid elements must be consistent from one element to the next within a surface if it surface is traversed. BAQUS/Explicit does not smooth rigid surfaces; they are faceted like the underlying elements. Coarse meshing of discrete rigi ied contact is a useful means of mesh refinement. BAQUS/Explicit adjusts the nodal coordinates without strain to remove any initial overclosures prior to the first step. If the adju elements can become severely distorted. n subsequent steps any nodal adjustments to remove initial overclosures induce strains that can potentially cause severe mes 7. Quasi-Static Analysis The explicit solution method is a true dynamic procedure originally developed to model high-speed impact events in which inert are propagated as stress waves between neighboring elements while solving for a state of dynamic equilibrium. Since the minim require a large number of increments. The explicit solution method has proven valuable in solving static problems as well--ABAQUS/Explicit solves certain types of sta advantage of the explicit procedure over the implicit procedure is the greater ease with which it resolves complicated contact pr procedure requires less system resources than the implicit procedure. Refer to Chapter 3, "Overview of Explicit Dynamics," for a Applying the explicit dynamic procedure to quasi-static problems requires some special considerations. Since a static solution i impractical to analyze the simulation in its natural time scale, which would require an excessive number of small time increments accelerated in some way. The problem is that as the event is accelerated, the state of static equilibrium evolves into a state of d dominant. The goal is to model the process in the shortest time period in which inertial forces remain insignificant. 7.1 Analogy for explicit dynamics To provide you with a more intuitive understanding of the differences between a slow, quasi-static loading case and a rapid loa Figure 7-1 Analogy for slow and fast loading cases. The figure shows two cases of an elevator full of passengers. In the slow case the door opens and you walk in. To make room, who push their neighbors, and so on. This disturbance passes through the elevator until the people next to the walls indicate tha until everyone has reached a new equilibrium position. If you increase your speed slightly, you will shove your neighbors more fo same position as in the slow case. In the fast case the door opens and you run into the elevator at high speed, permitting the occupants no time to rearrange them directly in front of the door, while the other occupants will be unaffected. The same thinking is true for quasi-static analyses. The speed of the analysis often can be increased substantially without sever result of the slow case and a somewhat accelerated case are nearly the same. However, if the analysis speed is increased to a localize, and the results are quite different from the quasi-static solution. 7.2 Loading rates The actual time taken for a physical process is called its natural time. Generally, it is safe to assume that performing an analysis accurate static results. After all, if the real-life event actually occurs in a natural time scale in which velocities are zero at the co that the analysis has, in fact, achieved a steady state. You can increase the loading rate so that the same physical event occurs the true static solution and dynamic effects remain insignificant. 7.2.1 Smooth amplitude curves For accuracy and efficiency quasi-static analyses require the application of loading that is as smooth as possible. Sudden, jerky inaccurate solutions. Applying the load in the smoothest possible manner requires that the acceleration changes only a small am smooth, it follows that the changes in velocity and displacement are also smooth. ABAQUS has a simple, built-in type of amplitude called SMOOTH STEP that automatically creates a smooth loading amplitude. W DEFINITION=SMOOTH STEP, ABAQUS/Explicit automatically connects each of your data pairs with curves whose first and secon

your data points. Since both of these derivatives are smooth, you can apply a displacement loading with SMOOTH STEP using o be smooth. Using this type of loading amplitude allows you to perform a quasi-static analysis without generating waves due to d following amplitude definition ABAQUS/Explicit creates the amplitude curve shown in Figure 7-2 : *AMPLITUDE, DEFINITION=SMOOTH STEP 0.0, 0.0, 1.0E-5, 1.0

Figure 7-2 Amplitude definition using *AMPLITUDE, DEFINITION=SMOOTH STEP. 7.2.2 Structural problems In a static analysis the lowest mode of the structure usually dominates the response. Knowing the frequency and, corresponding required to obtain the proper static response. To illustrate the problem of determining the proper loading rate, consider the defo as shown in Figure 7-3 . The actual test is quasi-static. Figure 7-3 Rigid cylinder impacting beam. The response of the beam varies greatly with the loading rate. At an extremely high impact velocity of 400 m/s, the deformation obtain a better quasi-static solution, consider the lowest mode. Figure 7-4 Impact velocity of 400 m/s. The frequency of the lowest mode is approximately 250 Hz, which corresponds to a period of 4 milliseconds. The natural freque in ABAQUS/Standard. To deform the beam by the desired 0.2 m in 4 milliseconds, the velocity of the cylinder is 50 m/s. While 5 become secondary to the overall stiffness of the structure, and the deformed shape--shown in Figure 7-5 --indicates a much Figure 7-5 Impact velocity of 50 m/s. While the overall structural response appears to be what we expect as a quasi-static solution, it is usually desirable to increase certain that the solution is truly quasi-static. To improve the results even further, the velocity of the rigid cylinder could be rampe definition--thereby easing the initial impact. 7.2.3 Metal forming problems Artificially increasing the speed of forming events is necessary to obtain an economical solution, but how large a speed-up can deformation of the sheet metal blank corresponds to the deformed shape of the lowest mode, the time period of the lowest stru However, in forming processes the rigid dies and punches can constrain the punch in such a way that its deformation may not r recommendation is to limit punch speeds to less than 1% of the sheet metal wave speed. For typical processes the punch spee approximately 5000 m/s. This recommendation, therefore, suggests a factor of 50 as an upper bound on the speed-up of the p The suggested approach to determining an acceptable punch velocity involves running a series of analyses at various punch spe order of fastest to slowest since the solution time is inversely proportional to the punch velocity. Examine the results of the analy strains vary with punch speed. Some indications of excessive punch speeds are unrealistic, localized stretching and thinning as speed of, for example, 50 m/s, and decrease it from there, at some point the solutions will become similar from one punch spee a steady-state solution. As inertial effects become less significant, differences in simulation results also become less significant As the loading rate is increased artificially, it becomes more and more important to apply the loads in a gradual and smooth man impose a constant velocity throughout the forming step. Such a loading causes a sudden impact load onto the sheet metal blan through the blank and may produce undesired results. The effect of any impact load on the results becomes more pronounced a from zero using a SMOOTH STEP amplitude minimizes these adverse effects. Springback Springback is often an important part of a forming analysis because the springback analysis determines the shape of the final, u simulations, springback poses some special difficulties. The main problem with performing springback simulations within ABAQU solution. Typically, the loads must be removed very carefully, and damping must be introduced to make the solution time reason and ABAQUS/Standard allows a much more efficient approach. Since springback involves no contact and usually includes only mild nonlinearities, ABAQUS/Standard can solve springback prob preferred approach to springback analyses is to import the completed forming model from ABAQUS/Explicit into ABAQUS/Stand import the forming results and perform the springback analysis. Using the *IMPORT option within the ABAQUS/Standard input fil the entire deformable mesh is imported. The nodes, elements, and section properties are imported automatically, but you must r springback analysis is complete, you can import the model back into ABAQUS/Explicit to continue with another forming stage. 7.3 Mass scaling Mass scaling enables an analysis to be performed economically without artificially increasing the loading rate. Mass scaling is th involving a rate-dependent material or rate-dependent damping, such as dashpots. In such simulations increasing the loading r same factor as the loading rate. When the properties of the model change with the strain rate, artificially increasing the loading r The following equations show how the stable time increment is related to the material density. As discussed in Chapter 3, "Overv minimum stable time increment of all elements. It can be expressed as where is the characteristic element length and is the dilatational wave speed of the material. The dilatational wave speed for a where is the material density.

According to the above equations, artificially increasing the material density, , by a factor of decreases the wave speed by a fac Remember that when the global stability limit is increased, fewer increments are required to perform the same analysis, which is the same influence on inertial effects as artificially increasing the loading rate. Therefore, excessive mass scaling, just like exces suggested approach to determining an acceptable mass scaling factor, then, is similar to the approach to determining an accep approach is that the speedup associated with mass scaling is the square root of the mass scaling factor, whereas the speedup rate scaling factor. For example, a mass scaling factor of 100 corresponds exactly to a loading rate scaling factor of 10. There are several ways to implement mass scaling in your model using the *FIXED MASS SCALING or *VARIABLE MASS SCALING is part of the history definition, it can be changed from step to step, allowing great flexibility. Refer to the ABAQUS/Explicit User 7.4 Energy balance The most general means of evaluating whether or not a simulation is producing an appropriate quasi-static response involves st balance equation in ABAQUS/Explicit: where is the internal energy (both elastic and plastic strain energy), is the energy absorbed by viscous dissipation, is the kinet work of external forces, and is the total energy in the system. To illustrate energy balance with a simple example, consider the uniaxial tensile test shown in Figure 7-6 . Figure 7-6 Uniaxial tensile test. The energy history for the quasi-static test would appear as shown in Figure 7-7 . Figure 7-7 Energy history for quasi-static tensile test. If a simulation is quasi-static, the work applied by the external forces is nearly equal to the internal energy of the system. The vi materials, discrete dashpots, or material damping are used. We have already established that the inertial forces are negligible in model is very small. The corollary to both of these conditions is that the kinetic energy is also small. As a general rule the kinetic fraction (typically 5% to 10%) of its internal energy throughout most of the process. When comparing the energies, remember that ABAQUS/Explicit reports a global energy balance, which includes the kinetic energ bodies are of interest when evaluating the results, the kinetic energy of the rigid bodies should be subtracted from when evalua For example, if you are simulating a transport problem with rolling rigid dies, the kinetic energy of the rigid bodies may be a sign cases you must subtract the kinetic energy associated with rigid body motions before a meaningful comparison with internal ene 7.5 Example: deep drawing of a can bottom You have been asked to simulate the deep drawing of a can bottom. Analyze this two-stage metal forming process as a quasi-s compared to stiffness effects. Figure 7-8 , Figure 7-9 , and Figure 7-10 provide the detailed model description necessary fo can use axisymmetric shell elements to model the blank and axisymmetric rigid surfaces to model the tools. The consequence o dimensional shell elements is that an axisymmetric simulation cannot predict wrinkling in the circumferential direction if there are Figure 7-8 Detailed model description of forming stage 1. (All dimensions are in meters.) Figure 7-9 Detailed forming description of forming stage 2. (All dimensions are in meters.) Figure 7-10 Material properties for the blank. Creating the final desired shape of the can bottom requires two separate forming stages. In the first forming stage a thin disk of stationary die into the shape of a cup. Once the cup has been formed, the blank is removed from the first set of tools so that it annealed to relieve the cold-working plastic strains generated during the first forming stage. After annealing, the blank is placed into a second set of rigid tools, and it is drawn another 0.010 m into the final shape of the b spring and dashpot; thus, it is allowed to move in the axial, or 2-, direction. As the punch moves downward, it pinches the blank As the blank moves downward, it contacts the fixed "domer" die, which forms the dome shape in the can bottom. After the seco spring back to its final, unloaded configuration. 7.5.1 Coordinate system The orientation of the model in the global coordinate system is shown in Figure 7-8 and Figure 7-9 . Since the physical proce and 3-directions correspond to radial, axial, and circumferential directions, respectively. 7.5.2 Mesh design The blank Since the blank is initially a flat disk, model it with axisymmetric shell elements (SAX1) as a straight line starting at the symmetry how many elements you should use. Examine the forming processes, and consider the deformations the blank will experience du the features of the rigid forms accurately. When using a uniform mesh--as we are in this example--the element size must be small enough to capture the smallest feature forming stage 2. The general rule that we will use is that there should be 10 elements to model that radius, making the maximum which is a disk with a radius of 0.045 m, the element length is 0.0009 m, which is adequate. Both dies and punches are compos there are no sharp bends, there is no need for a terribly fine mesh; no area of the blank is subjected to severe crimping. Assume The rigid dies and punches Model the rigid dies and punches with analytical rigid surfaces in forming stage 1 and axisymmetric rigid elements (RAX2) in form representation of the surface as opposed to the piecewise linear representation offered by rigid elements. In this example, it wo

forming stage with analytical rigid surfaces. However, most three-dimensional dies and punches require the generality of a rigid mesh the rigid surfaces finely enough so that they are accurate representations of the actual rigid tools. Mesh each straight segm elements. The rigid surfaces should be as close as possible to the blank without any overclosure at the start of the forming step. Rememb when determining contact; therefore, the midplane of the blank must be at least one-half of the shell thickness away from the ri ABAQUS/Explicit will attempt to resolve overclosures in a strain-free manner, and the resulting nodal location changes may disto the initial shape of the blank if the overclosures are great enough. 7.5.3 Node and element sets Figure 7-11 shows all of the sets necessary to apply the element properties, loads, and boundary conditions, as well as to req an element set called BLANK. Create separate node sets for each of the rigid body reference nodes. Give the name REFP1 to th the name REFD1 to the node set containing the reference node for die 1. Create a node set called BSYM containing the node of called BEND containing the node at the free end of the blank. Figure 7-11 Necessary node and element sets. In addition, create a node set called NHIST containing two nodes for postprocessing. These nodes are selected to provide a de free end of the blank. Studying the velocity history for typical points in the blank will help to design the loading definition and to node from the symmetry axis and the node at the free end to be in node set NHIST. The second forming stage requires the rigid tools PUNCH2, DIE2, and DOMER to be in element sets with the same names. The n REFP2, REFD2, and REFDOMER. Create element sets SPRING and DASHPOT containing the spring and dashpot elements, resp 7.5.4 Determining an appropriate step time ``Loading rates,'' Section 7.2, discusses the procedures for determining the appropriate step time for a quasi-static process. W duration if we know the lowest natural frequency, the fundamental frequency, of the blank. One way to obtain such information i analysis the punch deforms the blank into a shape similar to the lowest mode. Therefore, it is important that the time for the first if you wish to model structural, as opposed to localized, deformation. Create an ABAQUS/Standard input file called draw_freq_std.inp. The only elements necessary for the frequency analysis are the the forming analysis. The node and element sets are also the same. Create the mesh in your preprocessor based on the mesh c and element sets,'' Section 7.5.3. The following is an appropriate heading for the analysis: *HEADING Deep drawing of a can bottom -- frequency analysis ABAQUS/Standard SI units (kg, m, s, N) Element properties Give each element set appropriate section properties. Create shell section properties for the blank so that it refers to a material c integration, and has a thickness of 0.5 mm. *SHELL SECTION, ELSET=BLANK, MATERIAL=STEEL, SECTION INTEGRATION=GAUSS .0005, 5 Material properties The blank is modeled using only the elastic material model since a *FREQUENCY analysis is a linear perturbation step and, thus define the elastic and mass properties of the steel: *MATERIAL, NAME=STEEL *ELASTIC 210.0E9, 0.3 *DENSITY 7800., Boundary conditions Apply symmetry boundary conditions to node set BSYM, and fix node set BEND in the axial, or 2-, direction. *BOUNDARY BSYM, 1 BSYM, 6 BEND, 2 History definition The history definition defines the *FREQUENCY procedure type to extract the first eigenmode. Request *RESTART output and th database file, and suppress other output. *STEP

*FREQUENCY 1, *RESTART, WRITE *OUTPUT, FIELD *NODE OUTPUT, NSET=BLANK U, *EL PRINT, FREQUENCY=0 *NODE PRINT, FREQUENCY=0 *END STEP Run the *FREQUENCY analysis Save your frequency analysis input in a file named draw_freq_std.inp, and run the analysis using the following command: abaqus job=draw_freq_std Postprocessing When the analysis has completed, use ABAQUS/Viewer to plot the shape of mode 1. Start ABAQUS/Viewer by typing abaqus viewer odb=draw_freq_std at the operating system prompt. To plot the mode shape: 1.From the main menu bar, select Result->Step/Frame. The Step/Frame dialog box appears. Select mode 1, and click OK. 2.From the main menu bar, select Plot->Deformed Shape; or use the tool in the toolbox. ABAQUS/Viewer displays the deformed model shape for the first buckling mode. Superimpose the undeformed model shape on 3.From the main menu bar, select Options->Deformed Shape. The Deformed Shape Plot Options dialog box appears. 4.Toggle on Superimpose undeformed plot. 5.Click OK. ABAQUS/Viewer displays an image of the deformed model of the blank superimposed upon the undeformed model of the blank The frequency analysis shows that the blank has a fundamental frequency of 304 Hz, corresponding to a period of 0.0033 s. Fig know that the lower bound on the step time for the first forming stage is 0.0033 s. Figure 7-12 Displaced shape of mode 1 from the ABAQUS/Standard frequency analysis. 7.5.5 Reviewing the input file for the ABAQUS/Explicit forming stage 1 analysis--the model data The first forming input file, draw_stage1_attempt.inp, will model only the first forming stage; thus, it requires only the component blank as you used in the frequency extraction input file, draw_freq_std.inp. Create the additional mesh components provided in ` Section 7.5.3, using your preprocessor. In this section we will review your model definition, make any corrections, and include a Model description The following would be a suitable description for the simulation: *HEADING Deep drawing of a can bottom -- stage 1, attempt 1 SI units (kg, m, s, N) This description clearly labels the input data and indicates the units. Element properties Give each element set appropriate section properties. Create shell section properties for the blank so that it refers to a material c integration, and has a thickness of 0.5 mm. Five section points through the shell thickness should provide accurate results for a *SHELL SECTION, ELSET=BLANK, MATERIAL=STEEL, SECTION INTEGRATION=GAUSS .0005, 5 Material properties The blank, which is the only deformable body in this simulation, is modeled using an elastic-plastic material model. Use the follo *MATERIAL, NAME=STEEL *ELASTIC 210.0E9, 0.3 *DENSITY 7800., Include the plasticity data under the *PLASTIC option. *PLASTIC 0.912E8, 0.0

0.131E9, 0.159E-02 0.171E9, 0.649E-02 0.211E9, 0.177E-01 0.251E9, 0.395E-01 0.291E9, 0.776E-01 0.331E9, 0.139 0.391E9, 0.295 Figure 7-13 shows that the plasticity curve for this example is quite smooth. Figure 7-13 Stress versus plastic strain material curve. Because smoothness is desired for all aspects of a quasi-static problem, it is advantageous to have as smooth a plasticity curv changes only slightly from one increment to the next as plastic straining occurs, thus helping to produce a solution with less noi Surface definitions Use the *SURFACE option to define the analytical surfaces of the rigid tools. *SURFACE, TYPE=SEGMENTS, NAME=PUNCH1 START,0.032,0.03025 LINE,0.032,0.00625 CIRCL,0.026,0.00025,0.026,0.00625 LINE,0.,0.00025 *SURFACE, TYPE=SEGMENTS, NAME=DIE1 START,0.0326,-0.02825 LINE,0.0326,-0.00425 CIRCL,0.0366,-0.00025,0.0366,-0.00425 LINE,0.050,-0.00025 To bind an analytical surface to a rigid body, assign the name of the analytical surface to the ANALYTICAL SURFACE parameter specify the rigid body reference node that will control the motion of the rigid body. *RIGID BODY, ANALYTICAL SURFCE=PUNCH1, REF NODE=<reference node number> *RIGID BODY, ANALYTICAL SURFACE=DIE1, REF NODE=<reference node number> Define the surfaces on the top and bottom of the blank. The surfaces on the top and bottom of the blank must face PUNCH1 an *SURFACE, NAME=BLANK_TOP BLANK, <SPOS or SNEG> *SURFACE, NAME=BLANK_BOT BLANK, <SPOS or SNEG> Two surface definitions (one on the top and the other on the bottom of the blank) are required because double-sided surfaces c rigid elements. Initial fixed boundary conditions The initial fixed boundary conditions can be part of the model definition or history definition. Since the model is axisymmetric, yo die 1--from motion in the radial, or 1-, direction and from in-plane rotation (degree of freedom 6). In addition, constrain die 1 *BOUNDARY REFP1, 1 REFP1, 6 REFD1, 1, 6 BSYM, 1 BSYM, 6 7.5.6 Designing the history definition Several aspects of the loading history are not clearly defined from the model description. Therefore, the analysis will include som acceptable results and what does not. Unlike the other example problems in this manual, we will discuss this forming problem o step before continuing to the next step. 7.5.7 General approach to forming stage 1 The goal of the first forming stage is to form a cup with a depth of 0.015 m using punch 1 and die 1. We intentionally did not pro appropriate punch velocity for a quasi-static analysis is not necessarily the same as the actual punch velocity. Typical punch ve vary. To analyze the first forming stage successfully, we must consider the following questions: 1.What is an appropriate step time?

2.How should we move the punch? 3.How many increments and how much computer time will the analysis require? 4.What output should we request? Determining an appropriate step time In ``Determining an appropriate step time,'' Section 7.5.4, we performed a frequency analysis, which showed that the blank has period of 0.0033 s. This time period provides a lower bound on the step time for the first forming stage. Choosing the step time frequency, or 0.033 s, should ensure a quality quasi-static solution. This time period corresponds to a constant punch velocity o Moving the punch Even if the punch actually moves at a nearly constant velocity, it is desirable to use a different amplitude curve that allows the b loading amplitude to use, remember that smoothness is important in all aspects of a quasi-static analysis. The preferred approa distance in the desired amount of time. We will analyze the first forming stage using both approaches--a constant punch veloci SMOOTH STEP amplitude--and will compare the results. Computer time Before you run the forming analysis, you may wish to know how many increments the analysis will take and, consequently, how datacheck analysis to obtain the approximate value for the initial stable time increment, or you can estimate it using the relations increment, which in this case does not change much from increment to increment, you can determine how many increments are begins, you can get an idea of how much CPU time is required per increment and, consequently, how much CPU time the step r Using the relations stated in ``Mass scaling,'' Section 7.3, the stable time increment for this analysis is approximately 1 ?10-7 s. increments for a step time of 0.033 s. Output To help determine how closely the analysis approximates the quasi-static assumption, the various energy histories are useful. E strain energy. Another useful history plot is that of the velocities at typical nodes. Earlier you created the node set, NHIST, conta output gives you an idea of the severity of any velocity spikes in the solution. Write these energy and velocity histories as history thicknesses, and nodal displacements to the output database as field data for postprocessing with ABAQUS/Viewer. You should restart the analysis. For your first attempt request state information at 20 intervals. Once you are more confident with the first for node for the rigid surface PUNCH1 as the monitor node. 7.5.8 Overview of forming stage 1 attempts For the first forming stage we present a progression of three attempts to obtain an acceptable solution. In the first attempt we m severe oscillations of the first attempt, in the second attempt we smooth the punch velocity amplitude by setting the DEFINITION To further remove high-frequency oscillations, in the third attempt we introduce stiffness proportional damping. The results of th as a basis of comparison for analyses with varying degrees of mass scaling. If you do not wish to run analyses that do not prod but run only attempt 3. The history definition for each of the attempts contains the same contact definitions and output requests. The history definition b procedure; and a total step time of 10 times the fundamental natural frequency, or 0.033 s. *STEP *DYNAMIC, EXPLICIT , .033 Define contact between the top of the blank and the punch and between the bottom of the blank and the die. *CONTACT PAIR, INTERACTION=INTER PUNCH1, BLANK_TOP DIE1, BLANK_BOT Define a friction coefficient of 0.1 between all contacting surfaces. *SURFACE INTERACTION, NAME=INTER *FRICTION .1, Write the restart information at 10 evenly spaced intervals during the analysis by including the following line in your input file: *RESTART, WRITE, NUMBER INTERVAL=10 Request velocity and displacement history output for node set NHIST and energy history output every 1 ?10-5 s. *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V *ENERGY OUTPUT ALLKE,ALLIE,ALLWK,ALLVD,ALLAE,ALLPD,ETOTAL Request section force and shell thickness field output for element set BLANK and displacement output at 4 intervals during the s *OUTPUT, FIELD, NUMBER INTERVAL=4

*ELEMENT OUTPUT, ELSET=BLANK STH, SF *NODE OUTPUT U,RF *END STEP 7.5.9 Forming stage 1--attempt 1 Add the following amplitude to the model definition to simulate the punch moving at a constant velocity of 0.45 m/s: *AMPLITUDE, NAME=FORM1 0., 0., .033, 1. and add the following prescribed displacement boundary condition to the history definition: *BOUNDARY, TYPE=DISPLACEMENT, AMPLITUDE=FORM1 REFP1, 2, 2, -0.015 Save your input in a file named draw_stage1_attempt1.inp, and run the analysis. Thirty minutes or more may be required to run t Strategy for evaluating the results Before looking at the results that are ultimately of interest, such as stresses and deformed shapes, we need to determine whethe acceptable. One good approach is to compare the kinetic energy history to the internal energy history. In a metal forming analys punch 1 has no mass, the blank is the only body with mass; and the kinetic energy is solely due to the motion of the blank. To in the blank should be no greater than a few percent of the internal energy. For greater accuracy, especially when springback stres approach is very useful because it is general for all types of metal forming processes and does not require any intuitive understa be too complex to permit an intuitive feel for the results. While a good primary indication of the caliber of a quasi-static analysis, the ratio of kinetic energy to internal energy alone is no two energies independently to determine whether they are reasonable. This part of the evaluation takes on increased importance accurate springback stress solution is highly dependent on accurate plasticity results. Even if the kinetic energy is fairly small, if plasticity. Generally, we expect smooth loading to produce smooth results; if the loading is smooth but the energy results are os incapable of showing such behavior, you should also study the kinetic energy history itself to see whether it is smooth or noisy. If the kinetic energy does not indicate quasi-static behavior, it can be useful to look at velocity histories at some nodes to get a velocity histories can indicate which regions of the model are oscillating and causing the high kinetic energies. Evaluating the results for attempt 1 In ABAQUS/Viewer, select File->Open from the main menu bar; or use the tool from the toolbar to open the output database dr energy. To create an energy history plot: 1.From the main menu bar, select Plot->History Output. A history plot of the external work for the whole model appears. 2.From the main menu bar, select Result->History Output. The History Output dialog box appears. 3.From the list of available variables, select Kinetic energy: ALLKE for Whole Model. 4.Skipping frames between reads keeps any high frequency oscillation from becoming simply a blackened region on the graph. 5.Click Plot to create a history plot of ALLKE. A history plot of the kinetic energy for the whole model appears (see Figure 7-14 ). Figure 7-14 Kinetic energy history for forming stage 1, attempt 1. 6.Click Dismiss to close the dialog box. Similarly, create a history plot of the model internal energy, ALLIE (see Figure 7-15 ). Figure 7-15 Internal energy history for forming stage 1, attempt 1. Figure 7-14 shows the kinetic energy history, which is very noisy through the first half of the step before becoming smooth. Th the blank is an indication of the inadequacy of this analysis. In this analysis the punch velocity remains constant, while the kineti no mass--is far from constant. Figure 7-15 shows the internal energy history. Without anything further to which to compare the internal energy, it appears reas Comparing Figure 7-14 and Figure 7-15 shows that the kinetic energy is a small fraction (less than 1%) of the internal energy that kinetic energy must be small relative to internal energy has been satisfied, even for this severe loading case. Since the kinetic energy history is very noisy, it is useful to look at the velocity history of a node in the model to gain an understa of the first node in node set NHIST, which is a node in the central region of the blank (Figure 7-16 ). Figure 7-16 Velocity history of the first node in NHIST for forming stage 1, attempt 1. To create a history plot of the velocity: 1.From the main menu bar, select Result->History Output. 2.From the list of available variables in the History Output dialog box, select Spatial velocity: V2 at Node 521 in NSET NHIST.

3.Click Plot to create a history plot of V2. The velocity of the node oscillates between values as great as ?0 m/s, while the punch velocity remains constant. Such noisy be material includes plasticity. These results indicate that we must change the simulation in some way to obtain a smoother respon 7.5.10 Forming stage 1--attempt 2 For attempt 2 smooth the velocity amplitude of the punch. Instead of moving the punch with a constant velocity, use the SMOOT curves,'' Section 7.2.1, for an explanation of this amplitude definition. By specifying an amplitude of 0.0 at the beginning of the ABAQUS/Explicit creates an amplitude definition that is smooth in both its first and second derivatives. Therefore, using SMOOT velocity and acceleration are smooth. Change the amplitude definition to *AMPLITUDE, NAME=FORM1, DEFINITION=SMOOTH STEP 0., 0., .033, 1. Save your modified input in a file named draw_stage1_attempt2.inp, and run the analysis. Thirty minutes or more may be require Evaluating the results for attempt 2 The kinetic energy history shown in Figure 7-17 is still quite noisy, although somewhat different from that of attempt 1. The no the end of the analysis it again becomes noisy. Figure 7-17 Kinetic energy history for forming stage 1, attempt 2. The internal energy for attempt 2, shown in Figure 7-18 , is similar to that of attempt 1. The internal energy history has a few un Again, the ratio of kinetic energy to internal energy is quite small and appears to be acceptable. Figure 7-18 Internal energy history for forming stage 1, attempt 2. For many forming analyses a smooth loading velocity, such as that used in attempt 2, is adequate to produce a quasi-static for less well-behaved than many because of the lack of a blank holder. Figure 7-19 shows the velocity history of node set BEND, Figure 7-19 Velocity history of node set BEND for forming stage 1, attempt 2. The node oscillates greatly because the end of the blank is unconstrained. If this forming process had a blank holder, attempt 2 process lacks such a constraint, we must make additional changes to produce an acceptable quasi-static solution. 7.5.11 Forming stage 1--attempt 3 To decrease the high-frequency noise caused primarily by the oscillations of the blank's free end, add stiffness proportional dam smallest amount of damping possible to obtain the desired solution since increasing the stiffness damping decreases the stable (Another available type of material damping is mass proportional damping, which generally is used to decrease the low-frequen forming process in a viscous fluid. The problem with mass proportional damping in metal forming analyses is the difficulty in det adversely influenced the solution. Therefore, we discourage the use of mass proportional damping in forming analyses unless it To avoid a dramatic drop in the stable time increment, the stiffness proportional damping factor, , should be less than, or of the without damping. Since the initial stable time increment for attempt 2 reported in the status file is 1.05 ?10-7, use the following *DAMPING, BETA=1.E-7 The *DAMPING option should form part of the material definition of the problem. Save your modified input in a file, draw_stage1_attempt3.inp, and run the analysis. Evaluating the results for attempt 3 The kinetic energy history shown in Figure 7-20 is much smoother and corresponds to the punch velocity much better than the oscillation. The internal energy history, shown in Figure 7-21 , shows a smooth increase from zero up to the final value. Most of the interna small relative to the internal energy. These results appear to be acceptable. Figure 7-20 Kinetic energy history for forming stage 1, attempt 3. Figure 7-21 Internal energy history for forming stage 1, attempt 3. History plots of the other energy variables reveal that the energy absorbed by viscous dissipation is small relative to the internal damping included in this model has not adversely affected the results. In addition, the amount of energy used to control hourgla 7.5.12 Discussion of the three forming attempts Before continuing the forming process with the springback analysis, we will discuss what we have learned from the three attemp Our initial criteria for evaluating the acceptability of the results was that the kinetic energy should be small compared to the inter case, attempt 1, this condition seems to have been met adequately. For many forming analyses that are better constrained--for internal energy ratio small may be an adequate condition to obtain a good solution to the forming problem. However, in this sim effect on the solution, doubling the internal energy of the blank, as shown in Figure 7-22 . Without damping these oscillations, Figure 7-22 Comparison of internal energies for the three attempts of forming stage 1. The additional requirements--that the histories of kinetic energy and internal energy must be appropriate and reasonable--are v of evaluating the results. Enforcing these requirements in general for more complex forming processes may be difficult because of the forming process. Results of forming stage 1 Now that we are satisfied that the quasi-static solution for the first forming stage is adequate, study some of the other results o

force summation through the entire thickness of the element in the radial direction, or the local 1-direction. The axisymmetric m Figure 7-23 Contour plot of SF1 (section force in the radial direction) for forming stage 1, attempt 3 (swept model for visualizat To contour the section force: 1.First, restrict the display group to the elements representing the blank. From the main menu bar, select Tools->Display GroupThe Create Display Group dialog box appears. 2.Select the element set PART-1-1.BLANK, and click Replace to replace the current model display with the element set PART-1 Now, the element set PART-1-1.BLANK is the only member of the display group. 3.From the main menu bar, select Result->Field Output. The Field Output dialog box appears. 4.Select SF from the Output Variable list. 5.From the list of available components, select SF1. 6.Click OK to apply these settings. The Select Plot Mode dialog box appears. 7.Toggle on Contour, and click OK. To better visualize contours in axisymmetric models, sweep the shell elements to construct the equivalent three-dimensional vie fashion. 8.From the main menu bar, select View->ODB Display Options. The ODB Display Options dialog box appears. 9.Click the Sweep & Extrude tab. 10.In the General Sweep region, toggle on Sweep elements and set the sweep range from 0 to 90 degrees. 11.Click OK to apply these settings. The view shown in Figure 7-23 can be specified. 12.From the main menu bar, select View->Specify. The Specify View dialog box appears. 13.Using the Rotation Angles method, enter the X-, Y-, and Z- angles as 45, 0, 0. 14.Click OK. ABAQUS displays your model in the specified view. The plot shows that SF1 is nearly constant at a value of approximately 3.274 ?104 in the flat region near the symmetry line. We w because of the uniform stretching. Since the blank is formed in a smooth manner, we would expect SF1 to vary smoothly, as it d the free end of the blank. Figure 7-24 shows a contour plot of the shell thickness, STH. At the start of the analysis the blank had a uniform shell thicknes Figure 7-24 Contour plot of STH (shell thickness) for forming stage 1, and attempt 3 (swept model for visualization purposes on By the end of forming stage 1 the blank has thinned somewhat in the filleted region, which makes sense since this region has st in thickness because there was little drawing in that region. It is interesting that the shell actually became thicker in the vertical re end of the blank moved substantially toward the symmetry line during forming, resulting in compressive stresses in the hoop dire force in the hoop direction. If this force is strongly compressive, it is possible that a full three-dimensional solution would predic end of the blank is strongly compressive with a value of 2.05 ?105 N, indicating that there may be wrinkling near the free end. Figure 7-25 Contour plot of SF2 (section force in the hoop direction) for forming stage 1, attempt 3 (swept model for visualizat 7.5.13 Methods of speeding up the analysis Now that we have obtained an acceptable solution to the first forming stage, we can try to obtain similar acceptable results usin computer time to be run in their physical time scale because the actual time period of forming events is large by explicit dynamic often requires making changes to the input file to reduce the computer cost of the analysis. There are two ways to reduce the co 1.Artificially increase the punch velocity so that the same forming process occurs in a shorter step time. This method is called lo 2.Artificially increase the mass density of the elements so that the stability limit increases, allowing the analysis to take fewer inc Unless the model has rate-dependent materials or damping, these two methods effectively do the same thing. Our model has no damping due to the dashpot in the second forming stage. Since scaling the loading rate would change the velocities and, hence which will not affect the force in the dashpot. Determining acceptable mass scaling ``Loading rates,'' Section 7.2, and ``Metal forming problems,'' Section 7.2.3, discuss how to determine acceptable scaling of the static analysis. The goal is to model the process in the shortest time period in which inertial forces remain insignificant. There ar still obtaining a meaningful quasi-static solution. As discussed in ``Loading rates,'' Section 7.2, we can use the same methods to determine an appropriate mass scaling factor a factor. The difference between the two methods is that a load-rate scaling factor of f has the same effect as a mass scaling fac period of the fundamental frequency of the blank would be adequate to produce quasi-static results. By studying the model ene acceptable. This technique produced a punch velocity of approximately 0.5 m/s, which is typical of actual punch velocities. Now

compare the results against our unscaled solution to determine whether the scaled results are acceptable. We assume that scali objective is to use scaling to decrease the computer time and still produce acceptable results. Our goal is to determine what scaling values still produce acceptable results and at what point the scaled results become unacc scaling factors, we investigate a range of scaling factors on the stable time increment size from to 30; specifically, we choose factors of 10, 100, and 900, respectively. To apply mass scaling of 10, add the following option to the history definition, *FIXED MASS SCALING, FACTOR=10 and save the modified input in a file named draw_stage1_attempt3_sqrt10.inp; run draw_stage1_attempt3_10.inp and draw_stage 900, respectively. First we will look at the effect of mass scaling on the section forces and the displaced shape. We will then see whether the ener Evaluating the results with mass scaling One of the results of interest in this analysis is the radial section force, SF1. Since we have already seen the contour plot of SF1 can compare the results from each of the scaled analyses with the unscaled analysis results. These contour plots display the res these SF1 contour plots also provide the displaced shape of the mesh. Figure 7-26 shows SF1 for a speedup of (mass scalin scaling of 100), and Figure 7-28 shows SF1 for a speedup of 30 (mass scaling of 900). These plots indicate that a speedup o For a speedup of 10 the results are similar, but the section force at the symmetry axis drops noticeably. For a speedup of 30 the of speedup is clearly too high. Figure 7-26 Section force SF1 for speedup of (mass scaling of 10; swept model for visualization purposes only). Figure 7-27 Section force SF1 for speedup of 10 (mass scaling of 100; swept model for visualization purposes only). Figure 7-28 Section force SF1 for speedup of 30 (mass scaling of 900; swept model for visualization purposes only). Now that we have seen how mass scaling has affected some of the section force results, we will see whether the energy historie shows the histories of kinetic energy and internal energy for all of the scaled analyses. The internal energy is almost identical for zero for a speedup of to a maximum of approximately 7.0 for a speedup of 30. Figure 7-29 Kinetic and internal energy histories for speedups of , 10, and 30, corresponding to mass scaling factors of 10, 10 Discussion of speedup methods As the mass scaling increases, the solution time decreases. The quality of the results also decreases because dynamic effects b that improves the solution time without sacrificing the results. Clearly, a speedup of 30 is too great to produce quasi-static resu A smaller speedup, such as , does not alter the results significantly. These results would be adequate for most applications, incl of the results begins to diminish, while the general magnitudes and trends of the results remain intact. Correspondingly, the ratio results for this case would be adequate for many applications but not for accurate springback analysis. 7.5.14 Springback after forming stage 1 in ABAQUS/Standard While it is possible to perform springback analyses within ABAQUS/Explicit, ABAQUS/Standard is much more efficient at solving simulations without external loading or contact, ABAQUS/Standard can obtain a springback solution in just a few increments. Co time period that is long enough for the solution to reach a steady state. For efficiency ABAQUS has the capability to import resu ABAQUS/Standard, allowing us to perform forming analyses in ABAQUS/Explicit and springback analyses in ABAQUS/Standard. We will create a new input file that imports the results from the analysis with a speedup of (mass scaling of 10, draw_stage1_at Model definition The first option following *HEADING is the *IMPORT option, which reads the element definitions and the state from the correspo begin with the following: *HEADING Deep drawing of a can bottom -- first springback ABAQUS/Standard springback following draw_stage1_attempt3_sqrt10.inp SI units (kg, m, s, N) *IMPORT, STEP=1, INTERVAL=10, STATE=YES, UPDATE=NO BLANK, *IMPORT ELSET BLANK, *IMPORT NSET BSYM, NHIST Setting the STATE parameter equal to YES causes the state of the model--stresses, strains, etc.--to be imported. Setting the U displacements to be imported as well instead of being reset to zero. The data line following the *IMPORT option supplies the na imported. The data line following the *IMPORT ELSET option lists element set names to be imported; the *IMPORT NSET option History definition You must redefine the boundary conditions, which are not imported. Impose the same symmetry boundary conditions as were im

*BOUNDARY BSYM, 1 BSYM, 6 For output, write the final state of the model to the restart file. *RESTART, WRITE, FREQUENCY=999 The history definition is quite straightforward since no loads are applied. Use the following history definition: *STEP, NLGEOM *STATIC .1, 1. *BOUNDARY, FIXED BSYM, 2, 2 *OUTPUT, FIELD, FREQUENCY=10 *ELEMENT OUTPUT, ELSET=BLANK STH, SF *NODE OUTPUT U, *END STEP The NLGEOM parameter is used on the *STEP option since the ABAQUS/Explicit analysis used it (this is the default setting in AB parameter because we do not always know whether or not geometric nonlinearities will affect the results. To remove rigid body m blank at just a single point, such as node set BSYM, so that you impose no unnecessary constraints. It is important to use the F at its final position at the end of forming stage 1. If you were to omit the FIXED parameter, BSYM would return to its original loca from the previous stage or moving it back to its undeformed location will produce the same springback results; however, fixing B blank location through the stages of this simulation. Save your input in a file named draw_spring1_std.inp, and run the analysis using the following command: abaqus job=draw_spring1_std oldjob=draw_stage1_attempt3_sqrt10 Results of the springback analysis Figure 7-30 shows a contour plot of SF1 at the end of the springback analysis in ABAQUS/Standard. These results are necess In fact, springback results are highly sensitive to errors in the forming stage, more sensitive than the results of the forming stage Figure 7-30 Contour plot of SF1 following springback in ABAQUS/Standard (swept model for visualization purposes only). 7.5.15 Annealing and forming stage 2 Now that we are satisfied with the results for forming stage 1 with a speedup of (mass scaling of 10) and its corresponding spr stage 2 using ABAQUS/Explicit. We will create a new input file that imports the final springback condition from ABAQUS/Standar Model definition Import the model from the final increment of the springback analysis. Set the STATE parameter equal to YES so that the stress a parameter equal to NO so that the displacements are continuous. Begin your restart input file with the following *HEADING and * *HEADING Deep drawing of a can bottom -- stage 2 ABAQUS/Explicit annealing and forming stage 2 following draw_spring1_std.inp SI units (kg, m, s, N) *IMPORT, STATE=YES, UPDATE=NO BLANK, Since you are importing only the blank from the ABAQUS/Standard analysis, you must define the rigid tooling needed for forming Figure 7-11 . *RIGID BODY, ELSET=PUNCH2, REF NODE=<reference node number> *RIGID BODY, ELSET=DIE2, REF NODE=<reference node number> *RIGID BODY, ELSET=DOMER, REF NODE=<reference node number> Define the surfaces on the blank and the rigid tools. *SURFACE, NAME=BLANK_TOP BLANK, <SPOS or SNEG> *SURFACE, NAME=BLANK_BOT BLANK, <SPOS or SNEG> *SURFACE, NAME=PUNCH2 PUNCH2, <SPOS or SNEG> *SURFACE, NAME=DIE2

DIE2, <SPOS or SNEG> *SURFACE, NAME=DOMER DOMER, <SPOS or SNEG> You must also define the spring and dashpot on which die 2 sits. These are defined as a SPRINGA and a DASHPOTA element, e set REFD2) and one end attached to ground (node set SPEND). *ELEMENT, TYPE=SPRINGA, ELSET=SPRING <element number>,<node # of REFD2>,<node # of SPEND> *SPRING, ELSET=SPRING 2, 6.3E6, *ELEMENT, TYPE=DASHPOTA, ELSET=DASHPOT <element number>,<node # of REFD2>,<node # of SPEND> *DASHPOT, ELSET=DASHPOT 2, 1.E4, Define the 10 kg mass associated with die 2. *ELEMENT, TYPE=MASS, ELSET=MASS <element number>,<node # of REFD2> *MASS, ELSET=MASS 10., You must define any desired sets for postprocessing. *NSET, NSET=REF REFP2, REFD2, REFDOMER, SPEND Before proceeding, you must make certain that the rigid tools are positioned properly by defining a brief step that includes only occurs in this first step, the time period for the step can be very short. We have chosen 1 ?10-6 s as the step time. Include the input file: *AMPLITUDE, NAME=RIGID, DEFINITION=SMOOTH STEP 0., 0., 1.E-6, 1. In addition, define an amplitude for the forming stage. *AMPLITUDE, NAME=FORM2, DEFINITION=SMOOTH STEP 0., 0., .033, 1. History definition Since an *ANNEAL step cannot be the first step in an ABAQUS/Explicit import analysis, begin the history by moving the rigid too straightforward; the difficulty in this case is positioning the tools properly, given the unknown shape of the blank following spring blank is known: the blank has been drawn to a depth of 0.015 m, and it is clamped between two rigid tools whose positions are constrained by the rigid tools, and it deforms freely to minimize its internal strain energy. The changes in shape during springbac springback in ABAQUS/Viewer before positioning the rigid tools for the second forming stage. After the first step it is imperative that there be no overclosure of contact pairs because, if such overclosure exists, it is resolved velocity being imposed on an overclosed node, possibly distorting the associated elements and drastically increasing the kinetic bodies are far away from the blank at the start of the step because the analysis runs while no deformation occurs in the blank. Use Tools->Query from the main menu bar in ABAQUS/Viewer to determine the coordinates of some nodes on the blank at the e second forming stage as close as possible to the deformed blank without any overclosure. Even the slightest overclosure will ca thickness of the shell when positioning the rigid bodies, since ABAQUS/Explicit considers the current shell thickness when perfo that the curved parts of punch 2 and die 2 form concentric arcs with the blank in between. Punch 2 has a radius of 0.006 m, and of approximately 0.006 m after springback. Die 2 has a radius of 0.0065 m, and it contacts the outer radius of the blank, whichof approximately 0.0065 m. Include the *DIAGNOSTICS option with the CONTACT INITIAL OVERCLOSURE=DETAIL and DEFORM information about possible initial overclosures and resulting excessive deformation speeds. Use the amplitude definition named "RIGID" with the displacement boundary condition to move the rigid bodies into position. It i step since the final velocity in the positioning step is the initial velocity in the second forming stage. Use the following step to po *STEP *DYNAMIC, EXPLICIT , 1.E-6 *RESTART, WRITE, NUMBER INTERVAL=1 *BOUNDARY, TYPE=VELOCITY BSYM, 1, 2, 0.

BSYM, 6, 6, 0. REFDOMER, 1, 1, 0. REFDOMER, 6, 6, 0. REFD2, 1, 1, 0. REFD2, 6, 6, 0. SPEND, 1, 1, 0. REFP2, 1, 1, 0. REFP2, 6, 6, 0. *BOUNDARY, AMPLITUDE=RIGID REFDOMER, 2, 2, <user-determined value> REFD2, 2, 2, <user-determined value> SPEND, 2, 2, <user-determined value> REFP2, 2, 2, <user-determined value> *DIAGNOSTICS, CONTACT INITIAL OVERCLOSURE=DETAIL, DEFORMATION SPEED CHECK=DETAIL *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U, V *NODE OUTPUT, NSET=REF U, RF *ENERGY OUTPUT ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER INTERVAL=4 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U, *END STEP Physically, annealing is the process of heating a metal part to a high temperature to allow the microstructure to recrystallize, rem *ANNEAL procedure removes stresses and plastic strains from all deformable elements in the model without changing the shell t to be the same before and after annealing. The *ANNEAL step is quite simple because there are no loadings or output requests *STEP *ANNEAL *END STEP In the second forming stage the deformed blank is clamped between punch 2 and die 2, which sits on a spring and dashpot. As is forced to move in the negative 2-direction, pressing against the stationary domer in the flat region of the blank. At the end of previously flat region, while retaining the vertically drawn region from the first forming stage. Since the analysis is similar to the f scaling are the same as for the first forming stage. The loading rate in this stage of the forming simulation is actually slower than punch is shorter while the step time remains the same. Add the following to your input file for forming stage 2: *STEP *DYNAMIC, EXPLICIT , .033 *FIXED MASS SCALING, FACTOR=10 *DIAGNOSTICS, CONTACT INITIAL OVERCLOSURE=DETAIL, DEFORMATION SPEED CHECK=DETAIL *CONTACT PAIR, INTERACTION=INTER PUNCH2, BLANK_TOP DIE2, BLANK_BOT DOMER, BLANK_BOT *SURFACE INTERACTION, NAME=INTER *FRICTION .1, *BOUNDARY, TYPE=DISPLACEMENT, AMPLITUDE=FORM2, OP=NEW REFP2, 2, 2, -0.010

*BOUNDARY, TYPE=VELOCITY, OP=NEW REFP2, 1, 1, 0. REFP2, 6, 6, 0. REFD2, 1, 1, 0. REFD2, 6, 6, 0. SPEND, 1, 2, 0. REFDOMER, 1, 6, 0. BSYM, 1, 1, 0. BSYM, 6, 6, 0. *MONITOR, NODE=<node # for REFD2>, DOF=2 *RESTART, WRITE, NUMBER INTERVAL=10 *END STEP Save your input data in a file called draw_stage2.inp. Run the analysis using the following command: abaqus job=draw_stage2 oldjob=draw_spring1_std Evaluating the results for forming stage 2 Again you must evaluate the quality of the results as you did following the first forming stage to make certain that there are no se the details of checking the results again because the procedures are the same as were used to check the results from the first f kinetic energy, ALLKE, and the velocity of the reference node for die 2. Die 2 moves as it is pushed by the blank, which is pushed by punch 2. Since the bodies may not be exactly in contact at the sta When the blank impacts die 2, even with a small velocity, die 2 will oscillate somewhat since it sits on a spring and dashpot. Alth they have little effect on the solution because the predominant forming is occurring in the dome region. The deformed model fol Figure 7-31 Displaced model following forming stage 2. 7.5.16 Springback after forming stage 2 in ABAQUS/Standard Continue with another springback analysis in ABAQUS/Standard. Perform the springback analysis in ABAQUS/Standard just as y the *IMPORT option block to the following: *IMPORT, STATE=YES, UPDATE=NO BLANK, Save the springback input in a file called draw_spring2_std.inp, and run the springback analysis using the following command: abaqus job=draw_spring2_std oldjob=draw_stage2 Figure 7-32 shows a contour plot of SF1 following the final springback in ABAQUS/Standard. Figure 7-32 Contour plot of SF1 following final springback in ABAQUS/Standard (swept model for visualization purposes only). Figure 7-33 shows the final shape of the can bottom following the final springback in ABAQUS/Standard. Figure 7-33 Final deformed shape of can bottom (swept model for visualization purposes only). 7.6 Summary f a quasi-static analysis is performed in its natural time scale, the solution should be nearly the same as a truly static solution. t is often necessary to use load rate scaling or mass scaling to obtain a quasi-static solution using less CPU time. he loading rate often can be increased somewhat, as long as the solution does not localize. If the loading rate is increased to ass scaling is an alternative to increasing the loading rate. When using rate-dependent materials, mass scaling is preferable b properties. n a static analysis the lowest modes of the structure dominate the response. Knowing the lowest natural frequency and, corres time required to obtain the proper static response. t may be necessary to run a series of analyses at varying loading rates to determine an acceptable loading rate. he kinetic energy of the deforming material should not exceed a small fraction (typically 5% to 10%) of the internal energy thro sing the *AMPLITUDE option with the DEFINITION=SMOOTH STEP parameter is the most efficient way to prescribe displaceme f contact overclosures exist at the beginning of any step other than the first step, the contact overclosure is corrected in a sin severe element distortions. mport the model from ABAQUS/Explicit to ABAQUS/Standard to perform an efficient springback analysis. mport the model from ABAQUS/Standard to ABAQUS/Explicit to perform another forming stage following the springback analy Appendix A: Example Input Files This appendix provides complete input file listings for the examples contained in this guide. You can get a copy of any of these abaqus fetch job=<file name> where <file name> does not include the extension .inp. A.1 Overhead hoist frame frame_xpl.inp *HEADING

Two-dimensional overhead hoist frame SI units (kg, m, s, N) *PREPRINT, ECHO=YES, MODEL=YES, HISTORY=YES ** ** Model definition ** *NODE, NSET=NALL 101, 0., 0., 0. 102, 1., 0., 0. 103, 2., 0., 0. 104, 0.5, 0.866, 0. 105, 1.5, 0.866, 0. *ELEMENT, TYPE=T2D2, ELSET=FRAME 11, 101,102 12, 102,103 13, 101,104 14, 102,104 15, 102,105 16, 103,105 17, 104,105 *SOLID SECTION, ELSET=FRAME, MATERIAL=STEEL ** diameter = 5mm --> area = 1.963E-5 m^2 1.963E-5, *MATERIAL, NAME=STEEL *ELASTIC 200.E9, 0.3 *DENSITY 7800., ** ** History definition ** *STEP 10kN central load *DYNAMIC, EXPLICIT ,0.01 *BOUNDARY 101, ENCASTRE 103, 2 *CLOAD 102, 2, -10.E3 ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *FILE OUTPUT, NUMBER INTERVAL=20 *NODE FILE, NSET=NALL U, RF *EL FILE, ELSET=FRAME S, *END STEP A.2 Stress wave propagation in a bar wave_50x10x10.inp *HEADING Stress wave propagation in a bar -- 50x10x10 elements SI units (kg, m, s, N) *NODE,NSET=FLHS

1,0.,0. 21,0.,.2 *NGEN,NSET=FLHS 1,21,1 *NODE,NSET=FRHS 5001,1.,0. 5021,1.,.2 *NGEN,NSET=FRHS 5001,5021,1 *NFILL,NSET=NFRONT FLHS,FRHS,50,100 *NODE,NSET=BLHS 200001,0.,0.,-.2 200021,0.,.2,-.2 *NGEN,NSET=BLHS 200001,200021,1 *NODE,NSET=BRHS 205001,1.,0.,-.2 205021,1.,.2,-.2 *NGEN,NSET=BRHS 205001,205021,1 *NFILL,NSET=NBACK BLHS,BRHS,50,100 *NFILL,NSET=NALL NFRONT,NBACK,10,20000 *ELEMENT,TYPE=C3D8R,ELSET=BAR 1,1,3,103,101,20001,20003,20103,20101 *ELGEN,ELSET=BAR 1,10,2,2,50,100,100,10,20000,20000 *ELGEN,ELSET=ELOAD 1,10,2,2,10,20000,20000 *ELSET,ELSET=EOUT 81209,82509,83709 *NFILL,NSET=NLHS FLHS,BLHS,10,20000 *NFILL,NSET=NFIX FRHS,BRHS,10,20000 *NGEN,NSET=LBOT 1,200001,20000 *NGEN,NSET=RBOT 5001,205001,20000 *NFILL,NSET=NBOT LBOT,RBOT,50,100 *NGEN,NSET=LTOP 21,200021,20000 *NGEN,NSET=RTOP 5021,205021,20000 *NFILL,NSET=NTOP LTOP,RTOP,50,100 *SOLID SECTION,ELSET=BAR,MATERIAL=STEEL *MATERIAL,NAME=STEEL *ELASTIC 207.0E9,0.3 *DENSITY 7800.0,

*BOUNDARY NFIX,1,3 NFRONT,3,3 NBACK,3,3 NTOP,2,2 NBOT,2,2 *AMPLITUDE,NAME=BLAST 0.,1.,3.88E-5,1.,3.89E-5,0.,3.90E-5,0. *STEP Blast loading *DYNAMIC,EXPLICIT ,2.0E-4 *BULK VISCOSITY ,0.0 *DLOAD,AMPLITUDE=BLAST ELOAD,P3,1.0E5 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=4 *OUTPUT, HISTORY, TIME INTERVAL=1.0E-6 *ELEMENT OUTPUT, ELSET=EOUT S, ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *NSET,NSET=QA_TEST_NLHS NLHS, *OUTPUT, FIELD, NUMBER INTERVAL=2, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_NLHS U, *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=5.0E-5 ** *FILE OUTPUT,NUMBER INTERVAL=2,TIMEMARKS=YES *NODE FILE,NSET=NLHS U, *END STEP wave_25x5x5.inp *HEADING Stress wave propagation in a bar -- 25x5x5 elements SI units (kg, m, s, N) *NODE,NSET=FLHS 1,0.,0. 21,0.,.2 *NGEN,NSET=FLHS 1,21,1 *NODE,NSET=FRHS 5001,1.,0. 5021,1.,.2 *NGEN,NSET=FRHS 5001,5021,1 *NFILL,NSET=NFRONT FLHS,FRHS,50,100 *NODE,NSET=BLHS 200001,0.,0.,-.2 200021,0.,.2,-.2 *NGEN,NSET=BLHS 200001,200021,1

*NODE,NSET=BRHS 205001,1.,0.,-.2 205021,1.,.2,-.2 *NGEN,NSET=BRHS 205001,205021,1 *NFILL,NSET=NBACK BLHS,BRHS,50,100 *NFILL,NSET=NALL NFRONT,NBACK,10,20000 *ELEMENT,TYPE=C3D8R,ELSET=BAR 1,1,5,205,201,40001,40005,40205,40201 *ELGEN,ELSET=BAR 1,5,4,4,25,200,200,5,40000,40000 *ELGEN,ELSET=ELOAD 1,5,4,4,5,40000,40000 *NFILL,NSET=NLHS FLHS,BLHS,10,20000 *NFILL,NSET=NFIX FRHS,BRHS,10,20000 *NGEN,NSET=LBOT 1,200001,20000 *NGEN,NSET=RBOT 5001,205001,20000 *NFILL,NSET=NBOT LBOT,RBOT,50,100 *NGEN,NSET=LTOP 21,200021,20000 *NGEN,NSET=RTOP 5021,205021,20000 *NFILL,NSET=NTOP LTOP,RTOP,50,100 *SOLID SECTION,ELSET=BAR,MATERIAL=STEEL *MATERIAL,NAME=STEEL *ELASTIC 207.0E9,0.3 *DENSITY 7800.0, *BOUNDARY NFIX,1,3 NFRONT,3,3 NBACK,3,3 NTOP,2,2 NBOT,2,2 *AMPLITUDE,NAME=BLAST 0.,1.,3.88E-5,1.,3.89E-5,0.,3.90E-5,0. *STEP Blast loading *DYNAMIC,EXPLICIT ,2.0E-4 *BULK VISCOSITY ,0.0 *DLOAD,AMPLITUDE=BLAST ELOAD,P3,1.0E5 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=10 *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=1.0E-6

********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *NSET,NSET=QA_TEST_NLHS NLHS, *OUTPUT, FIELD, NUMBER INTERVAL=4, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_NLHS U, *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=5.0E-5 ** *FILE OUTPUT,NUMBER INTERVAL=4,TIMEMARKS=YES *NODE FILE,NSET=NLHS U, *END STEP wave_50x5x5.inp *HEADING Stress wave propagation in a bar -- 25x5x5 elements SI units (kg, m, s, N) *NODE,NSET=FLHS 1,0.,0. 21,0.,.2 *NGEN,NSET=FLHS 1,21,1 *NODE,NSET=FRHS 5001,1.,0. 5021,1.,.2 *NGEN,NSET=FRHS 5001,5021,1 *NFILL,NSET=NFRONT FLHS,FRHS,50,100 *NODE,NSET=BLHS 200001,0.,0.,-.2 200021,0.,.2,-.2 *NGEN,NSET=BLHS 200001,200021,1 *NODE,NSET=BRHS 205001,1.,0.,-.2 205021,1.,.2,-.2 *NGEN,NSET=BRHS 205001,205021,1 *NFILL,NSET=NBACK BLHS,BRHS,50,100 *NFILL,NSET=NALL NFRONT,NBACK,10,20000 *ELEMENT,TYPE=C3D8R,ELSET=BAR 1,1,5,105,101,40001,40005,40105,40101 *ELGEN,ELSET=BAR 1,5,4,4,50,100,100,5,40000,40000 *ELGEN,ELSET=ELOAD 1,5,4,4,5,40000,40000 *NFILL,NSET=NLHS FLHS,BLHS,10,20000 *NFILL,NSET=NFIX FRHS,BRHS,10,20000 *NGEN,NSET=LBOT

1,200001,20000 *NGEN,NSET=RBOT 5001,205001,20000 *NFILL,NSET=NBOT LBOT,RBOT,50,100 *NGEN,NSET=LTOP 21,200021,20000 *NGEN,NSET=RTOP 5021,205021,20000 *NFILL,NSET=NTOP LTOP,RTOP,50,100 *SOLID SECTION,ELSET=BAR,MATERIAL=STEEL *MATERIAL,NAME=STEEL *ELASTIC 207.0E9,0.3 *DENSITY 7800.0, *BOUNDARY NFIX,1,3 NFRONT,3,3 NBACK,3,3 NTOP,2,2 NBOT,2,2 *AMPLITUDE,NAME=BLAST 0.,1.,3.88E-5,1.,3.89E-5,0.,3.90E-5,0. *STEP Blast loading *DYNAMIC,EXPLICIT ,2.0E-4 *BULK VISCOSITY ,0.0 *DLOAD,AMPLITUDE=BLAST ELOAD,P3,1.0E5 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=10 *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=1.0E-6 ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *NSET,NSET=QA_TEST_NLHS NLHS, *OUTPUT, FIELD, NUMBER INTERVAL=4, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_NLHS U, *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=5.0E-5 ** *FILE OUTPUT,NUMBER INTERVAL=4,TIMEMARKS=YES *NODE FILE,NSET=NLHS U, *END STEP wave_50x10x5.inp *HEADING Stress wave propagation in a bar -- 50x10x5 elements SI units (kg, m, s, N) *NODE,NSET=FLHS 1,0.,0.

21,0.,.2 *NGEN,NSET=FLHS 1,21,1 *NODE,NSET=FRHS 5001,1.,0. 5021,1.,.2 *NGEN,NSET=FRHS 5001,5021,1 *NFILL,NSET=NFRONT FLHS,FRHS,50,100 *NODE,NSET=BLHS 200001,0.,0.,-.2 200021,0.,.2,-.2 *NGEN,NSET=BLHS 200001,200021,1 *NODE,NSET=BRHS 205001,1.,0.,-.2 205021,1.,.2,-.2 *NGEN,NSET=BRHS 205001,205021,1 *NFILL,NSET=NBACK BLHS,BRHS,50,100 *NFILL,NSET=NALL NFRONT,NBACK,10,20000 *ELEMENT,TYPE=C3D8R,ELSET=BAR 1,1,3,103,101,40001,40003,40103,40101 *ELGEN,ELSET=BAR 1,10,2,2,50,100,100,5,40000,40000 *ELGEN,ELSET=ELOAD 1,10,2,2,5,40000,40000 *NFILL,NSET=NLHS FLHS,BLHS,10,20000 *NFILL,NSET=NFIX FRHS,BRHS,10,20000 *NGEN,NSET=LBOT 1,200001,20000 *NGEN,NSET=RBOT 5001,205001,20000 *NFILL,NSET=NBOT LBOT,RBOT,50,100 *NGEN,NSET=LTOP 21,200021,20000 *NGEN,NSET=RTOP 5021,205021,20000 *NFILL,NSET=NTOP LTOP,RTOP,50,100 *SOLID SECTION,ELSET=BAR,MATERIAL=STEEL *MATERIAL,NAME=STEEL *ELASTIC 207.0E9,0.3 *DENSITY 7800.0, *BOUNDARY NFIX,1,3 NFRONT,3,3

NBACK,3,3 NTOP,2,2 NBOT,2,2 *AMPLITUDE,NAME=BLAST 0.,1.,3.88E-5,1.,3.89E-5,0.,3.90E-5,0. *STEP Blast loading *DYNAMIC,EXPLICIT ,2.0E-4 *BULK VISCOSITY ,0.0 *DLOAD,AMPLITUDE=BLAST ELOAD,P3,1.0E5 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=10 *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=1.0E-6 ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *NSET,NSET=QA_TEST_NLHS NLHS, *OUTPUT, FIELD, NUMBER INTERVAL=4, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_NLHS U, *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=5.0E-5 ** *FILE OUTPUT,NUMBER INTERVAL=4,TIMEMARKS=YES *NODE FILE,NSET=NLHS U, *END STEP A.3 Potential instabilities with springs and dashpots stability.inp *HEADING Spring instability SI units (kg, m, s, N) *NODE,NSET=NALL 1, 0.0, 0.0 2, 1.0, 0.0 3, 0.0, 1.0 4, 0.1, 1.0 *ELEMENT,TYPE=SPRINGA,ELSET=SPRING 1,1,2 *SPRING,ELSET=SPRING 1.0E5, *ELEMENT,TYPE=MASS,ELSET=MASS 2,2 *MASS,ELSET=MASS 5.0, *ELEMENT,TYPE=T2D2,ELSET=TRUSS 3,3,4 *SOLID SECTION,ELSET=TRUSS,MATERIAL=MAT1 1.0, *MATERIAL,NAME=MAT1 *DENSITY 10., *ELASTIC

1.0E4,0.0 *NSET,NSET=N2 2, ** *STEP Load the truss with constant velocity Load the spring/mass with constant force *DYNAMIC,EXPLICIT ,2.0 *BOUNDARY, TYPE=DISPLACEMENT NALL,2,2 1,1,3 3,1,1 *BOUNDARY, TYPE=VELOCITY 4,1,1,1 *CLOAD 2,1,1.0E3 *BULK VISCOSITY 0.0, *MONITOR,DOF=1,NODE=2 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=1 *OUTPUT, HISTORY, TIME INTERVAL=1.0E-3 *NODE OUTPUT,NSET=N2 U, *ENERGY OUTPUT ETOTAL, ALLKE, ALLIE, ALLWK ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *FILE OUTPUT,NUMBER INTERVAL=5 *NODE FILE U, *END STEP stability_addmass.inp *HEADING Spring instability *NODE,NSET=NALL 1, 0.0, 0.0 2, 1.0, 0.0 3, 0.0, 1.0 4, 0.1, 1.0 *ELEMENT,TYPE=SPRINGA,ELSET=SPRING 1,1,2 *SPRING,ELSET=SPRING 1.0E5, *ELEMENT,TYPE=MASS,ELSET=MASS 2,2 *MASS,ELSET=MASS 1760.0, *ELEMENT,TYPE=T2D2,ELSET=TRUSS 3,3,4 *SOLID SECTION,ELSET=TRUSS,MATERIAL=MAT1 1.0, *MATERIAL,NAME=MAT1 *DENSITY

10., *ELASTIC 1.0E4,0.0 *NSET,NSET=N2 2, ** ** *NSET, NSET=QA_TEST_ALLNODES NALL, N2 ** *STEP Load the truss with constant velocity Load the spring/mass with constant force *DYNAMIC,EXPLICIT ,2.0 *BOUNDARY, TYPE=DISPLACEMENT NALL,2,2 1,1,3 3,1,1 *BOUNDARY, TYPE=VELOCITY 4,1,1,1 *CLOAD 2,1,1.0E3 *BULK VISCOSITY 0.0, *MONITOR,DOF=1,NODE=2 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=1 *OUTPUT, HISTORY, TIME INTERVAL=1.0E-3 *NODE OUTPUT,NSET=N2 U, *ENERGY OUTPUT ETOTAL, ALLKE, ALLIE, ALLWK ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *FILE OUTPUT,NUMBER INTERVAL=5 *NODE FILE U, *OUTPUT, FIELD, NUMBER INTERVAL=5 *NODE OUTPUT, NSET=QA_TEST_ALLNODES U, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.4 ***OUTPUT,HISTORY,VARIABLE=PRESELECT *END STEP A.4 Hourglassing in a rubber block hourglass_square.inp *HEADING Hourglassing example 10 x 10 regular mesh SI units (kg, m, s, N) *NODE,NSET=NALL 1,0.,0. 21,.10,0. 1001,0.,.10 1021,.10,.10

*NGEN,NSET=NBOT 1,21,1 *NGEN,NSET=NTOP 1001,1021,1 *NFILL,NSET=NALL NBOT,NTOP,20,50 *NSET,NSET=NRHS,GENERATE 21,1021,50 *ELEMENT,TYPE=CPE4R,ELSET=EALL 1,1,3,103,101 *ELGEN,ELSET=EALL 1,10,2,2,10,100,100 *SOLID SECTION,ELSET=EALL,MATERIAL=RUBBER .10, *MATERIAL,NAME=RUBBER ** hyperelastic constants in Pa *HYPERELASTIC,POLYNOMIAL,N=1 3.2E6,.8E6 ** density is 1500 kg/m^3 *DENSITY 1500., ** ** RIGID ELEMENT USED TO COMPRESS THE BLOCK ** *NODE, NSET=NRIGID 20001,-.10,0. 20002,.10,.20 20003,-.02,.12 *NSET,NSET=NREF 20003, *ELEMENT,TYPE=R2D2,ELSET=ERIGID 20001,20002,20001 *AMPLITUDE,NAME=CRUSH,DEFINITION=SMOOTH STEP 0.,0.,.1,1. *BOUNDARY NBOT,2 NRHS,1 NREF,6 ** *SURFACE,NAME=RIGID ERIGID,SPOS *SURFACE,NAME=SOLID EALL, *RIGID BODY,ELSET=ERIGID,REFNODE=20003 ** *ELSET,ELSET=QA_TEST_ALLELEMS EALL, ** *NSET, NSET=QA_TEST_ALLNODES NALL, NBOT, NTOP, NALL NRHS, NRIGID, NREF ** *STEP Compress the rubber block *DYNAMIC,EXPLICIT

,0.1 *CONTACT PAIR,INTERACTION=INTER SOLID,RIGID *SURFACE INTERACTION,NAME=INTER ** displace a total distance of .012 in both directions *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=CRUSH NREF,1,1,.012 NREF,2,2,-.012 ** *OUTPUT, FIELD, NUMBER INTERVAL=20, VARIABLE=PRESELECT *OUTPUT, HISTORY, FREQUENCY=1 *ENERGY OUTPUT ALLAE, ALLKE, ALLSE, ALLVD, ALLFD, ALLWK, ETOTAL ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *OUTPUT, FIELD, NUMBER INTERVAL=1, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_ALLNODES U, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.1 ** *FILE OUTPUT,NUMBER INTERVAL=1 *NODE FILE U, *END STEP hourglass_square_fine.inp *HEADING Hourglassing example 20 x 20 regular mesh SI units (kg, m, s, N) *NODE,NSET=NALL 1,0.,0. 21,.10,0. 1001,0.,.10 1021,.10,.10 *NGEN,NSET=NBOT 1,21,1 *NGEN,NSET=NTOP 1001,1021,1 *NFILL,NSET=NALL NBOT,NTOP,20,50 *NSET,NSET=NRHS,GEN 21,1021,50 *ELEMENT,TYPE=CPE4R,ELSET=EALL 1,1,2,52,51 *ELGEN,ELSET=EALL 1,20,1,1,20,50,50 *SOLID SECTION,ELSET=EALL,MATERIAL=RUBBER .10, *MATERIAL,NAME=RUBBER ** hyperelastic constants in Pa *HYPERELASTIC,POLYNOMIAL,N=1 3.2E6,.8E6 ** density is 1500 kg/m^3

*DENSITY 1500., ** ** RIGID ELEMENT USED TO COMPRESS THE BLOCK ** *NODE, NSET=NRIGID 20001,-.10,0. 20002,.10,.20 20003,-.02,.12 *NSET,NSET=NREF 20003, *ELEMENT,TYPE=R2D2,ELSET=ERIGID 20001,20002,20001 *AMPLITUDE,NAME=CRUSH,DEFINITION=SMOOTH STEP 0.,0.,.1,1. *BOUNDARY NBOT,2 NRHS,1 NREF,6 ** *SURFACE,NAME=RIGID ERIGID,SPOS *SURFACE,NAME=SOLID EALL, *RIGID BODY,ELSET=ERIGID,REFNODE=20003 ** *ELSET,ELSET=QA_TEST_ALLELEMS EALL, ** *NSET, NSET=QA_TEST_ALLNODES NALL, NBOT, NTOP, NALL NRHS, NRIGID, NREF ** *STEP Compress the rubber block *DYNAMIC,EXPLICIT ,0.1 *CONTACT PAIR,INTERACTION=INTER SOLID,RIGID *SURFACE INTERACTION,NAME=INTER ** displace a total distance of .012 in both directions *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=CRUSH NREF,1,1,.012 NREF,2,2,-.012 ** *OUTPUT, FIELD, NUMBER INTERVAL=20, VARIABLE=PRESELECT *OUTPUT, HISTORY, FREQUENCY=1 *ENERGY OUTPUT ALLAE, ALLKE, ALLSE, ALLVD, ALLFD, ALLWK, ETOTAL ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *OUTPUT, FIELD, NUMBER INTERVAL=1, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_ALLNODES

U, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.1 ** *FILE OUTPUT,NUMBER INTERVAL=1 *NODE FILE U, *END STEP hourglass_flat_corner.inp *HEADING Hourglassing example 10 x 10 regular mesh with triangular corner SI units (kg, m, s, N) *NODE,NSET=NALL 1,0.,0. 21,.10,0. 1001,0.,.10 1021,.10,.10 *NGEN,NSET=NBOT 1,21,1 *NGEN,NSET=NTOP 1001,1021,1 *NFILL,NSET=NALL NBOT,NTOP,20,50 *NSET,NSET=NRHS,GENERATE 21,1021,50 *ELEMENT,TYPE=CPE4R,ELSET=EALL 1,1,3,103,101 *ELGEN,ELSET=EALL 1,10,2,2,9,100,100 *ELEMENT,TYPE=CPE4R,ELSET=EALL 903,903,905,1005,1003 *ELGEN,ELSET=EALL 903,9,2,2 *ELEMENT,TYPE=CPE3,ELSET=EALL 901,901,903,1003 *SOLID SECTION,ELSET=EALL,MATERIAL=RUBBER .10, *MATERIAL,NAME=RUBBER ** hyperelastic constants in Pa *HYPERELASTIC,POLYNOMIAL,N=1 3.2E6,.8E6 ** density is 1500 kg/m^3 *DENSITY 1500., ** ** RIGID ELEMENT USED TO COMPRESS THE BLOCK ** *NODE, NSET=NRIGID 20001,-.095,-.005 20002,.105,.195 20003,-.02,.12 *NSET,NSET=NREF 20003, *ELEMENT,TYPE=R2D2,ELSET=ERIGID 20001,20002,20001

*AMPLITUDE,NAME=CRUSH,DEFINITION=SMOOTH STEP 0.,0.,.081,1. *BOUNDARY NBOT,2 NRHS,1 NREF,6 ** *SURFACE,NAME=RIGID ERIGID,SPOS *SURFACE,NAME=SOLID EALL, *RIGID BODY,ELSET=ERIGID,REFNODE=20003 *STEP Compress the rubber block *DYNAMIC,EXPLICIT ,.081 *CONTACT PAIR,INTERACTION=INTER SOLID,RIGID *SURFACE INTERACTION,NAME=INTER ** displace a total distance of .012 in both directions *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=CRUSH NREF,1,1,.007 NREF,2,2,-.007 ** *OUTPUT, FIELD, NUMBER INTERVAL=20, VARIABLE=PRESELECT *OUTPUT, HISTORY, FREQUENCY=1 *ENERGY OUTPUT ALLAE, ALLKE, ALLSE, ALLVD, ALLFD, ALLWK, ETOTAL ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *NSET,NSET=QA_TEST_ALLNODES NALL, *OUTPUT, FIELD, NUMBER INTERVAL=1, TIMEMARKS=YES *NODE OUTPUT,NSET=QA_TEST_ALLNODES U, *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=.081 ** *FILE OUTPUT,NUMBER INTERVAL=1 *NODE FILE U, *END STEP hourglass_flat_corner_fine.inp *HEADING Hourglassing example 20 x 20 regular mesh with triangular corner SI units (kg, m, s, N) *NODE,NSET=NALL 1,0.,0. 21,.10,0. 1001,0.,.10 1021,.10,.10 *NGEN,NSET=NBOT 1,21,1

*NGEN,NSET=NTOP 1001,1021,1 *NFILL,NSET=NALL NBOT,NTOP,20,50 *NSET,NSET=NRHS,GENERATE 21,1021,50 *ELEMENT,TYPE=CPE4R,ELSET=EALL 1,1,2,52,51 *ELGEN,ELSET=EALL 1,20,1,1,19,50,50 *ELEMENT,TYPE=CPE4R,ELSET=EALL 952,952,953,1003,1002 *ELGEN,ELSET=EALL 952,19,1,1 *ELEMENT,TYPE=CPE3,ELSET=EALL 951,951,952,1002 *SOLID SECTION,ELSET=EALL,MATERIAL=RUBBER .10, *MATERIAL,NAME=RUBBER ** hyperelastic constants in Pa *HYPERELASTIC,POLYNOMIAL,N=1 3.2E6,.8E6 ** density is 1500 kg/m^3 *DENSITY 1500., ** ** RIGID ELEMENT USED TO COMPRESS THE BLOCK ** *NODE, NSET=NRIGID 20001,-.0975,-.0025 20002,.1025,.1975 20003,-.02,.12 *NSET,NSET=NREF 20003, *ELEMENT,TYPE=R2D2,ELSET=ERIGID 20001,20002,20001 *AMPLITUDE,NAME=CRUSH, DEFINITION=SMOOTH STEP 0.,0.,.095,1. *BOUNDARY NBOT,2 NRHS,1 NREF,6 ** *SURFACE,NAME=RIGID ERIGID,SPOS *SURFACE,NAME=SOLID EALL, *RIGID BODY,ELSET=ERIGID,REFNODE=20003 *STEP Compress the rubber block *DYNAMIC,EXPLICIT ,.095 *CONTACT PAIR,INTERACTION=INTER SOLID,RIGID *SURFACE INTERACTION,NAME=INTER

** displace a total distance of .012 in both directions *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=CRUSH NREF,1,1,.0095 NREF,2,2,-.0095 ** *OUTPUT, FIELD, NUMBER INTERVAL=20, VARIABLE=PRESELECT *OUTPUT, HISTORY, FREQUENCY=1 *ENERGY OUTPUT ALLAE, ALLKE, ALLSE, ALLVD, ALLFD, ALLWK, ETOTAL ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *NSET,NSET=QA_TEST_ALLNODES NALL, *OUTPUT, FIELD, NUMBER INTERVAL=1, TIMEMARKS=YES *NODE OUTPUT,NSET=QA_TEST_ALLNODES U, *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=.081 ** *FILE OUTPUT,NUMBER INTERVAL=1 *NODE FILE U, *END STEP hourglass_rounded.inp *HEADING Hourglassing example Coarse rounded mesh SI units (kg, m, s, N) ** *NODE, NSET=QA_TEST_ALLNODES 1,5.78316e-10 , 0.0 2,0.1 , 0.0 3,0.01 , -8.17963e-09 4,0.02 , -1.77636e-17 5,0.03 , 0.0 6,0.04 , -8.88178e-18 7,0.05 , 0.0 8,0.06 , 0.0 9,0.07 , 0.0 10,0.08 , 0.0 11,0.09 , 1.77636e-17 12,0.1 , 0.1 13,0.1 , 0.01 14,0.1 , 0.02 15,0.1 , 0.03 16,0.1 , 0.04 17,0.1 , 0.05 18,0.1 , 0.06 19,0.1 , 0.07 20,0.1 , 0.08 21,0.1 , 0.09 22,0.01 , 0.1 23,0.09 , 0.1 24,0.08 , 0.1

25,0.07 , 0.1 26,0.06 , 0.1 27,0.05 , 0.1 28,0.04 , 0.1 29,0.03 , 0.1 30,0.02 , 0.1 31,-8.88178e-18 ,0.09 32,0.00292893 , 0.0970711 33,0.0 , 0.08 34,8.88178e-18 , 0.07 35,-3.55271e-17 ,0.06 36,1.77636e-17 , 0.05 37,-1.77636e-17 ,0.04 38,1.77636e-17 , 0.03 39,-1.77636e-17 ,0.02 40,8.88178e-18 , 0.01 41,0.01 , 0.01 42,0.02 , 0.01 43,0.03 , 0.01 44,0.04 , 0.01 45,0.05 , 0.01 46,0.06 , 0.01 47,0.07 , 0.01 48,0.08 , 0.01 49,0.09 , 0.01 50,0.09 , 0.02 51,0.09 , 0.03 52,0.09 , 0.04 53,0.09 , 0.05 54,0.09 , 0.06 55,0.09 , 0.07 56,0.09 , 0.08 57,0.09 , 0.09 58,0.08 , 0.09 59,0.07 , 0.09 60,0.06 , 0.09 61,0.0503229 , 0.0904665 62,0.0409745 , 0.0908973 63,0.0316622 , 0.0910528 64,0.0227124 , 0.0916174 65,0.0146448 , 0.0916235 66,0.00792875 , 0.0912914 67,0.00647801 , 0.0859062 68,0.00714392 , 0.0773471 69,0.00770258 , 0.068442 70,0.00838407 , 0.0594883 71,0.00909115 , 0.0499915 72,0.00948701 , 0.0399975 73,0.00971194 , 0.0299992 74,0.00986479 , 0.0199993 75,0.0200909 , 0.0201311 76,0.0301478 , 0.0201596 77,0.040162 , 0.0201569 78,0.0501655 , 0.0201507 79,0.0601664 , 0.0201465

80,0.0701666 , 81,0.0801666 , 82,0.0800105 , 83,0.0799231 , 84,0.0792047 , 85,0.0798683 , 86,0.0798618 , 87,0.0798831 , 88,0.069845 , 89,0.0601382 , 90,0.0506204 , 91,0.0410646 , 92,0.0317326 , 93,0.022659 , 94,0.0141192 , 95,0.0154997 , 96,0.0161073 , 97,0.0172161 , 98,0.0184092 , 99,0.0192782 , 100,0.0197972 , 101,0.030197 , 102,0.0403787 , 103,0.0503986 , 104,0.0604724 , 105,0.0705017 , 106,0.0236526 , 107,0.0250679 , 108,0.0264482 , 109,0.0278873 , 110,0.0291762 , 111,0.0325028 , 112,0.0338676 , 113,0.0356554 , 114,0.0374627 , 115,0.0389564 , 116,0.0417333 , 117,0.0433002 , 118,0.0449965 , 119,0.0476309 , 120,0.0488952 , 121,0.0511686 , 122,0.0523448 , 123,0.053426 , 124,0.0580377 , 125,0.059107 , 126,0.0691865 , 127,0.0684489 , 128,0.0702447 , 129,0.0702772 , 130,0.0607197 , 131,0.0613644 , 132,0.0607822 , ** **

0.0201443 0.0201189 0.0301555 0.0401642 0.0501634 0.0601666 0.0701667 0.0801667 0.0800097 0.0808345 0.0815274 0.0817862 0.0820925 0.0830882 0.0837189 0.075066 0.0668381 0.0585046 0.0496626 0.0401657 0.0300608 0.0303886 0.0304388 0.0299076 0.0298293 0.0303151 0.0744256 0.0655438 0.0573525 0.0491529 0.0404657 0.0734996 0.0655433 0.0577415 0.0494021 0.0403291 0.0730185 0.0649617 0.0567479 0.0483452 0.0396955 0.0725486 0.0641522 0.0563392 0.0480735 0.0389287 0.0401552 0.0502223 0.0601388 0.0702667 0.0724696 0.0641471 0.0573286

*ELEMENT, TYPE=CPE4R , ELSET=EALL 1, 3, 41, 40, 1 2, 4, 42, 41, 3 3, 5, 43, 42, 4 4, 6, 44, 43, 5 5, 7, 45, 44, 6 6, 8, 46, 45, 7 7, 9, 47, 46, 8 8, 10, 48, 47, 9 9, 11, 49, 48, 10 10, 13, 49, 11, 2 11, 14, 50, 49, 13 12, 15, 51, 50, 14 13, 16, 52, 51, 15 14, 17, 53, 52, 16 15, 18, 54, 53, 17 16, 19, 55, 54, 18 17, 20, 56, 55, 19 18, 21, 57, 56, 20 19, 23, 57, 21, 12 20, 24, 58, 57, 23 21, 25, 59, 58, 24 22, 26, 60, 59, 25 23, 27, 61, 60, 26 24, 28, 62, 61, 27 25, 29, 63, 62, 28 26, 30, 64, 63, 29 27, 22, 65, 64, 30 28, 32, 66, 65, 22 29, 31, 67, 66, 32 30, 33, 68, 67, 31 31, 34, 69, 68, 33 32, 35, 70, 69, 34 33, 36, 71, 70, 35 34, 37, 72, 71, 36 35, 38, 73, 72, 37 36, 39, 74, 73, 38 37, 41, 74, 39, 40 38, 42, 75, 74, 41 39, 43, 76, 75, 42 40, 44, 77, 76, 43 41, 45, 78, 77, 44 42, 46, 79, 78, 45 43, 47, 80, 79, 46 44, 48, 81, 80, 47 45, 50, 81, 48, 49 46, 51, 82, 81, 50 47, 52, 83, 82, 51 48, 53, 84, 83, 52 49, 54, 85, 84, 53 50, 55, 86, 85, 54 51, 56, 87, 86, 55 52, 58, 87, 56, 57 53, 59, 88, 87, 58 54, 60, 89, 88, 59

55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109,

61, 62, 63, 64, 65, 67, 68, 69, 70, 71, 72, 73, 75, 76, 77, 78, 79, 80, 82, 95, 96, 97, 98, 99, 101, 106, 107, 108, 109, 110, 102, 111, 112, 113, 114, 115, 103, 116, 117, 118, 119, 120, 104, 83, 84, 85, 86, 88, 121, 122, 123, 124, 104, 130, 131,

90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 100, 101, 102, 103, 104, 105, 105, 106, 107, 108, 109, 110, 110, 111, 112, 113, 114, 115, 115, 116, 117, 118, 119, 120, 120, 121, 122, 123, 124, 125, 125, 126, 127, 128, 129, 129, 130, 131, 132, 127, 105, 129, 128,

89, 90, 91, 92, 93, 65, 94, 95, 96, 97, 98, 99, 73, 100, 101, 102, 103, 104, 80, 93, 106, 107, 108, 109, 99, 92, 111, 112, 113, 114, 110, 91, 116, 117, 118, 119, 115, 90, 121, 122, 123, 124, 120, 105, 126, 127, 128, 86, 89, 130, 131, 132, 126, 88, 129,

60 61 62 63 64 66 67 68 69 70 71 72 74 75 76 77 78 79 81 94 95 96 97 98 100 93 106 107 108 109 101 92 111 112 113 114 102 91 116 117 118 119 103 82 83 84 85 87 90 121 122 123 125 89 130

110, 111,

132, 124,

127, 125,

128, 126,

131 127

** ** BOT ** *NSET,NSET=NBOT 1, 3, 4, 5, 6, 7, 8, 9, 10, 11, ** ** RHS ** *NSET,NSET=NRHS 2, 2, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, *SOLID SECTION,ELSET=EALL,MATERIAL=RUBBER .10, *MATERIAL,NAME=RUBBER ** hyperelastic constants in Pa *HYPERELASTIC,POLYNOMIAL,N=1 3.2E6,.8E6 ** density is 1500 kg/m^3 *DENSITY 1500., ** ** RIGID ELEMENT USED TO COMPRESS THE BLOCK ** *NODE, NSET=NRIGID 20001,-.097071,-.002929 20002,.102929,.197971 20003,-.02,.12 *NSET,NSET=NREF 20003, *ELEMENT,TYPE=R2D2,ELSET=ERIGID 20001,20002,20001 *AMPLITUDE,NAME=CRUSH,DEFINITION=SMOOTH STEP 0.,0.,.1,1. *BOUNDARY

NBOT,2 NRHS,1 NREF,6 ** *SURFACE,NAME=RIGID ERIGID,SPOS *SURFACE,NAME=SOLID EALL, *RIGID BODY,ELSET=ERIGID,REFNODE=20003 ** *ELSET,ELSET=QA_TEST_ALLELEMS EALL, ** *NSET, NSET=QA_TEST_ALLNODES QA_TEST_ALLNODES, NBOT, NRHS, NRIGID NREF, ** *STEP Compress the rubber block *DYNAMIC,EXPLICIT ,0.1 *CONTACT PAIR,INTERACTION=INTER SOLID,RIGID *SURFACE INTERACTION,NAME=INTER ** displace a total distance of .012 in both directions *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=CRUSH NREF,1,1,.009071 NREF,2,2,-.009071 ** *OUTPUT, FIELD, NUMBER INTERVAL=20, VARIABLE=PRESELECT *OUTPUT, HISTORY, FREQUENCY=1 *ENERGY OUTPUT ALLAE, ALLKE, ALLSE, ALLVD, ALLFD, ALLWK, ETOTAL ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *OUTPUT, FIELD, NUMBER INTERVAL=1, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_ALLNODES U, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.1 ** *FILE OUTPUT,NUMBER INTERVAL=1 *NODE FILE U, *END STEP hourglass_rounded_fine.inp *HEADING Hourglassing example Fine rounded mesh SI units (kg, m, s, N) ** *NODE, NSET=QA_TEST_ALLNODES 1,5.78316e-10 , 5.78316e-10 2,0.1 , 0.0

3,0.00476191 , -8.74009e-09 4,0.00952381 , -8.88178e-18 5,0.0142857 , 0.0 6,0.0190476 , 1.77636e-17 7,0.0238095 , 0.0 8,0.0285714 , 0.0 9,0.0333333 , -1.77636e-17 10,0.0380952 , 8.88178e-18 11,0.0428571 , 0.0 12,0.047619 , -8.88178e-18 13,0.052381 , -8.88178e-18 14,0.0571429 , -8.88178e-18 15,0.0619048 , 0.0 16,0.0666667 , 8.88178e-18 17,0.0714286 , 0.0 18,0.0761905 , 0.0 19,0.0809524 , -8.88178e-18 20,0.0857143 , 8.88178e-18 21,0.0904762 , -8.88178e-18 22,0.0952381 , 8.88178e-18 23,0.1 , 0.1 24,0.1 , 0.005 25,0.1 , 0.01 26,0.1 , 0.015 27,0.1 , 0.02 28,0.1 , 0.025 29,0.1 , 0.03 30,0.1 , 0.035 31,0.1 , 0.04 32,0.1 , 0.045 33,0.1 , 0.05 34,0.1 , 0.055 35,0.1 , 0.06 36,0.1 , 0.065 37,0.1 , 0.07 38,0.1 , 0.075 39,0.1 , 0.08 40,0.1 , 0.085 41,0.1 , 0.09 42,0.1 , 0.095 43,0.01 , 0.1 44,0.095 , 0.1 45,0.09 , 0.1 46,0.085 , 0.1 47,0.08 , 0.1 48,0.075 , 0.1 49,0.07 , 0.1 50,0.065 , 0.1 51,0.06 , 0.1 52,0.055 , 0.1 53,0.05 , 0.1 54,0.045 , 0.1 55,0.04 , 0.1 56,0.035 , 0.1 57,0.03 , 0.1

58,0.025 , 0.1 59,0.02 , 0.1 60,0.015 , 0.1 61,0.0 , 0.09 62,0.005 , 0.0986602 63,0.00133975 , 0.095 64,-8.88178e-18 , 0.085 65,0.0 , 0.08 66,-1.77636e-17 , 0.075 67,1.77636e-17 , 0.07 68,-8.88178e-18 , 0.065 69,1.77636e-17 , 0.06 70,0.0 , 0.055 71,0.0 , 0.05 72,-8.88178e-18 , 0.045 73,-8.88178e-18 , 0.04 74,0.0 , 0.035 75,8.88178e-18 , 0.03 76,8.88178e-18 , 0.025 77,0.0 , 0.02 78,-8.88178e-18 , 0.015 79,0.0 , 0.00999998 80,8.88178e-18 , 0.00499998 81,0.00476256 , 0.00491069 82,0.00952401 , 0.00484735 83,0.0142858 , 0.00481 84,0.0190476 , 0.00478884 85,0.0238095 , 0.00477698 86,0.0285714 , 0.00477033 87,0.0333333 , 0.00476662 88,0.0380952 , 0.00476454 89,0.0428571 , 0.00476338 90,0.047619 , 0.00476273 91,0.052381 , 0.00476237 92,0.0571429 , 0.00476216 93,0.0619048 , 0.00476205 94,0.0666667 , 0.00476199 95,0.0714286 , 0.00476195 96,0.0761905 , 0.00476193 97,0.0809524 , 0.00476191 98,0.0857143 , 0.00476191 99,0.0904762 , 0.00476191 100,0.0952384 , 0.00481482 101,0.095 , 0.095 102,0.09 , 0.095 103,0.085 , 0.095 104,0.08 , 0.095 105,0.075 , 0.095 106,0.07 , 0.095 107,0.065 , 0.095 108,0.06 , 0.095 109,0.055 , 0.095 110,0.05 , 0.095 111,0.045 , 0.095 112,0.04 , 0.095

113,0.035 , 0.095 114,0.03 , 0.095 115,0.025 , 0.095 116,0.0202439 , 0.095 117,0.0154874 , 0.095 118,0.0111804 , 0.0951625 119,0.00736927 , 0.094562 120,0.00395472 , 0.0927692 121,0.00358206 , 0.0885104 122,0.00399329 , 0.0837514 123,0.00446495 , 0.0790828 124,0.00485982 , 0.0744614 125,0.00486418 , 0.0697215 126,0.00476997 , 0.0649983 127,0.0048656 , 0.0599996 128,0.0049239 , 0.0549999 129,0.00495728 , 0.05 130,0.00497607 , 0.045 131,0.00498661 , 0.04 132,0.00499251 , 0.035 133,0.00499581 , 0.03 134,0.00499766 , 0.025 135,0.00499869 , 0.02 136,0.00499927 , 0.015 137,0.00494708 , 0.00999964 138,0.0950956 , 0.00995332 139,0.0950312 , 0.0149882 140,0.0950093 , 0.019997 141,0.0950026 , 0.0249993 142,0.0950007 , 0.0299998 143,0.0950002 , 0.035 144,0.095 , 0.04 145,0.095 , 0.045 146,0.095 , 0.05 147,0.095 , 0.055 148,0.095 , 0.06 149,0.095 , 0.065 150,0.095 , 0.07 151,0.095 , 0.075 152,0.095 , 0.08 153,0.095 , 0.085 154,0.095 , 0.09 155,0.00969311 , 0.00981222 156,0.0144388 , 0.00968934 157,0.0191861 , 0.00962749 158,0.0239355 , 0.00959954 159,0.0286871 , 0.00958807 160,0.0334406 , 0.00958417 161,0.038196 , 0.00958357 162,0.0429528 , 0.0095843 163,0.0477108 , 0.00958545 164,0.0524697 , 0.00958663 165,0.0572293 , 0.00958769 166,0.0619894 , 0.00958857 167,0.06675 , 0.00958929

168,0.071511 , 0.00958985 169,0.0762721 , 0.00959029 170,0.0810334 , 0.00959062 171,0.0857949 , 0.00959088 172,0.0905497 , 0.00964399 173,0.09 , 0.09 174,0.085 , 0.09 175,0.08 , 0.09 176,0.075 , 0.09 177,0.07 , 0.09 178,0.065 , 0.09 179,0.06 , 0.09 180,0.055 , 0.09 181,0.05 , 0.0897364 182,0.0450128 , 0.0904875 183,0.04 , 0.09 184,0.035 , 0.09 185,0.03 , 0.09 186,0.025 , 0.09 187,0.0201463 , 0.090056 188,0.0156477 , 0.0901038 189,0.0114117 , 0.0905262 190,0.00749454 , 0.0898208 191,0.0902107 , 0.0148139 192,0.0900408 , 0.0199138 193,0.0899907 , 0.0249623 194,0.0899851 , 0.0299841 195,0.0899898 , 0.0349934 196,0.0899944 , 0.0399973 197,0.0899973 , 0.0449989 198,0.0899988 , 0.0499996 199,0.0899995 , 0.0549998 200,0.0899998 , 0.0599999 201,0.0899999 , 0.065 202,0.09 , 0.07 203,0.09 , 0.075 204,0.09 , 0.08 205,0.09 , 0.085 206,0.00784526 , 0.0851443 207,0.00853498 , 0.080095 208,0.00967079 , 0.0750524 209,0.00973984 , 0.0700113 210,0.00978195 , 0.0650577 211,0.00985999 , 0.0600674 212,0.00992457 , 0.0550568 213,0.00996406 , 0.0500415 214,0.00998455 , 0.0450281 215,0.00999408 , 0.0400182 216,0.00999813 , 0.0350114 217,0.00999968 , 0.030007 218,0.0100002 , 0.0250042 219,0.0100003 , 0.0200025 220,0.00997261 , 0.0149884 221,0.0147512 , 0.0147034 222,0.019503 , 0.0145233

223,0.0242525 , 224,0.0290008 , 225,0.0337491 , 226,0.0384983 , 227,0.0432485 , 228,0.0480006 , 229,0.0527528 , 230,0.0575067 , 231,0.0622617 , 232,0.0670176 , 233,0.0717744 , 234,0.076532 , 235,0.0812902 , 236,0.085741 , 237,0.085 , 238,0.08 , 239,0.075 , 240,0.07 , 241,0.065 , 242,0.06 , 243,0.055 , 244,0.0497093 , 245,0.0450058 , 246,0.0400137 , 247,0.0350001 , 248,0.03 , 249,0.0252767 , 250,0.0207309 , 251,0.0161891 , 252,0.0117988 , 253,0.0126945 , 254,0.0140849 , 255,0.0148683 , 256,0.0149574 , 257,0.0149849 , 258,0.0150056 , 259,0.0150204 , 260,0.015026 , 261,0.0150246 , 262,0.01502 , 263,0.0150148 , 264,0.0150103 , 265,0.0149875 , 266,0.085504 , 267,0.0853652 , 268,0.0853371 , 269,0.0852776 , 270,0.0849545 , 271,0.0849735 , 272,0.0849854 , 273,0.084992 , 274,0.0849957 , 275,0.0849976 , 276,0.0849987 , 277,0.0849993 ,

0.0144475 0.0144221 0.014416 0.0144166 0.0144193 0.0146643 0.0144258 0.0144288 0.0144315 0.0144338 0.0144358 0.0144375 0.014439 0.0144407 0.085 0.085 0.085 0.085 0.085 0.085 0.085 0.0841437 0.085899 0.0853578 0.0852705 0.085 0.0850155 0.0852459 0.0852459 0.0858309 0.0809067 0.075896 0.0706397 0.0654483 0.0603158 0.0552272 0.0501605 0.0451102 0.0400736 0.0350481 0.0300309 0.0250196 0.0200064 0.0196104 0.024735 0.0298033 0.0348925 0.0399478 0.0449707 0.0499835 0.0549907 0.0599948 0.064997 0.0699983 0.0749991

278,0.0849995 , 279,0.0198316 , 280,0.0246018 , 281,0.0293587 , 282,0.0341133 , 283,0.0388232 , 284,0.0435806 , 285,0.048402 , 286,0.0531241 , 287,0.0578769 , 288,0.06263 , 289,0.0673835 , 290,0.0721376 , 291,0.0765606 , 292,0.0810093 , 293,0.0799997 , 294,0.0749999 , 295,0.0699999 , 296,0.0646637 , 297,0.0593474 , 298,0.0537918 , 299,0.0483624 , 300,0.0444397 , 301,0.0401636 , 302,0.0353663 , 303,0.0305006 , 304,0.0257917 , 305,0.0212672 , 306,0.0170234 , 307,0.018767 , 308,0.0196936 , 309,0.0200256 , 310,0.0201145 , 311,0.0201281 , 312,0.0201216 , 313,0.0201075 , 314,0.0200893 , 315,0.0200699 , 316,0.0200521 , 317,0.0200298 , 318,0.0807222 , 319,0.0806742 , 320,0.0805401 , 321,0.0798441 , 322,0.0798919 , 323,0.0799312 , 324,0.0799574 , 325,0.0799738 , 326,0.0799837 , 327,0.0799898 , 328,0.0799927 , 329,0.024923 , 330,0.0295902 , 331,0.0343522 , 332,0.0391312 ,

0.0799995 0.0196089 0.0193683 0.0192786 0.0192557 0.0195078 0.0195896 0.0201023 0.0192628 0.0192661 0.0192693 0.0192723 0.0192751 0.0192767 0.0192394 0.0799999 0.08 0.08 0.0800006 0.079986 0.0798593 0.0785067 0.0818527 0.0810108 0.0805349 0.0803992 0.0803748 0.0804467 0.0805618 0.0756879 0.0709885 0.0658202 0.0606433 0.0554937 0.0503724 0.0452747 0.0401981 0.0351399 0.0300972 0.025063 0.0243303 0.0294858 0.0347306 0.0397629 0.0448468 0.0499016 0.054937 0.0599597 0.0649743 0.0699836 0.0749896 0.0245338 0.0245417 0.0245466 0.0246084

333,0.0439817 , 334,0.0489837 , 335,0.0535794 , 336,0.0583711 , 337,0.0630566 , 338,0.0675354 , 339,0.0720923 , 340,0.0764514 , 341,0.0231647 , 342,0.0244391 , 343,0.025058 , 344,0.0252451 , 345,0.0252871 , 346,0.0252752 , 347,0.0252422 , 348,0.0252006 , 349,0.0251587 , 350,0.0251078 , 351,0.0278267 , 352,0.0292771 , 353,0.0298018 , 354,0.0302456 , 355,0.0304024 , 356,0.0304301 , 357,0.0304052 , 358,0.0303559 , 359,0.0303824 , 360,0.0301692 , 361,0.0327424 , 362,0.0340745 , 363,0.034687 , 364,0.0351511 , 365,0.035471 , 366,0.0355743 , 367,0.0355787 , 368,0.0355385 , 369,0.0355759 , 370,0.0352385 , 371,0.0377128 , 372,0.0389279 , 373,0.0395415 , 374,0.0399766 , 375,0.0404684 , 376,0.0406805 , 377,0.0407381 , 378,0.0407215 , 379,0.0407927 , 380,0.0403158 , 381,0.0429949 , 382,0.0438735 , 383,0.044453 , 384,0.0450048 , 385,0.0455152 , 386,0.0458113 , 387,0.0459334 ,

0.0247635 0.0255213 0.0237035 0.024126 0.0241092 0.0243842 0.0241132 0.0240275 0.075773 0.0710298 0.0662511 0.0610671 0.0558784 0.0506961 0.0455317 0.0403993 0.0352949 0.0299049 0.0757026 0.0711284 0.0665854 0.061445 0.0562607 0.0510698 0.0458803 0.0406974 0.035272 0.0298353 0.0758878 0.0712924 0.0667319 0.0618474 0.0566767 0.0514845 0.0462867 0.0410886 0.03552 0.029942 0.0760459 0.0715307 0.0669765 0.0621835 0.0570852 0.0519031 0.0467066 0.0415054 0.0358567 0.0301612 0.0770777 0.0724468 0.0678221 0.0628627 0.0577175 0.0525224 0.0473068

388,0.0459511 , 389,0.0460155 , 390,0.0453922 , 391,0.0491573 , 392,0.0496202 , 393,0.049974 , 394,0.0505053 , 395,0.050887 , 396,0.0510886 , 397,0.0511606 , 398,0.051237 , 399,0.0504385 , 400,0.0546041 , 401,0.0547817 , 402,0.0551609 , 403,0.0556313 , 404,0.0560376 , 405,0.056301 , 406,0.0561574 , 407,0.0564581 , 408,0.0560631 , 409,0.0543058 , 410,0.058694 , 411,0.0633064 , 412,0.0749954 , 413,0.0699971 , 414,0.0649982 , 415,0.0599988 , 416,0.0599357 , 417,0.0601894 , 418,0.0606345 , 419,0.0607717 , 420,0.0610156 , 421,0.0611409 , 422,0.0611731 , 423,0.0614801 , 424,0.0676202 , 425,0.0719442 , 426,0.076306 , 427,0.066424 , 428,0.0712383 , 429,0.0760571 , 430,0.0749899 , 431,0.0699917 , 432,0.0649967 , 433,0.074983 , 434,0.0699849 , 435,0.0649983 , 436,0.0749727 , 437,0.0699751 , 438,0.0653074 , 439,0.0749561 , 440,0.0702522 , 441,0.0655331 , 442,0.0749296 ,

0.0420846 0.0363563 0.0305318 0.0735438 0.0685562 0.0633561 0.0582801 0.0531059 0.0479006 0.0426802 0.0370734 0.0312928 0.0748046 0.0695641 0.0643739 0.0591955 0.053992 0.0487611 0.0435158 0.0382622 0.0330352 0.0276458 0.028688 0.0291063 0.0749972 0.0750003 0.0750009 0.0749833 0.0699344 0.0648894 0.0598121 0.0546921 0.0495413 0.0443579 0.0391413 0.0339643 0.0295716 0.0293197 0.029008 0.0348155 0.0345775 0.0342656 0.0699945 0.0700006 0.0699991 0.0649904 0.0650003 0.064999 0.0599843 0.0599996 0.0600472 0.0549751 0.0550192 0.0550718 0.0499607

443,0.0703401 444,0.0657277 445,0.0751652 446,0.0703957 447,0.0657962 448,0.0751578 449,0.0703148 450,0.0657913

, , , , , , , ,

0.0500151 0.0500776 0.0448009 0.0449914 0.0450509 0.0396081 0.039945 0.0400013

** ** *ELEMENT, TYPE=CPE4R , ELSET=EALL 1, 3, 81, 80, 1 2, 4, 82, 81, 3 3, 5, 83, 82, 4 4, 6, 84, 83, 5 5, 7, 85, 84, 6 6, 8, 86, 85, 7 7, 9, 87, 86, 8 8, 10, 88, 87, 9 9, 11, 89, 88, 10 10, 12, 90, 89, 11 11, 13, 91, 90, 12 12, 14, 92, 91, 13 13, 15, 93, 92, 14 14, 16, 94, 93, 15 15, 17, 95, 94, 16 16, 18, 96, 95, 17 17, 19, 97, 96, 18 18, 20, 98, 97, 19 19, 21, 99, 98, 20 20, 22, 100, 99, 21 21, 24, 100, 22, 2 22, 44, 101, 42, 23 23, 45, 102, 101, 44 24, 46, 103, 102, 45 25, 47, 104, 103, 46 26, 48, 105, 104, 47 27, 49, 106, 105, 48 28, 50, 107, 106, 49 29, 51, 108, 107, 50 30, 52, 109, 108, 51 31, 53, 110, 109, 52 32, 54, 111, 110, 53 33, 55, 112, 111, 54 34, 56, 113, 112, 55 35, 57, 114, 113, 56 36, 58, 115, 114, 57 37, 59, 116, 115, 58 38, 60, 117, 116, 59 39, 43, 118, 117, 60 40, 62, 119, 118, 43 41, 63, 120, 119, 62 42, 61, 121, 120, 63 43, 64, 122, 121, 61 44, 65, 123, 122, 64

45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99,

66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 81, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 101, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 138, 102, 103, 104,

124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 172, 173, 174, 175,

123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 79, 100, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 41, 137, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 99, 154, 173, 174,

65 66 67 68 69 70 71 72 73 74 75 76 77 78 80 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 42 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 100 101 102 103

100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154,

105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 121, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 173, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 155, 156, 157, 158, 159, 160, 161, 162,

176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 205, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 220, 221, 222, 223, 224, 225, 226, 227,

175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 119, 172, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 153, 190, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 136, 220, 221, 222, 223, 224, 225, 226,

104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 120 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 154 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 137 155 156 157 158 159 160 161

155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 205, 206, 207, 208, 209,

163, 164, 165, 166, 167, 168, 169, 170, 171, 191, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 206, 207, 208, 209, 210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 221, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 204, 237,

228, 229, 230, 231, 232, 233, 234, 235, 236, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 278,

227, 228, 229, 230, 231, 232, 233, 234, 235, 171, 205, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 189, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 219, 236, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 204,

162 163 164 165 166 167 168 169 170 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 190 206 207 208 209 210 211 212 213 214 215 216 217 218 220 191 192 193 194 195 196 197 198 199 200 201 202 203 205

210, 211, 212, 213, 214, 215, 216, 217, 218, 219, 220, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264,

222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 266, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 253, 254, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 279, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 293, 280,

279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 317, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 328, 328, 329,

265, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 235, 278, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 251, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 264, 292, 318, 319, 320, 321, 322, 323, 324, 325, 326, 327, 277, 317,

221 222 223 224 225 226 227 228 229 230 231 232 233 234 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 252 253 254 255 256 257 258 259 260 261 262 263 265 266 267 268 269 270 271 272 273 274 275 276 278 279

265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319,

281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 318, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 329, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 330, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 331, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370,

330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 340, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 370, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380,

329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 291, 305, 341, 342, 343, 344, 345, 346, 347, 348, 349, 316, 304, 351, 352, 353, 354, 355, 356, 357, 358, 359, 350, 303, 361, 362, 363, 364, 365, 366, 367, 368, 369, 360, 302, 371, 372, 373, 374, 375, 376, 377, 378, 379,

280 281 282 283 284 285 286 287 288 289 290 292 306 307 308 309 310 311 312 313 314 315 317 305 341 342 343 344 345 346 347 348 349 329 304 351 352 353 354 355 356 357 358 359 330 303 361 362 363 364 365 366 367 368 369

320, 321, 322, 323, 324, 325, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 336, 337, 338, 339, 340, 341, 342, 343, 344, 345, 346, 347, 348, 349, 350, 351, 352, 353, 354, 355, 356, 357, 358, 359, 360, 361, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 373, 374,

332, 371, 372, 373, 374, 375, 376, 377, 378, 379, 380, 333, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 334, 391, 392, 393, 394, 395, 396, 397, 398, 399, 335, 409, 336, 337, 294, 295, 296, 297, 400, 401, 402, 403, 404, 405, 406, 407, 408, 410, 338, 339, 340, 319, 424,

380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 390, 299, 391, 392, 393, 394, 395, 396, 397, 398, 399, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409, 410, 410, 411, 412, 413, 414, 415, 415, 416, 417, 418, 419, 420, 421, 422, 423, 411, 424, 425, 426, 426, 427,

370, 301, 381, 382, 383, 384, 385, 386, 387, 388, 389, 380, 300, 299, 391, 392, 393, 394, 395, 396, 397, 398, 390, 298, 400, 401, 402, 403, 404, 405, 406, 407, 399, 408, 409, 410, 328, 412, 413, 414, 297, 415, 416, 417, 418, 419, 420, 421, 422, 423, 411, 424, 425, 340, 423,

331 302 371 372 373 374 375 376 377 378 379 332 301 381 382 383 384 385 386 387 388 389 333 299 391 392 393 394 395 396 397 398 334 399 335 336 293 294 295 296 298 400 401 402 403 404 405 406 407 408 337 338 339 318 411

375, 376, 377, 378, 379, 380, 381, 382, 383, 384, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 395, 396, 397, 398, 399, 400, 401, 402, 403, 404, 405, 406, 407, 408, 409,

425, 426, 320, 412, 413, 414, 416, 430, 431, 432, 417, 433, 434, 435, 418, 436, 437, 438, 419, 439, 440, 441, 420, 442, 443, 444, 421, 445, 446, 447, 422, 448, 449, 450, 450,

428, 429, 429, 430, 431, 432, 432, 433, 434, 435, 435, 436, 437, 438, 438, 439, 440, 441, 441, 442, 443, 444, 444, 445, 446, 447, 447, 448, 449, 450, 450, 429, 428, 427, 422,

427, 428, 426, 327, 430, 431, 414, 326, 433, 434, 432, 325, 436, 437, 435, 324, 439, 440, 438, 323, 442, 443, 441, 322, 445, 446, 444, 321, 448, 449, 447, 320, 429, 428, 423,

424 425 319 328 412 413 415 327 430 431 416 326 433 434 417 325 436 437 418 324 439 440 419 323 442 443 420 322 445 446 421 321 448 449 427

** ** BOT ** *NSET,NSET=NBOT 1, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,

18, 19, 20, 21, 22, ** ** RHS ** *NSET,NSET=NRHS 2, 2, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, *SOLID SECTION,ELSET=EALL,MATERIAL=RUBBER .10, *MATERIAL,NAME=RUBBER ** hyperelastic constants in Pa *HYPERELASTIC,POLYNOMIAL,N=1 3.2E6,.8E6 ** density is 1500 kg/m^3 *DENSITY 1500., ** ** RIGID ELEMENT USED TO COMPRESS THE BLOCK ** *NODE, NSET=NRIGID 20001,-.09683,-.00317 20002,.10317,.19683 20003,-.02,.12 *NSET,NSET=NREF 20003, *ELEMENT,TYPE=R2D2,ELSET=ERIGID 20001,20002,20001 *AMPLITUDE,NAME=CRUSH,DEFINITION=SMOOTH STEP 0.,0.,.1,1. *BOUNDARY NBOT,2

NRHS,1 NREF,6 ** *SURFACE,NAME=RIGID ERIGID,SPOS *SURFACE,NAME=SOLID EALL, *RIGID BODY,ELSET=ERIGID,REFNODE=20003 ** *ELSET,ELSET=QA_TEST_ALLELEMS EALL, ** *NSET, NSET=QA_TEST_ALLNODES QA_TEST_ALLNODES, NBOT, NRHS, NRIGID NREF, ** *STEP Compress the rubber block *DYNAMIC,EXPLICIT ,0.1 *CONTACT PAIR,INTERACTION=INTER SOLID,RIGID *SURFACE INTERACTION,NAME=INTER ** displace a total distance of .012 in both directions *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=CRUSH NREF,1,1,.00883 NREF,2,2,-.00883 ** *OUTPUT, FIELD, NUMBER INTERVAL=20, VARIABLE=PRESELECT *OUTPUT, HISTORY, FREQUENCY=1 *ENERGY OUTPUT ALLAE, ALLKE, ALLSE, ALLVD, ALLFD, ALLWK, ETOTAL ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *OUTPUT, FIELD, NUMBER INTERVAL=1, TIMEMARKS=YES *NODE OUTPUT, NSET=QA_TEST_ALLNODES U, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.1 ** *FILE OUTPUT,NUMBER INTERVAL=1 *NODE FILE U, *END STEP A.5 Blast loading on a stiffened plate blast_base.inp *HEADING Blast load on a flat plate with stiffeners S4R elements (20x20 mesh) Normal stiffeners (20x2) SI units (kg, m, s, N) ** ** Define corners and edges of plate **

*NODE, NSET=CORNERS 1,0.0, 0.0,0.0 21,0.0, 2.0, 0.0 801, 0.0, 0.0, 2.0 821, 0.0, 2.0, 2.0 *NGEN,NSET=EDGE1 1,21,1 *NGEN,NSET=EDGE2 801,821,1 *NFILL,NSET=NALL EDGE1,EDGE2,20,40 *NGEN,NSET=EDGE3 1,801,40 *NGEN,NSET=EDGE4 21,821,40 *NSET,NSET=EDGE EDGE1,EDGE2,EDGE3,EDGE4 ** ** Define a node set for output ** *NSET, NSET=NOUT 411 , ** ** Define stiffeners ** *NODE, NSET=STIFFN 4201,0.1, 0.0, 0.5 4221,0.1, 2.0, 0.5 4401,0.1, 0.0, 1.0 4421,0.1, 2.0, 1.0 4601,0.1, 0.0, 1.5 4621,0.1, 2.0, 1.5 **Generate inner edges of stiffeners *NGEN,NSET=ST1EDGE1 4201,4221,1 *NGEN,NSET=ST2EDGE1 4401,4421,1 *NGEN,NSET=ST3EDGE1 4601,4621,1 ** Define node sets on plate for outer edge of stiffeners *NSET, NSET=ST1EDGE2, GENERATE 201,221,1 *NSET, NSET=ST2EDGE2, GENERATE 401,421,1 *NSET, NSET=ST3EDGE2, GENERATE 601,621,1 *NFILL ST1EDGE2,ST1EDGE1,2,2000 *NFILL ST2EDGE2,ST2EDGE1,2,2000 *NFILL ST3EDGE2,ST3EDGE1,2,2000 ** ** Define master element type for plate (with associated nodes) ** and generate shell elements

*ELEMENT,TYPE=S4R,ELSET=PLATE 1,1,2,42,41 *ELGEN,ELSET=PLATE 1,20,1,1,20,40,20 ** Define master element type for stiffeners and generate *ELEMENT,TYPE=S4R,ELSET=STIFF1 1001, 201,2201,2202,202 *ELGEN,ELSET=STIFF1 1001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF2 2001, 401,2401,2402,402 *ELGEN,ELSET=STIFF2 2001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF3 3001, 601,2601,2602,602 *ELGEN,ELSET=STIFF3 3001,20,1,1,2,2000,20 ** Define elset for el history output ** Elements on central stiffener *ELSET, ELSET=STIFFMAX 2009, 2010, 2029, 2030 ** *SHELL SECTION,MATERIAL=STEEL,ELSET=PLATE,OFFSET=SPOS 0.025 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF1 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF2 0.0125, *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF3 0.0125, ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,.3 *PLASTIC 300.0E6, 0.000 350.0E6, 0.025 375.0E6, 0.100 394.0E6, 0.200 400.0E6, 0.350 *DENSITY 7800.0, ** *BOUNDARY EDGE, ENCASTRE ** Specify time variation for blast load *AMPLITUDE, NAME=BLAST 0.0, 0.0, 1.0E-03, 7.0E5, 10E-03, 7.0E05, 20E-03, 0.0, 50E-03, 0.0 *STEP Apply blast loading ** Explicit analysis with a time duration of 50 ms *DYNAMIC, EXPLICIT , 50E-03 ** Apply distributed load (surface pressure) to the element set PLATE *DLOAD, AMPLITUDE=BLAST

PLATE, P, 1.0 *MONITOR, NODE=411, DOF=1 *OUTPUT, FIELD, NUMBER INTERVAL=25 *ELEMENT OUTPUT S,PE *NODE OUTPUT U *OUTPUT, HISTORY, TIME INTERVAL=1.0E-4 *ELEMENT OUTPUT, ELSET=STIFFMAX PEEQ, MISES, ERV, *NODE OUTPUT, NSET=NOUT U,V *ENERGY OUTPUT ALLKE, ALLSE, ALLWK, ALLPD, ALLIE, ALLVD, ALLAE, ETOTAL ** *NSET,NSET=QA_TEST 370,371,372, 410,411,412, 450,451,452, *OUTPUT, FIELD, NUMBER INTERVAL=5, TIMEMARKS=YES *NODE OUTPUT,NSET=QA_TEST U *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=10E-03 *END STEP blast_damp.inp *HEADING Blast load on a flat plate with stiffeners S4R elements (20x20 mesh) Normal stiffeners (20x2) SI units (kg, m, s, N) Includes damping Longer analysis time (150ms) ** ** Define corners and edges of plate ** *NODE, NSET=CORNERS 1,0.0, 0.0,0.0 21,0.0, 2.0, 0.0 801, 0.0, 0.0, 2.0 821, 0.0, 2.0, 2.0 *NGEN,NSET=EDGE1 1,21,1 *NGEN,NSET=EDGE2 801,821,1 *NFILL,NSET=NALL EDGE1,EDGE2,20,40 *NGEN,NSET=EDGE3 1,801,40 *NGEN,NSET=EDGE4 21,821,40 *NSET,NSET=EDGE EDGE1,EDGE2,EDGE3,EDGE4 ** ** Define a node set for output **

*NSET, NSET=NOUT 411 , ** ** Define stiffeners ** *NODE, NSET=STIFFN 4201,0.1, 0.0, 0.5 4221,0.1, 2.0, 0.5 4401,0.1, 0.0, 1.0 4421,0.1, 2.0, 1.0 4601,0.1, 0.0, 1.5 4621,0.1, 2.0, 1.5 **Generate inner edges of stiffeners *NGEN,NSET=ST1EDGE1 4201,4221,1 *NGEN,NSET=ST2EDGE1 4401,4421,1 *NGEN,NSET=ST3EDGE1 4601,4621,1 ** Define node sets on plate for outer edge of stiffeners *NSET, NSET=ST1EDGE2, GENERATE 201,221,1 *NSET, NSET=ST2EDGE2, GENERATE 401,421,1 *NSET, NSET=ST3EDGE2, GENERATE 601,621,1 *NFILL ST1EDGE2,ST1EDGE1,2,2000 *NFILL ST2EDGE2,ST2EDGE1,2,2000 *NFILL ST3EDGE2,ST3EDGE1,2,2000 ** ** Define master element type for plate (with associated nodes) ** and generate shell elements *ELEMENT,TYPE=S4R,ELSET=PLATE 1,1,2,42,41 *ELGEN,ELSET=PLATE 1,20,1,1,20,40,20 ** Define master element type for stiffeners and generate *ELEMENT,TYPE=S4R,ELSET=STIFF1 1001, 201,2201,2202,202 *ELGEN,ELSET=STIFF1 1001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF2 2001, 401,2401,2402,402 *ELGEN,ELSET=STIFF2 2001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF3 3001, 601,2601,2602,602 *ELGEN,ELSET=STIFF3 3001,20,1,1,2,2000,20 ** Define elset for el history output ** Elements on central stiffener *ELSET, ELSET=STIFFMAX

2009, 2010, 2029, 2030 ** *SHELL SECTION,MATERIAL=STEEL,ELSET=PLATE,OFFSET=SPOS 0.025 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF1 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF2 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF3 0.0125 , ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,.3 *PLASTIC 300.0E6, 0.000 350.0E6, 0.025 375.0E6, 0.100 394.0E6, 0.200 400.0E6, 0.350 *DENSITY 7800.0, *DAMPING, ALPHA=50.0, BETA=0.0 ** *BOUNDARY EDGE, ENCASTRE ** Specify time variation for blast load *AMPLITUDE, NAME=BLAST 0.0, 0.0, 1.0E-03, 7.0E5, 10E-03, 7.0E05, 20E-03, 0.0, 50E-03, 0.0 *STEP Apply blast loading ** Explicit analysis with a time duration of 150 ms *DYNAMIC, EXPLICIT ,150E-03 ** Apply distributed load (surface pressure) to the element set PLATE *DLOAD, AMPLITUDE=BLAST PLATE, P, 1.0 *MONITOR, NODE=411, DOF=1 *OUTPUT, FIELD, NUMBER INTERVAL=25 *ELEMENT OUTPUT S,PE *NODE OUTPUT U *OUTPUT, HISTORY, TIME INTERVAL=1.0E-4 *ELEMENT OUTPUT, ELSET=STIFFMAX PEEQ, MISES, ERV, *NODE OUTPUT, NSET=NOUT U,V *ENERGY OUTPUT ALLKE, ALLSE, ALLWK, ALLPD, ALLIE, ALLVD, ALLAE, ETOTAL ** *NSET,NSET=QA_TEST 370,371,372, 410,411,412, 450,451,452,

*OUTPUT, FIELD, NUMBER INTERVAL=5, TIMEMARKS=YES *NODE OUTPUT,NSET=QA_TEST U *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=30E-03 *END STEP blast_long.inp *HEADING Blast load on a flat plate with stiffeners S4R elements (20x20 mesh) Normal stiffeners (20x2) SI units (kg, m, s, N) Longer analysis time to compare to blast_damp ** ** Define corners and edges of plate ** *NODE, NSET=CORNERS 1,0.0, 0.0,0.0 21,0.0, 2.0, 0.0 801, 0.0, 0.0, 2.0 821, 0.0, 2.0, 2.0 *NGEN,NSET=EDGE1 1,21,1 *NGEN,NSET=EDGE2 801,821,1 *NFILL,NSET=NALL EDGE1,EDGE2,20,40 *NGEN,NSET=EDGE3 1,801,40 *NGEN,NSET=EDGE4 21,821,40 *NSET,NSET=EDGE EDGE1,EDGE2,EDGE3,EDGE4 ** ** Define a node set for output ** *NSET, NSET=NOUT 411 , ** ** Define stiffeners ** *NODE, NSET=STIFFN 4201,0.1, 0.0, 0.5 4221,0.1, 2.0, 0.5 4401,0.1, 0.0, 1.0 4421,0.1, 2.0, 1.0 4601,0.1, 0.0, 1.5 4621,0.1, 2.0, 1.5 **Generate inner edges of stiffeners *NGEN,NSET=ST1EDGE1 4201,4221,1 *NGEN,NSET=ST2EDGE1 4401,4421,1 *NGEN,NSET=ST3EDGE1 4601,4621,1 ** Define node sets on plate for outer edge of stiffeners

*NSET, NSET=ST1EDGE2, GENERATE 201,221,1 *NSET, NSET=ST2EDGE2, GENERATE 401,421,1 *NSET, NSET=ST3EDGE2, GENERATE 601,621,1 *NFILL ST1EDGE2,ST1EDGE1,2,2000 *NFILL ST2EDGE2,ST2EDGE1,2,2000 *NFILL ST3EDGE2,ST3EDGE1,2,2000 ** ** Define master element type for plate (with associated nodes) ** and generate shell elements *ELEMENT,TYPE=S4R,ELSET=PLATE 1,1,2,42,41 *ELGEN,ELSET=PLATE 1,20,1,1,20,40,20 ** Define master element type for stiffeners and generate *ELEMENT,TYPE=S4R,ELSET=STIFF1 1001, 201,2201,2202,202 *ELGEN,ELSET=STIFF1 1001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF2 2001, 401,2401,2402,402 *ELGEN,ELSET=STIFF2 2001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF3 3001, 601,2601,2602,602 *ELGEN,ELSET=STIFF3 3001,20,1,1,2,2000,20 ** Define elset for el history output ** Elements on central stiffener *ELSET, ELSET=STIFFMAX 2009, 2010, 2029, 2030 ** *SHELL SECTION,MATERIAL=STEEL,ELSET=PLATE,OFFSET=SPOS 0.025 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF1 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF2 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF3 0.0125 , ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,.3 *PLASTIC 300.0E6, 0.000 350.0E6, 0.025 375.0E6, 0.100 394.0E6, 0.200 400.0E6, 0.350

*DENSITY 7800.0 , ** *BOUNDARY EDGE, ENCASTRE ** Specify time variation for blast load *AMPLITUDE, NAME=BLAST 0.0, 0.0, 1.0E-03, 7.0E5, 10E-03, 7.0E05, 20E-03, 0.0, 50E-03, 0.0 *STEP Apply blast loading ** Explicit analysis with a time duration of 150 ms *DYNAMIC, EXPLICIT ,150E-03 ** Apply distributed load (surface pressure) to the element set PLATE *DLOAD, AMPLITUDE=BLAST PLATE, P, 1.0 *MONITOR, NODE=411, DOF=1 *OUTPUT, FIELD, NUMBER INTERVAL=25 *ELEMENT OUTPUT S,PE *NODE OUTPUT U *OUTPUT, HISTORY, TIME INTERVAL=1.0E-4 *ELEMENT OUTPUT, ELSET=STIFFMAX PEEQ, MISES, ERV, *NODE OUTPUT, NSET=NOUT U,V *ENERGY OUTPUT ALLKE, ALLSE, ALLWK, ALLPD, ALLIE, ALLVD, ALLAE, ETOTAL ** *NSET,NSET=QA_TEST 370,371,372, 410,411,412, 450,451,452, *OUTPUT, FIELD, NUMBER INTERVAL=5, TIMEMARKS=YES *NODE OUTPUT,NSET=QA_TEST U *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=30E-03 *END STEP blast_refined.inp *HEADING Blast load on a flat plate with stiffeners S4R elements (20x20 mesh) Refined stiffeners (20x4) SI units (kg, m, s, N) ** ** Define corners and edges of plate ** *NODE, NSET=CORNERS 1,0.0, 0.0,0.0 21,0.0, 2.0, 0.0 801, 0.0, 0.0, 2.0 821, 0.0, 2.0, 2.0 *NGEN,NSET=EDGE1 1,21,1

*NGEN,NSET=EDGE2 801,821,1 *NFILL, NSET=NALL EDGE1,EDGE2,20,40 *NGEN,NSET=EDGE3 1,801,40 *NGEN,NSET=EDGE4 21,821,40 *NSET,NSET=EDGE EDGE1,EDGE2,EDGE3,EDGE4 ** ** Define a node set for output ** *NSET, NSET=NOUT 411 , ** ** define stiffeners ** *NODE, NSET=STIFFN 4201,0.1, 0.0, 0.5 4221,0.1, 2.0, 0.5 4401,0.1, 0.0, 1.0 4421,0.1, 2.0, 1.0 4601,0.1, 0.0, 1.5 4621,0.1, 2.0, 1.5 **Generate inner edges of stiffeners *NGEN,NSET=ST1EDGE1 4201,4221,1 *NGEN,NSET=ST2EDGE1 4401,4421,1 *NGEN,NSET=ST3EDGE1 4601,4621,1 ** Define node sets on plate for outer edge of stiffeners *NSET, NSET=ST1EDGE2, GENERATE 201,221,1 *NSET, NSET=ST2EDGE2, GENERATE 401,421,1 *NSET, NSET=ST3EDGE2, GENERATE 601,621,1 *NFILL ST1EDGE2,ST1EDGE1,4,1000 *NFILL ST2EDGE2,ST2EDGE1,4,1000 *NFILL ST3EDGE2,ST3EDGE1,4,1000 ** ** Define master element type for plate (with associated nodes) ** and generate shell elements *ELEMENT,TYPE=S4R,ELSET=PLATE 1,1,2,42,41 *ELGEN,ELSET=PLATE 1,20,1,1,20,40,20 ** Define master element type for stiffeners and generate *ELEMENT,TYPE=S4R,ELSET=STIFF1 1001, 201,1201,1202,202

*ELGEN,ELSET=STIFF1 1001,20,1,1,4,1000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF2 2001, 401,1401,1402,402 *ELGEN,ELSET=STIFF2 2001,20,1,1,4,1000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF3 3001, 601,1601,1602,602 *ELGEN,ELSET=STIFF3 3001,20,1,1,4,1000,20 ** Define elset for el history output ** Elements on central stiffener *ELSET, ELSET=STIFFMAX 2009, 2010, 2029,2030,2050,2051,2070,2071 ** *SHELL SECTION,MATERIAL=STEEL,ELSET=PLATE,OFFSET=SPOS 0.025 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF1 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF2 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF3 0.0125 , ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,.3 *PLASTIC 300.0E6, 0.000 350.0E6, 0.025 375.0E6, 0.100 394.0E6, 0.200 400.0E6, 0.350 *DENSITY 7800.0, ** *BOUNDARY EDGE, ENCASTRE ** Specify time variation for blast load *AMPLITUDE, NAME=BLAST 0.0, 0.0, 1.0E-03, 7.0E5, 10E-03, 7.0E05, 20E-03, 0.0, 50E-03, 0.0 *STEP Apply blast loading ** Explicit analysis with a time duration of 50 ms *DYNAMIC, EXPLICIT , 50E-03 ** Apply distributed load (surface pressure) to the element set PLATE *DLOAD, AMPLITUDE=BLAST PLATE, P, 1.0 *MONITOR, NODE=411, DOF=1 *OUTPUT, FIELD, NUMBER INTERVAL=25 *ELEMENT OUTPUT S,PE *NODE OUTPUT U

*OUTPUT, HISTORY, TIME INTERVAL=1.0E-4 *ELEMENT OUTPUT, ELSET=STIFFMAX PEEQ, MISES, ERV, *NODE OUTPUT, NSET=NOUT U,V *ENERGY OUTPUT ALLKE, ALLSE, ALLWK, ALLPD, ALLIE, ALLVD, ALLAE, ETOTAL ** *NSET,NSET=QA_TEST 370,371,372, 410,411,412, 450,451,452, *OUTPUT, FIELD, NUMBER INTERVAL=5, TIMEMARKS=YES *NODE OUTPUT,NSET=QA_TEST U *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=10E-03 *END STEP blast_rate.inp *HEADING Blast load on a flat plate with stiffeners S4R elements (20x20 mesh) Normal stiffeners (20x2) SI units (kg, m, s, N) Includes damping Includes rate-dependent plasticity ** ** Define corners and edges of plate ** *NODE, NSET=CORNERS 1,0.0, 0.0,0.0 21,0.0, 2.0, 0.0 801, 0.0, 0.0, 2.0 821, 0.0, 2.0, 2.0 *NGEN,NSET=EDGE1 1,21,1 *NGEN,NSET=EDGE2 801,821,1 *NFILL,NSET=NALL EDGE1,EDGE2,20,40 *NGEN,NSET=EDGE3 1,801,40 *NGEN,NSET=EDGE4 21,821,40 *NSET,NSET=EDGE EDGE1,EDGE2,EDGE3,EDGE4 ** ** Define a node set for output ** *NSET, NSET=NOUT 411 , ** ** Define stiffeners ** *NODE, NSET=STIFFN 4201,0.1, 0.0, 0.5

4221,0.1, 2.0, 0.5 4401,0.1, 0.0, 1.0 4421,0.1, 2.0, 1.0 4601,0.1, 0.0, 1.5 4621,0.1, 2.0, 1.5 **Generate inner edges of stiffeners *NGEN,NSET=ST1EDGE1 4201,4221,1 *NGEN,NSET=ST2EDGE1 4401,4421,1 *NGEN,NSET=ST3EDGE1 4601,4621,1 ** Define node sets on plate for outer edge of stiffeners *NSET, NSET=ST1EDGE2, GENERATE 201,221,1 *NSET, NSET=ST2EDGE2, GENERATE 401,421,1 *NSET, NSET=ST3EDGE2, GENERATE 601,621,1 *NFILL ST1EDGE2,ST1EDGE1,2,2000 *NFILL ST2EDGE2,ST2EDGE1,2,2000 *NFILL ST3EDGE2,ST3EDGE1,2,2000 ** ** Define master element type for plate (with associated nodes) ** and generate shell elements *ELEMENT,TYPE=S4R,ELSET=PLATE 1,1,2,42,41 *ELGEN,ELSET=PLATE 1,20,1,1,20,40,20 ** Define master element type for stiffeners and generate *ELEMENT,TYPE=S4R,ELSET=STIFF1 1001, 201,2201,2202,202 *ELGEN,ELSET=STIFF1 1001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF2 2001, 401,2401,2402,402 *ELGEN,ELSET=STIFF2 2001,20,1,1,2,2000,20 *ELEMENT,TYPE=S4R,ELSET=STIFF3 3001, 601,2601,2602,602 *ELGEN,ELSET=STIFF3 3001,20,1,1,2,2000,20 ** Define elset for el history output ** Elements on central stiffener *ELSET, ELSET=STIFFMAX 2009, 2010, 2029, 2030 ** *SHELL SECTION,MATERIAL=STEEL,ELSET=PLATE,OFFSET=SPOS 0.025 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF1 0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF2

0.0125 , *SHELL SECTION,MATERIAL=STEEL,ELSET=STIFF3 0.0125 , ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,.3 *PLASTIC 300.0E6, 0.000 350.0E6, 0.025 375.0E6, 0.100 394.0E6, 0.200 400.0E6, 0.350 *RATE DEPENDENT 40.0,5.0 *DENSITY 7800.0, *DAMPING, ALPHA=50.0, BETA=0.0 ** *BOUNDARY EDGE, ENCASTRE ** Specify time variation for blast load *AMPLITUDE, NAME=BLAST 0.0, 0.0, 1.0E-03, 7.0E5, 10E-03, 7.0E05, 20E-03, 0.0, 50E-03, 0.0 *STEP Apply blast loading ** Explicit analysis with a time duration of 50 ms *DYNAMIC, EXPLICIT , 50E-03 ** Apply distributed load (surface pressure) to the element set PLATE *DLOAD, AMPLITUDE=BLAST PLATE, P, 1.0 *MONITOR, NODE=411, DOF=1 *OUTPUT, FIELD, NUMBER INTERVAL=25 *ELEMENT OUTPUT S,PE *NODE OUTPUT U *OUTPUT, HISTORY, TIME INTERVAL=1.0E-4 *ELEMENT OUTPUT, ELSET=STIFFMAX PEEQ, MISES, ERV, *NODE OUTPUT, NSET=NOUT U,V *ENERGY OUTPUT ALLKE, ALLSE, ALLWK, ALLPD, ALLIE, ALLVD, ALLAE, ETOTAL ** *NSET,NSET=QA_TEST 370,371,372, 410,411,412, 450,451,452, *OUTPUT, FIELD, NUMBER INTERVAL=5, TIMEMARKS=YES *NODE OUTPUT,NSET=QA_TEST U *OUTPUT, HISTORY, VARIABLE=PRESELECT, TIME INTERVAL=10E-03 *END STEP

A.6 Circuit board drop test circuit.inp *HEADING Circuit board drop test 1.0 meter drop SI units (kg, m, s, N) *SYSTEM 0.,0.,0., .5,.707,.25 -.5,.707,-.5 *NODE 1, 0.005, -0.010, 0.012 11, 0.005, -0.010, 0.162 501, 0.105, -0.010, 0.012 511, 0.105, -0.010, 0.162 1001, 0., 0., 0. 1004, 0., -0.009, 0. 1005, 0., -0.011, 0. 1008, 0., -0.02, 0. 1061, 0., 0., 0.024 1064, 0., -0.009, 0.024 1065, 0., -0.011, 0.024 1068, 0., -0.02, 0.024 16001, 0.110, 0., 0. 16004, 0.110, -0.009, 0. 16005, 0.110, -0.011, 0. 16008, 0.110, -0.02, 0. 16061, 0.110, 0., 0.024 16064, 0.110, -0.009, 0.024 16065, 0.110, -0.011, 0.024 16068, 0.110, -0.02, 0.024 *NGEN, NSET=BOARDA 1,501,50 *NGEN, NSET=BOARDB 11,511,50 *NFILL,NSET=BOARD BOARDA, BOARDB, 10, 1 ** *NGEN, NSET=PACKAA 1001,1004,1 1005,1008,1 *NGEN, NSET=PACKAB 1061,1064,1 1065,1068,1 *NFILL,NSET=PACKA PACKAA,PACKAB,6,10 *NGEN, NSET=PACKBA 16001,16008,1 *NGEN, NSET=PACKBB 16061,16068,1 *NFILL,NSET=PACKB PACKBA,PACKBB,6,10 *NFILL,NSET=PACK PACKA,PACKB,15,1000 ** ** Reset coordinate system

** *SYSTEM *NODE 50000, -0.10, -0.10, -0.0001 50001, -0.10, 0.10, -0.0001 50002, 0.15, 0.10, -0.0001 50003, 0.15, -0.10, -0.0001 99999, 0., 0., -0.001 ** ** Element definitions ** *ELEMENT,TYPE=S4R,ELSET=BOARD 1,1,51,52,2 *ELGEN,ELSET=BOARD 1,10,50,1,10,1,10 *ELEMENT,TYPE=C3D8R,ELSET=PACK 500, 1001,1002,2002,2001, 1011,1012,2012,2011 530, 1031,1032,2032,2031, 1041,1042,2042,2041 534, 1035,1036,2036,2035, 1045,1046,2046,2045 *ELGEN,ELSET=PACK 500, 7,1,1, 3,10,10, 15,1000,100 530, 3,1,1, 3,10,10, 15,1000,100 534, 3,1,1, 3,10,10, 15,1000,100 *ELEMENT,TYPE=MASS,ELSET=CHIPS 6001,60 6002,357 6003,403 *ELEMENT,TYPE=R3D4,ELSET=FLOOR 50000, 50000,50003,50002,50001 ** ** Section properties ** *SOLID SECTION,ELSET=PACK,MATERIAL=FOAM *SHELL SECTION,ELSET=BOARD,MATERIAL=PCB,ORIENTATION=OR1 0.002, *ORIENTATION,NAME=OR1,SYSTEM=RECTANGULAR,DEFINITION=COORDINATES .5,.707,.25,-.5,.707,-.5 3,0 *MASS,ELSET=CHIPS 0.005, *RIGID BODY, ELSET=FLOOR, REFNODE=99999 *NSET,NSET=REF 99999, ** ** Materials ** *MATERIAL,NAME=PCB *ELASTIC 45.E9,0.3 *DENSITY 500., *MATERIAL,NAME=FOAM *ELASTIC 3.E6,0.0 *DENSITY

100., *FOAM 1.0,0.02E6,0.22E6, *FOAM HARDENING 0.20745E5,0.0 0.42916E5,0.2 0.75427E5,0.4 0.11738E6,0.6 0.16653E6,0.8 0.22000E6,1.0 0.24745E6,1.1 0.27494E6,1.2 0.30217E6,1.3 0.32890E6,1.4 0.35492E6,1.5 0.38006E6,1.6 0.40418E6,1.7 0.42720E6,1.8 0.44905E6,1.9 0.46969E6,2.0 0.50729E6,2.2 0.54008E6,2.4 0.56834E6,2.6 0.59247E6,2.8 0.61291E6,3.0 0.65083E6,3.5 0.67484E6,4.0 0.70810E6,6.0 0.71340E6,11.0 ** ** Element Faces for Contact ** *ELSET,GENERATE,ELSET=LOWBOARD 1,10,1 ** ** Define output locations and variables ** *NSET,NSET=CHIPS 60,357,403 *ELSET,ELSET=BOTPART 18, *ELSET,ELSET=PARTS 18,57,81 ** ** ** Define contact surfaces ** *SURFACE, NAME=LOWBOARD LOWBOARD, *SURFACE, NAME=PACK PACK, *SURFACE, NAME=FLOOR FLOOR,SPOS ** **

*BOUNDARY REF,ENCASTRE ** *INITIAL CONDITIONS,TYPE=VELOCITY BOARD,3,-4.43 PACK,3,-4.43 **************************** *STEP *DYNAMIC, EXPLICIT ,0.02 ** ** ** Define contact pairs ** *CONTACT PAIR,INTERACTION=FRIC,WEIGHT=0.5 PACK,LOWBOARD *CONTACT PAIR,INTERACTION=FRIC PACK,FLOOR ** ** Define surface interactions ** *SURFACE INTERACTION,NAME=FRIC *FRICTION 0.3, *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=20 *OUTPUT, HISTORY, TIME INTERVAL=0.1E-3 *NODE OUTPUT, NSET=CHIPS U, V, A *ELEMENT OUTPUT, ELSET=BOTPART 1,5 LE, S, SP, LEP *ENERGY OUTPUT ALLIE, ALLKE, ALLPD, ALLAE, ALLSE ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *FILE OUTPUT,NUMBER INTERVAL=3,TIMEMARKS=YES *ENERGY FILE *NODE FILE,NSET=CHIPS U, *END STEP A.7 Crushing of a tube tube_buckle.inp *HEADING Tube crush (x-direction) -- buckling analysis Tube dimensions: 0.4m x 0.1m x 0.1m SI units (kg, m, s, N) *NODE, NSET=QA_TEST_ALLNODES 33, 0.0 , 0.05, -0.05 833, 0.0 , 0.05, 0.05 1633, 0.0, -0.05, 0.05 2433, 0.0, -0.05, -0.05 3233, 0.0, 0.05, -0.05 ** *NODE, NSET=TOP

1, 0.4, 0.05, -0.05 101, 0.4, 0.05, -0.0375 201, 0.4, 0.05, -0.025 301, 0.4, 0.05, -0.0125 401, 0.4, 0.05, 0.0 501, 0.4, 0.05, 0.0125 601, 0.4, 0.05, 0.025 701, 0.4, 0.05, 0.0375 801, 0.4, 0.05, 0.05 ** 901, 0.4, 0.0375, 0.05 1001, 0.4, 0.025, 0.05 1101, 0.4, 0.0125, 0.05 1201, 0.4, 0.0, 0.05 1301, 0.4, -0.0125, 0.05 1401, 0.4, -0.025, 0.05 1501, 0.4, -0.0375, 0.05 1601, 0.4, -0.05, 0.05 ** 1701, 0.4, -0.05, 0.0375 1801, 0.4, -0.05, 0.025 1901, 0.4, -0.05, 0.0125 2001, 0.4, -0.05, 0.0 2101, 0.4, -0.05, -0.0125 2201, 0.4, -0.05, -0.025 2301, 0.4, -0.05, -0.0375 2401, 0.4, -0.05, -0.05 ** 2501, 0.4, -0.0375, -0.05 2601, 0.4, -0.025, -0.05 2701, 0.4, -0.0125, -0.05 2801, 0.4, 0.0, -0.05 2901, 0.4, 0.0125, -0.05 3001, 0.4, 0.025, -0.05 3101, 0.4, 0.0375, -0.05 3201, 0.4, 0.05, -0.05 ** *NGEN, NSET=BOT 33, 833, 100 833, 1633, 100 1633, 2433, 100 2433, 3233, 100 ** *NFILL, NSET=ALL TOP, BOT, 32, 1 ** ** SHELL ELEMENT DEFINITIONS ** *ELEMENT, TYPE=S4R, ELSET=YPOS 1, 1, 2, 102, 101 *ELGEN, ELSET=YPOS 1, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=ZPOS 801, 801,802,902,901

*ELGEN, ELSET=ZPOS 801, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=YNEG 1601, 1601, 1602, 1702, 1701 *ELGEN, ELSET=YNEG 1601, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=ZNEG 2401, 2401,2402,2502,2501 *ELGEN, ELSET=ZNEG 2401, 32, 1, 1, 7, 100, 100 *ELEMENT, TYPE=S4R, ELSET=ZNEG 3101, 3101, 3102, 2, 1 *ELGEN, ELSET=ZNEG 3101, 32, 1, 1 ** *ELSET, ELSET=TUBE ZPOS, ZNEG, YPOS, YNEG ** *SHELL SECTION,ELSET=TUBE,MATERIAL=STEEL 0.001,3 ** ** MATERIAL BEHAVIOR ** *MATERIAL,NAME=STEEL *ELASTIC 207.E9,.3 *PLASTIC 1.587E8,0.0 1.631E8,0.015 1.863E8,0.033 1.932E8,0.044 2.020E8,0.062 2.070E8,1.500 *DENSITY 7800., ** ** RIGID SURFACE ELEMENTS ** *NODE, NSET=QA_TEST_ALLNODES 20000, 0.41, 0.0, 0.0 20001, 0.4, -0.1, 0.1 20002, 0.4, 0.1, 0.1 20003, 0.4, 0.1,-0.1 20004, 0.4, -0.1,-0.1 ** 30000, -0.1, 0.0, 0.0 30001, 0.0, -0.1, 0.1 30002, 0.0, 0.1, 0.1 30003, 0.0, 0.1, -0.1 30004, 0.0, -0.1, -0.1 *NSET,NSET=REFTOP 20000, *NSET,NSET=REFBOT

30000, ** *ELEMENT,TYPE=R3D4,ELSET=RIGTOP 20001,20001,20002,20003,20004 ** *RIGID BODY,ELSET=RIGTOP,REFNODE=20000 ** *ELEMENT, TYPE=R3D4,ELSET=RIGBOT 30001,30001,30004,30003,30002 ** *RIGID BODY,ELSET=RIGBOT,REFNODE=30000 ** ** MASS FOR RIGID BODIES ** *ELEMENT,TYPE=MASS,ELSET=MASS 35001,30000 ** *MASS,ELSET=MASS 500., ** ** MISC NSETS AND ELSETS ** *NSET, NSET=PRIN TOP, REFTOP, BOT, REFBOT ** ** BOUNDARY CONDITIONS ** *BOUNDARY REFTOP, 1,6 REFBOT, 2,6 ** ** RIGID SURFACES ** *SURFACE, NAME=TOPSU RIGTOP, SPOS *SURFACE, NAME=BOTSU RIGBOT, SPOS ** ** IMPACT CONTACT ** *SURFACE, TYPE=NODE, NAME=TOP TOP, *CONTACT PAIR, ADJUST=.01, INTERACTION=TUBE TOP, TOPSU ** *SURFACE INTERACTION, NAME=TUBE ** ** BOTTOM CONTACT ** *SURFACE, TYPE=NODE, NAME=BOT BOT, ** the bottom end of the tube is fixed to the bottom rigid plate *TIE, NAME=FIXEDBOTTOM, POSITION TOLERANCE=0.01, ADJUST=YES, NO ROTATION BOT, BOTSU

** ** ** HISTORY DEFINITION ** *NSET, NSET=QA_TEST_ALLNODES QA_TEST_ALLNODES, TOP, BOT, ALL REFTOP, REFBOT, PRIN ** *STEP Buckling analysis of the tube *BUCKLE,EIGENSOLVER=SUBSPACE 10, *CLOAD REFBOT, 1, 500 *NODE FILE U, *NODE PRINT, FREQUENCY=0 *EL PRINT, FREQUENCY=0 *OUTPUT,FIELD,VARIABLE=PRESELECT,FREQUENCY=9999 *OUTPUT,FIELD *NODE OUTPUT U, *OUTPUT,HISTORY *NODE OUTPUT, NSET=QA_TEST_ALLNODES U, *END STEP tube_crush.inp *HEADING Tube crush (x-direction) -- crushing analysis Tube dimensions: 0.4m x 0.1m x 0.1m SI units (kg, m, s, N) *NODE 33, 0.0 , 0.05, -0.05 833, 0.0 , 0.05, 0.05 1633, 0.0, -0.05, 0.05 2433, 0.0, -0.05, -0.05 3233, 0.0, 0.05, -0.05 ** *NODE, NSET=TOP 1, 0.4, 0.05, -0.05 101, 0.4, 0.05, -0.0375 201, 0.4, 0.05, -0.025 301, 0.4, 0.05, -0.0125 401, 0.4, 0.05, 0.0 501, 0.4, 0.05, 0.0125 601, 0.4, 0.05, 0.025 701, 0.4, 0.05, 0.0375 801, 0.4, 0.05, 0.05 ** 901, 0.4, 0.0375, 0.05 1001, 0.4, 0.025, 0.05 1101, 0.4, 0.0125, 0.05 1201, 0.4, 0.0, 0.05 1301, 0.4, -0.0125, 0.05 1401, 0.4, -0.025, 0.05

1501, 0.4, -0.0375, 0.05 1601, 0.4, -0.05, 0.05 ** 1701, 0.4, -0.05, 0.0375 1801, 0.4, -0.05, 0.025 1901, 0.4, -0.05, 0.0125 2001, 0.4, -0.05, 0.0 2101, 0.4, -0.05, -0.0125 2201, 0.4, -0.05, -0.025 2301, 0.4, -0.05, -0.0375 2401, 0.4, -0.05, -0.05 ** 2501, 0.4, -0.0375, -0.05 2601, 0.4, -0.025, -0.05 2701, 0.4, -0.0125, -0.05 2801, 0.4, 0.0, -0.05 2901, 0.4, 0.0125, -0.05 3001, 0.4, 0.025, -0.05 3101, 0.4, 0.0375, -0.05 3201, 0.4, 0.05, -0.05 ** *NGEN, NSET=BOT 33, 833, 100 833, 1633, 100 1633, 2433, 100 2433, 3233, 100 ** *NFILL, NSET=ALL TOP, BOT, 32, 1 ** ** SHELL ELEMENT DEFINITIONS ** *ELEMENT, TYPE=S4R, ELSET=YPOS 1, 1, 2, 102, 101 *ELGEN, ELSET=YPOS 1, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=ZPOS 801, 801,802,902,901 *ELGEN, ELSET=ZPOS 801, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=YNEG 1601, 1601, 1602, 1702, 1701 *ELGEN, ELSET=YNEG 1601, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=ZNEG 2401, 2401,2402,2502,2501 *ELGEN, ELSET=ZNEG 2401, 32, 1, 1, 7, 100, 100 *ELEMENT, TYPE=S4R, ELSET=ZNEG 3101, 3101, 3102, 2, 1 *ELGEN, ELSET=ZNEG 3101, 32, 1, 1

** *ELSET, ELSET=TUBE ZPOS, ZNEG, YPOS, YNEG ** *SHELL SECTION,ELSET=TUBE,MATERIAL=STEEL 0.001,3 ** ** MATERIAL BEHAVIOR ** *MATERIAL,NAME=STEEL *ELASTIC 207.E9,.3 *PLASTIC 1.587E8,0.0 1.631E8,0.015 1.863E8,0.033 1.932E8,0.044 2.020E8,0.062 2.070E8,1.500 *DENSITY 7800., ** ** RIGID BODY ELEMENTS ** *NODE 20000, 0.41, 0.0, 0.0 20001, 0.4, -0.1, 0.1 20002, 0.4, 0.1, 0.1 20003, 0.4, 0.1,-0.1 20004, 0.4, -0.1,-0.1 ** 30000, -0.01, 0.0, 0.0 30001, 0.0, -0.1, 0.1 30002, 0.0, 0.1, 0.1 30003, 0.0, 0.1, -0.1 30004, 0.0, -0.1, -0.1 *NSET,NSET=REFTOP 20000, *NSET,NSET=REFBOT 30000, ** *ELEMENT,TYPE=R3D4,ELSET=RIGTOP 20001,20001,20002,20003,20004 ** *RIGID BODY,ELSET=RIGTOP,REFNODE=20000 ** *ELEMENT, TYPE=R3D4,ELSET=RIGBOT 30001,30001,30004,30003,30002 ** *RIGID BODY,ELSET=RIGBOT,TIE NSET=BOT,REFNODE=30000 ** ** MASS FOR RIGID BODIES ** *ELEMENT,TYPE=MASS,ELSET=MASS 35001,30000

** *MASS,ELSET=MASS 500., ** ** MISC NSETS AND ELSETS ** *NSET, NSET=PRIN TOP, REFTOP, BOT, REFBOT ** *INITIAL CONDITIONS,TYPE=VELOCITY ** 1 mph = 0.44701 m/s ** 20 mph = 8.9408 m/s REFBOT,1,8.9408 ALL,1,8.9408 ** ** BOUNDARY CONDITIONS ** *BOUNDARY REFTOP, 1,6 REFBOT, 2,6 ** ** IMPERFECTIONS BASED ON BUCKLING MODES ** *IMPERFECTION, FILE=tube_buckle, STEP=1 1, 2.0E-5 2, 0.8E-5 3, 0.4E-5 4, 0.18E-5 5, 0.16E-5 6, 0.10E-5 7, 0.10E-5 8, 0.08E-5 9, 0.02E-5 10, 0.02E-5 ** ** *SURFACE, NAME=TOPSU RIGTOP, SPOS *SURFACE, NAME=TUBE TUBE, ** ** HISTORY DEFINITON ** *STEP Impact of square tube with free deceleration Initial velocity=8.9408m/s (20mph) *DYNAMIC, EXPLICIT ,0.03 ** ** SELF CONTACT ** *CONTACT PAIR, INTERACTION=SELF TUBE, *SURFACE INTERACTION, NAME=SELF *FRICTION

0.1, ** ** IMPACT CONTACT ** ** faces of the rail (both inside and out) near the impact wall ** may contact the wall *CONTACT PAIR, INTERACTION=IMPACT TUBE, TOPSU *SURFACE INTERACTION, NAME=IMPACT ** ** *OUTPUT, HISTORY, TIME INTERVAL=0.00008 *ENERGY OUTPUT ALLWK, ALLKE, ETOTAL, ALLPD, ALLIE, ALLVD, ALLAE, ALLSE *NODE OUTPUT, NSET=PRIN U, V, A, RF1, RF2, RF3 *MONITOR, NODE=30000, DOF=1 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=20 ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *FILE OUTPUT, NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U, *END STEP tube_crush_noimper.inp *HEADING Tube crush (x-direction) -- crushing analysis Tube dimensions: 0.4m x 0.1m x 0.1m SI units (kg, m, s, N) *NODE 33, 0.0 , 0.05, -0.05 833, 0.0 , 0.05, 0.05 1633, 0.0, -0.05, 0.05 2433, 0.0, -0.05, -0.05 3233, 0.0, 0.05, -0.05 ** *NODE, NSET=TOP 1, 0.4, 0.05, -0.05 101, 0.4, 0.05, -0.0375 201, 0.4, 0.05, -0.025 301, 0.4, 0.05, -0.0125 401, 0.4, 0.05, 0.0 501, 0.4, 0.05, 0.0125 601, 0.4, 0.05, 0.025 701, 0.4, 0.05, 0.0375 801, 0.4, 0.05, 0.05 ** 901, 0.4, 0.0375, 0.05 1001, 0.4, 0.025, 0.05 1101, 0.4, 0.0125, 0.05 1201, 0.4, 0.0, 0.05 1301, 0.4, -0.0125, 0.05 1401, 0.4, -0.025, 0.05

1501, 0.4, -0.0375, 0.05 1601, 0.4, -0.05, 0.05 ** 1701, 0.4, -0.05, 0.0375 1801, 0.4, -0.05, 0.025 1901, 0.4, -0.05, 0.0125 2001, 0.4, -0.05, 0.0 2101, 0.4, -0.05, -0.0125 2201, 0.4, -0.05, -0.025 2301, 0.4, -0.05, -0.0375 2401, 0.4, -0.05, -0.05 ** 2501, 0.4, -0.0375, -0.05 2601, 0.4, -0.025, -0.05 2701, 0.4, -0.0125, -0.05 2801, 0.4, 0.0, -0.05 2901, 0.4, 0.0125, -0.05 3001, 0.4, 0.025, -0.05 3101, 0.4, 0.0375, -0.05 3201, 0.4, 0.05, -0.05 ** *NGEN, NSET=BOT 33, 833, 100 833, 1633, 100 1633, 2433, 100 2433, 3233, 100 ** *NFILL, NSET=ALL TOP, BOT, 32, 1 ** ** SHELL ELEMENT DEFINITIONS ** *ELEMENT, TYPE=S4R, ELSET=YPOS 1, 1, 2, 102, 101 *ELGEN, ELSET=YPOS 1, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=ZPOS 801, 801,802,902,901 *ELGEN, ELSET=ZPOS 801, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=YNEG 1601, 1601, 1602, 1702, 1701 *ELGEN, ELSET=YNEG 1601, 32, 1, 1, 8, 100, 100 ** *ELEMENT, TYPE=S4R, ELSET=ZNEG 2401, 2401,2402,2502,2501 *ELGEN, ELSET=ZNEG 2401, 32, 1, 1, 7, 100, 100 *ELEMENT, TYPE=S4R, ELSET=ZNEG 3101, 3101, 3102, 2, 1 *ELGEN, ELSET=ZNEG 3101, 32, 1, 1

** *ELSET, ELSET=TUBE ZPOS, ZNEG, YPOS, YNEG ** *SHELL SECTION,ELSET=TUBE,MATERIAL=STEEL 0.001,3 ** ** MATERIAL BEHAVIOR ** *MATERIAL,NAME=STEEL *ELASTIC 207.E9,.3 *PLASTIC 1.587E8,0.0 1.631E8,0.015 1.863E8,0.033 1.932E8,0.044 2.020E8,0.062 2.070E8,1.500 *DENSITY 7800., ** ** RIGID BODY ELEMENTS ** *NODE 20000, 0.41, 0.0, 0.0 20001, 0.4, -0.1, 0.1 20002, 0.4, 0.1, 0.1 20003, 0.4, 0.1,-0.1 20004, 0.4, -0.1,-0.1 ** 30000, -0.01, 0.0, 0.0 30001, 0.0, -0.1, 0.1 30002, 0.0, 0.1, 0.1 30003, 0.0, 0.1, -0.1 30004, 0.0, -0.1, -0.1 *NSET,NSET=REFTOP 20000, *NSET,NSET=REFBOT 30000, ** *ELEMENT,TYPE=R3D4,ELSET=RIGTOP 20001,20001,20002,20003,20004 ** *RIGID BODY,ELSET=RIGTOP,REFNODE=20000 ** *ELEMENT, TYPE=R3D4,ELSET=RIGBOT 30001,30001,30004,30003,30002 ** *RIGID BODY,ELSET=RIGBOT,TIE NSET=BOT,REFNODE=30000 ** ** MASS FOR RIGID BODIES ** *ELEMENT,TYPE=MASS,ELSET=MASS 35001,30000

** *MASS,ELSET=MASS 500., ** ** MISC NSETS AND ELSETS ** *NSET, NSET=PRIN TOP, REFTOP, BOT, REFBOT ** *INITIAL CONDITIONS,TYPE=VELOCITY ** 1 mph = 0.44701 m/s ** 20 mph = 8.9408 m/s REFBOT,1,8.9408 ALL,1,8.9408 ** ** BOUNDARY CONDITIONS ** *BOUNDARY REFTOP, 1,6 REFBOT, 2,6 ** ** *SURFACE, NAME=TOPSU RIGTOP, SPOS *SURFACE, NAME=TUBE TUBE, ** ** HISTORY DEFINITON ** *STEP Impact of square tube with free deceleration Initial velocity=8.9408m/s (20mph) *DYNAMIC, EXPLICIT ,0.03 ** ** SELF CONTACT ** *CONTACT PAIR, INTERACTION=SELF TUBE, *SURFACE INTERACTION, NAME=SELF *FRICTION 0.1, ** ** IMPACT CONTACT ** ** faces of the rail (both inside and out) near the impact wall may ** contact the wall *CONTACT PAIR, INTERACTION=IMPACT TUBE, TOPSU *SURFACE INTERACTION, NAME=IMPACT ** ** *OUTPUT, HISTORY, TIME INTERVAL=0.00008 *ENERGY OUTPUT ALLWK, ALLKE, ETOTAL, ALLPD, ALLIE, ALLVD, ALLAE, ALLSE

*NODE OUTPUT, NSET=PRIN U, V, A, RF1, RF2, RF3 *MONITOR, NODE=30000, DOF=1 *OUTPUT, FIELD, VARIABLE=PRESELECT, NUMBER INTERVAL=20 ********************************** ** OUTPUT FOR HKS QA PURPOSES ********************************** *FILE OUTPUT, NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U, *END STEP A.8 Deep drawing of a can bottom draw_freq_std.inp *HEADING Deep drawing of a can bottom -- frequency analysis ABAQUS/Standard SI units (kg, m, s, N) ** ** BLANK ** *NODE,NSET=BLANK 501,.000,.000 601,.045,.000 *NGEN,NSET=BLANK 501,601,1 *NSET,NSET=BSYM 501, *NSET,NSET=BEND 601, *ELEMENT,TYPE=SAX1,ELSET=BLANK 501,501,503 *ELGEN,ELSET=BLANK 501,50,2,2 *SHELL SECTION,ELSET=BLANK,MATERIAL=STEEL, SECTION INTEGRATION=GAUSS .0005,5 *MATERIAL,NAME=STEEL *ELASTIC 210.0E9, 0.3 *DENSITY 7800., *BOUNDARY BSYM,1 BSYM,6 BEND,2 *STEP *FREQUENCY 1, *RESTART,WRITE *EL PRINT,FREQ=0 *NODE PRINT,FREQ=0 *NODE FILE,FREQ=999 U, *OUTPUT, FIELD

*NODE OUTPUT, NSET=BLANK U, *END STEP draw_stage1_attempt1.inp *HEADING Deep drawing of a can bottom -- stage 1, attempt 1 SI units (kg, m, s, N) ** *NODE,NSET=REFP1 1001,0.,0.024 *NODE,NSET=REFD1 1101,0.050,-0.012 ** ** BLANK ** *NODE,NSET=BLANK 501,.000,.000 601,.045,.000 *NGEN,NSET=BLANK 501,601,1 *ELEMENT,TYPE=SAX1,ELSET=BLANK 501,501,503 *ELGEN,ELSET=BLANK 501,50,2,2 *SHELL SECTION,ELSET=BLANK,MATERIAL=STEEL, SECTION INTEGRATION=GAUSS .0005,5 *NSET,NSET=BSYM 501, *NSET,NSET=NHIST 521,601 ** ** PUNCH1 -- PUNCH FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=PUNCH1, TYPE=SEGMENTS START,0.032,0.03025 LINE,0.032,0.00625 CIRCL,0.026,0.00025,0.026,0.00625 LINE,0.,0.00025 *RIGID BODY, REF NODE=1001, ANALYTICAL SURFACE=PUNCH1 ** ** DIE1 -- DIE FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=DIE1, TYPE=SEGMENTS START,0.0326,-0.02825 LINE,0.0326,-0.00425 CIRCL,0.0366,-0.00025,0.0366,-0.00425 LINE,0.050,-0.00025 *RIGID BODY, REF NODE=1101, ANALYTICAL SURFACE=DIE1 ** ** SURFACE DEFINITIONS ** *SURFACE, NAME=BLANK_TOP BLANK,SPOS *SURFACE, NAME=BLANK_BOT

BLANK,SNEG ** ** BLANK MATERIAL ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,0.3 *DENSITY 7800., *PLASTIC 0.912E8, 0.0 0.131E9, 0.159E-02 0.171E9, 0.649E-02 0.211E9, 0.177E-01 0.251E9, 0.395E-01 0.291E9, 0.776E-01 0.331E9, 0.139 0.391E9, 0.295 ** ** AMPLITUDE DEFINITIONS ** *AMPLITUDE,NAME=FORM1 0.,0.,.033,1. ** ** INITIAL FIXED BOUNDARY CONDITIONS ** *BOUNDARY REFP1,1 REFP1,6 REFD1,1,6 BSYM,1 BSYM,6 ** ** STEP 1 -- FIRST FORMING STAGE ** ** *ELSET,ELSET=QA_TEST_ALLELEMS BLANK, ** *NSET, NSET=QA_TEST_ALLNODES REFP1, REFD1, BLANK, BLANK BSYM, NHIST ** *STEP *DYNAMIC,EXPLICIT ,.033 ** ** Contact Pair Definition ** *CONTACT PAIR,INTERACTION=INTER PUNCH1,BLANK_TOP DIE1,BLANK_BOT *SURFACE INTERACTION,NAME=INTER *FRICTION .1,

*BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=FORM1 REFP1,2,2,-.015 *RESTART,WRITE,NUMBER INTERVAL=10 *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V *ENERGY OUTPUT ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER=4 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U,RF ** ** Output for HKS QA purposes ** *FILE OUTPUT,NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U,RF *EL FILE S, *OUTPUT, FIELD, NUMBER INTERVAL=1 *NODE OUTPUT, NSET=QA_TEST_ALLNODES U,RF *ELEMENT OUTPUT, ELSET=QA_TEST_ALLELEMS S, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.033 ***OUTPUT,HISTORY,VARIABLE=PRESELECT *END STEP draw_spring1_std.inp *HEADING Deep drawing of a can bottom -- first springback ABAQUS/Standard springback following draw_stage1_attempt3_sqrt10.inp SI units (kg, m, s, N) *IMPORT,STEP=1,INTERVAL=10,STATE=YES,UPDATE=NO BLANK, *IMPORT ELSET BLANK, *IMPORT NSET BSYM,NHIST,BLANK *BOUNDARY BSYM,1 BSYM,6 *RESTART,WRITE,FREQ=999 *STEP,NLGEOM *STATIC .1,1. *BOUNDARY,FIXED BSYM,2,2 *OUTPUT, FIELD, FREQUENCY=10 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT

U, ** ** Output for HKS QA purposes ** *NODE FILE,FREQ=999 U, *EL FILE,FREQ=999 S, *END STEP draw_stage1_attempt2.inp *HEADING Deep drawing of a can bottom -- stage 1, attempt 2 SI units (kg, m, s, N) ** *NODE,NSET=REFP1 1001,0.,0.024 *NODE,NSET=REFD1 1101,0.050,-0.012 ** ** BLANK ** *NODE,NSET=BLANK 501,.000,.000 601,.045,.000 *NGEN,NSET=BLANK 501,601,1 *ELEMENT,TYPE=SAX1,ELSET=BLANK 501,501,503 *ELGEN,ELSET=BLANK 501,50,2,2 *SHELL SECTION,ELSET=BLANK,MATERIAL=STEEL,SECTION INTEGRATION=GAUSS .0005,5 *NSET,NSET=BSYM 501, *NSET,NSET=NHIST 521,601 ** ** PUNCH1 -- PUNCH FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=PUNCH1, TYPE=SEGMENTS START,0.032,0.03025 LINE,0.032,0.00625 CIRCL,0.026,0.00025,0.026,0.00625 LINE,0.,0.00025 *RIGID BODY, REF NODE=1001, ANALYTICAL SURFACE=PUNCH1 ** ** DIE1 -- DIE FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=DIE1, TYPE=SEGMENTS START,0.0326,-0.02825 LINE,0.0326,-0.00425 CIRCL,0.0366,-0.00025,0.0366,-0.00425 LINE,0.050,-0.00025 *RIGID BODY, REF NODE=1101, ANALYTICAL SURFACE=DIE1 **

** SURFACE DEFINITIONS ** *SURFACE, NAME=BLANK_TOP BLANK,SPOS *SURFACE, NAME=BLANK_BOT BLANK,SNEG ** ** BLANK MATERIAL ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,0.3 *DENSITY 7800., *PLASTIC 0.912E8, 0.0 0.131E9, 0.159E-02 0.171E9, 0.649E-02 0.211E9, 0.177E-01 0.251E9, 0.395E-01 0.291E9, 0.776E-01 0.331E9, 0.139 0.391E9, 0.295 ** ** AMPLITUDE DEFINITIONS ** *AMPLITUDE,NAME=FORM1, DEFINITION=SMOOTH STEP 0.,0.,.033,1. ** ** INITIAL FIXED BOUNDARY CONDITIONS ** *BOUNDARY REFP1,1 REFP1,6 REFD1,1,6 BSYM,1 BSYM,6 ** ** STEP 1 -- FIRST FORMING STAGE ** ** *ELSET,ELSET=QA_TEST_ALLELEMS BLANK, ** *NSET, NSET=QA_TEST_ALLNODES REFP1, REFD1, BLANK, BLANK BSYM, NHIST ** *STEP *DYNAMIC,EXPLICIT ,.033 ** ** Contact Pair Definition ** *CONTACT PAIR,INTERACTION=INTER

PUNCH1,BLANK_TOP DIE1,BLANK_BOT *SURFACE INTERACTION,NAME=INTER *FRICTION .1, *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=FORM1 REFP1,2,2,-.015 *RESTART,WRITE,NUMBER INTERVAL=10 *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V *ENERGY OUTPUT ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER=4 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U,RF ** ** Output for HKS QA purposes ** *FILE OUTPUT,NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U,RF *EL FILE S, *OUTPUT, FIELD, NUMBER INTERVAL=1 *NODE OUTPUT, NSET=QA_TEST_ALLNODES U,RF *ELEMENT OUTPUT, ELSET=QA_TEST_ALLELEMS S, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.033 ***OUTPUT,HISTORY,VARIABLE=PRESELECT *END STEP draw_stage1_attempt3.inp *HEADING Deep drawing of a can bottom -- stage 1, attempt 3 SI units (kg, m, s, N) ** *NODE,NSET=REFP1 1001,0.,0.024 *NODE,NSET=REFD1 1101,0.050,-0.012 ** ** BLANK ** *NODE,NSET=BLANK 501,.000,.000 601,.045,.000 *NGEN,NSET=BLANK 501,601,1 *ELEMENT,TYPE=SAX1,ELSET=BLANK 501,501,503 *ELGEN,ELSET=BLANK

501,50,2,2 *SHELL SECTION,ELSET=BLANK,MATERIAL=STEEL,SECTION INTEGRATION=GAUSS .0005,5 *NSET,NSET=BSYM 501, *NSET,NSET=NHIST 521,601 ** ** PUNCH1 -- PUNCH FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=PUNCH1, TYPE=SEGMENTS START,0.032,0.03025 LINE,0.032,0.00625 CIRCL,0.026,0.00025,0.026,0.00625 LINE,0.,0.00025 *RIGID BODY, REF NODE=1001, ANALYTICAL SURFACE=PUNCH1 ** ** DIE1 -- DIE FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=DIE1, TYPE=SEGMENTS START,0.0326,-0.02825 LINE,0.0326,-0.00425 CIRCL,0.0366,-0.00025,0.0366,-0.00425 LINE,0.050,-0.00025 *RIGID BODY, REF NODE=1101, ANALYTICAL SURFACE=DIE1 ** ** SURFACE DEFINITIONS ** *SURFACE, NAME=BLANK_TOP BLANK,SPOS *SURFACE, NAME=BLANK_BOT BLANK,SNEG ** ** BLANK MATERIAL ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,0.3 *DENSITY 7800., *DAMPING,BETA=1.E-7 *PLASTIC 0.912E8, 0.0 0.131E9, 0.159E-02 0.171E9, 0.649E-02 0.211E9, 0.177E-01 0.251E9, 0.395E-01 0.291E9, 0.776E-01 0.331E9, 0.139 0.391E9, 0.295 ** ** AMPLITUDE DEFINITIONS ** *AMPLITUDE,NAME=FORM1, DEFINITION=SMOOTH STEP 0.,0.,.033,1.

** ** INITIAL FIXED BOUNDARY CONDITIONS ** *BOUNDARY REFP1,1 REFP1,6 REFD1,1,6 BSYM,1 BSYM,6 ** ** STEP 1 -- FIRST FORMING STAGE ** *STEP *DYNAMIC,EXPLICIT ,.033 ** ** Contact Pair Definition ** *CONTACT PAIR,INTERACTION=INTER PUNCH1,BLANK_TOP DIE1,BLANK_BOT *SURFACE INTERACTION,NAME=INTER *FRICTION .1, *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=FORM1 REFP1,2,2,-.015 *RESTART,WRITE,NUMBER INTERVAL=10 *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V *ENERGY OUTPUT ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER=4 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U,RF ** ** Output for HKS QA purposes ** *FILE OUTPUT,NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U,RF *EL FILE S, *END STEP draw_stage1_attempt3_sqrt10.inp *HEADING Deep drawing of a can bottom -- stage 1, attempt 3 Mass scaling factor = 10 (speedup of sqrt10) SI units (kg, m, s, N) ** *NODE,NSET=REFP1 1001,0.,0.024

*NODE,NSET=REFD1 1101,0.050,-0.012 ** ** BLANK ** *NODE,NSET=BLANK 501,.000,.000 601,.045,.000 *NGEN,NSET=BLANK 501,601,1 *ELEMENT,TYPE=SAX1,ELSET=BLANK 501,501,503 *ELGEN,ELSET=BLANK 501,50,2,2 *SHELL SECTION,ELSET=BLANK,MATERIAL=STEEL,SECTION INTEGRATION=GAUSS .0005,5 *NSET,NSET=BSYM 501, *NSET,NSET=NHIST 521,601 ** ** PUNCH1 -- PUNCH FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=PUNCH1, TYPE=SEGMENTS START,0.032,0.03025 LINE,0.032,0.00625 CIRCL,0.026,0.00025,0.026,0.00625 LINE,0.,0.00025 *RIGID BODY, REF NODE=1001, ANALYTICAL SURFACE=PUNCH1 ** ** DIE1 -- DIE FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=DIE1, TYPE=SEGMENTS START,0.0326,-0.02825 LINE,0.0326,-0.00425 CIRCL,0.0366,-0.00025,0.0366,-0.00425 LINE,0.050,-0.00025 *RIGID BODY, REF NODE=1101, ANALYTICAL SURFACE=DIE1 ** ** SURFACE DEFINITIONS ** *SURFACE, NAME=BLANK_TOP BLANK,SPOS *SURFACE, NAME=BLANK_BOT BLANK,SNEG ** ** BLANK MATERIAL ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,0.3 *DENSITY 7800., *DAMPING,BETA=1.E-7 *PLASTIC

0.912E8, 0.0 0.131E9, 0.159E-02 0.171E9, 0.649E-02 0.211E9, 0.177E-01 0.251E9, 0.395E-01 0.291E9, 0.776E-01 0.331E9, 0.139 0.391E9, 0.295 ** ** AMPLITUDE DEFINITIONS ** *AMPLITUDE,NAME=FORM1, DEFINITION=SMOOTH STEP 0.,0.,.033,1. ** ** INITIAL FIXED BOUNDARY CONDITIONS ** *BOUNDARY REFP1,1 REFP1,6 REFD1,1,6 BSYM,1 BSYM,6 ** ** STEP 1 -- FIRST FORMING STAGE ** ** *ELSET,ELSET=QA_TEST_ALLELEMS BLANK, ** *NSET, NSET=QA_TEST_ALLNODES REFP1, REFD1, BLANK, BLANK BSYM, NHIST ** *STEP *DYNAMIC,EXPLICIT ,.033 *FIXED MASS SCALING, FACTOR=10 ** ** Contact Pair Definition ** *CONTACT PAIR,INTERACTION=INTER PUNCH1,BLANK_TOP DIE1,BLANK_BOT *SURFACE INTERACTION,NAME=INTER *FRICTION .1, *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=FORM1 REFP1,2,2,-.015 *RESTART,WRITE,NUMBER INTERVAL=10 *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V *ENERGY OUTPUT ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER=4

*ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U,RF ** ** Output for HKS QA purposes ** *FILE OUTPUT,NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U,RF *EL FILE S, *OUTPUT, FIELD, NUMBER INTERVAL=1 *NODE OUTPUT, NSET=QA_TEST_ALLNODES U,RF *ELEMENT OUTPUT, ELSET=QA_TEST_ALLELEMS S, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.033 ***OUTPUT,HISTORY,VARIABLE=PRESELECT *END STEP draw_stage1_attempt3_10.inp *HEADING Deep drawing of a can bottom -- stage 1, attempt 3 Mass scaling factor = 100 (speedup of 10) SI units (kg, m, s, N) ** *NODE,NSET=REFP1 1001,0.,0.024 *NODE,NSET=REFD1 1101,0.050,-0.012 ** ** BLANK ** *NODE,NSET=BLANK 501,.000,.000 601,.045,.000 *NGEN,NSET=BLANK 501,601,1 *ELEMENT,TYPE=SAX1,ELSET=BLANK 501,501,503 *ELGEN,ELSET=BLANK 501,50,2,2 *SHELL SECTION,ELSET=BLANK,MATERIAL=STEEL,SECTION INTEGRATION=GAUSS .0005,5 *NSET,NSET=BSYM 501, *NSET,NSET=NHIST 521,601 ** ** PUNCH1 -- PUNCH FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=PUNCH1, TYPE=SEGMENTS START,0.032,0.03025 LINE,0.032,0.00625

CIRCL,0.026,0.00025,0.026,0.00625 LINE,0.,0.00025 *RIGID BODY, REF NODE=1001, ANALYTICAL SURFACE=PUNCH1 ** ** DIE1 -- DIE FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=DIE1, TYPE=SEGMENTS START,0.0326,-0.02825 LINE,0.0326,-0.00425 CIRCL,0.0366,-0.00025,0.0366,-0.00425 LINE,0.050,-0.00025 *RIGID BODY, REF NODE=1101, ANALYTICAL SURFACE=DIE1 ** ** SURFACE DEFINITIONS ** *SURFACE, NAME=BLANK_TOP BLANK,SPOS *SURFACE, NAME=BLANK_BOT BLANK,SNEG ** ** BLANK MATERIAL ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,0.3 *DENSITY 7800., *DAMPING,BETA=1.E-7 *PLASTIC 0.912E8, 0.0 0.131E9, 0.159E-02 0.171E9, 0.649E-02 0.211E9, 0.177E-01 0.251E9, 0.395E-01 0.291E9, 0.776E-01 0.331E9, 0.139 0.391E9, 0.295 ** ** AMPLITUDE DEFINITIONS ** *AMPLITUDE,NAME=FORM1, DEFINITION=SMOOTH STEP 0.,0.,.033,1. ** ** INITIAL FIXED BOUNDARY CONDITIONS ** *BOUNDARY REFP1,1 REFP1,6 REFD1,1,6 BSYM,1 BSYM,6 ** ** STEP 1 -- FIRST FORMING STAGE ** **

*ELSET,ELSET=QA_TEST_ALLELEMS BLANK, ** *NSET, NSET=QA_TEST_ALLNODES REFP1, REFD1, BLANK, BLANK BSYM, NHIST ** *STEP *DYNAMIC,EXPLICIT ,.033 *FIXED MASS SCALING, FACTOR=100 ** ** Contact Pair Definition ** *CONTACT PAIR,INTERACTION=INTER PUNCH1,BLANK_TOP DIE1,BLANK_BOT *SURFACE INTERACTION,NAME=INTER *FRICTION .1, *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=FORM1 REFP1,2,2,-.015 *RESTART,WRITE,NUMBER INTERVAL=10 *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V *ENERGY OUTPUT ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER=4 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U,RF ** ** Output for HKS QA purposes ** *FILE OUTPUT,NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U,RF *EL FILE S, *OUTPUT, FIELD, NUMBER INTERVAL=1 *NODE OUTPUT, NSET=QA_TEST_ALLNODES U,RF *ELEMENT OUTPUT, ELSET=QA_TEST_ALLELEMS S, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.033 ***OUTPUT,HISTORY,VARIABLE=PRESELECT *END STEP draw_stage1_attempt3_30.inp *HEADING Deep drawing of a can bottom -- stage 1, attempt 3 Mass scaling factor = 900 (speedup of 30) SI units (kg, m, s, N)

** *NODE,NSET=REFP1 1001,0.,0.024 *NODE,NSET=REFD1 1101,0.050,-0.012 ** ** BLANK ** *NODE,NSET=BLANK 501,.000,.000 601,.045,.000 *NGEN,NSET=BLANK 501,601,1 *ELEMENT,TYPE=SAX1,ELSET=BLANK 501,501,503 *ELGEN,ELSET=BLANK 501,50,2,2 *SHELL SECTION,ELSET=BLANK,MATERIAL=STEEL,SECTION INTEGRATION=GAUSS .0005,5 *NSET,NSET=BSYM 501, *NSET,NSET=NHIST 521,601 ** ** PUNCH1 -- PUNCH FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=PUNCH1, TYPE=SEGMENTS START,0.032,0.03025 LINE,0.032,0.00625 CIRCL,0.026,0.00025,0.026,0.00625 LINE,0.,0.00025 *RIGID BODY, REF NODE=1001, ANALYTICAL SURFACE=PUNCH1 ** ** DIE1 -- DIE FOR FORMING STAGE 1 (Analytical Rigid Surface) ** *SURFACE, NAME=DIE1, TYPE=SEGMENTS START,0.0326,-0.02825 LINE,0.0326,-0.00425 CIRCL,0.0366,-0.00025,0.0366,-0.00425 LINE,0.050,-0.00025 *RIGID BODY, REF NODE=1101, ANALYTICAL SURFACE=DIE1 ** ** SURFACE DEFINITIONS ** *SURFACE, NAME=BLANK_TOP BLANK,SPOS *SURFACE, NAME=BLANK_BOT BLANK,SNEG ** ** BLANK MATERIAL ** *MATERIAL,NAME=STEEL *ELASTIC 210.0E9,0.3 *DENSITY

7800., *DAMPING,BETA=1.E-7 *PLASTIC 0.912E8, 0.0 0.131E9, 0.159E-02 0.171E9, 0.649E-02 0.211E9, 0.177E-01 0.251E9, 0.395E-01 0.291E9, 0.776E-01 0.331E9, 0.139 0.391E9, 0.295 ** ** AMPLITUDE DEFINITIONS ** *AMPLITUDE,NAME=FORM1, DEFINITION=SMOOTH STEP 0.,0.,.033,1. ** ** INITIAL FIXED BOUNDARY CONDITIONS ** *BOUNDARY REFP1,1 REFP1,6 REFD1,1,6 BSYM,1 BSYM,6 ** ** STEP 1 -- FIRST FORMING STAGE ** ** *ELSET,ELSET=QA_TEST_ALLELEMS BLANK, ** *NSET, NSET=QA_TEST_ALLNODES REFP1, REFD1, BLANK, BLANK BSYM, NHIST ** *STEP *DYNAMIC,EXPLICIT ,.033 *FIXED MASS SCALING, FACTOR=900 ** ** Contact Pair Definition ** *CONTACT PAIR,INTERACTION=INTER PUNCH1,BLANK_TOP DIE1,BLANK_BOT *SURFACE INTERACTION,NAME=INTER *FRICTION .1, *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=FORM1 REFP1,2,2,-.015 *RESTART,WRITE,NUMBER INTERVAL=10 *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V

*ENERGY OUTPUT ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER=4 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U,RF ** ** Output for HKS QA purposes ** *FILE OUTPUT,NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U,RF *EL FILE S, *OUTPUT, FIELD, NUMBER INTERVAL=1 *NODE OUTPUT, NSET=QA_TEST_ALLNODES U,RF *ELEMENT OUTPUT, ELSET=QA_TEST_ALLELEMS S, *OUTPUT,HISTORY,VARIABLE=PRESELECT,TIMEINT=0.033 ***OUTPUT,HISTORY,VARIABLE=PRESELECT *END STEP draw_stage2.inp *HEADING Deep drawing of a can bottom -- stage 2 ABAQUS/Explicit forming stage 2 following draw_spring1_std.inp SI units (kg, m, s, N) *IMPORT,STATE=YES,UPDATE=NO BLANK, *IMPORT NSET BSYM,NHIST,BLANK ** ** PUNCH2 -- PUNCH FOR FORMING STAGE 2 ** *NODE,NSET=PUNCH2 201,.023875,.030 202,.023875,.002 232,.025875,.000 262,.031875,.006 263,.031875,.030 298,.025875,.002 299,.025875,.006 1201,.023875,.030 *NGEN,NSET=PUNCH2,LINE=C 202,232,1,298 232,262,1,299 *ELEMENT,TYPE=RAX2,ELSET=PUNCH2 201,201,202 *ELGEN,ELSET=PUNCH2 201,62,1,1 *RIGID BODY,ELSET=PUNCH2,REF NODE=1201 *NSET,NSET=REFP2 1201,

** ** DIE2 -- DIE FOR FORMING STAGE 2 ** *NODE,NSET=DIE2 301,.026,-.010 302,.026, .000 332,.0325, .0065 333,.050, .0065 399,.026, .0065 1301,.026,-.010 1302,.026,-.030 *NGEN,NSET=DIE2,LINE=C 302,332,1,399 *ELEMENT,TYPE=RAX2,ELSET=DIE2 301,301,302 *ELGEN,ELSET=DIE2 301,32,1,1 *RIGID BODY,ELSET=DIE2,REF NODE=1301 *NSET,NSET=REFD2 1301, ** ** DOMER -- DOMER DIE FOR FORMING STAGE 2 ** *NODE,NSET=DOMER 401, .000, .008 431, .022, .000 432, .022, -.010 499, .000, -.02625 1401,.000, .008 *NGEN,NSET=DOMER,LINE=C 401,431,1,499 *ELEMENT,TYPE=RAX2,ELSET=DOMER 401,401,402 *ELGEN,ELSET=DOMER 401,31,1,1 *RIGID BODY,ELSET=DOMER,REF NODE=1401 *NSET,NSET=REFDOMER 1401, ** ** Surface Definitions ** *SURFACE, NAME=BLANK_TOP BLANK,SPOS *SURFACE, NAME=BLANK_BOT BLANK,SNEG *SURFACE, NAME=PUNCH2 PUNCH2,SNEG *SURFACE, NAME=DIE2 DIE2,SPOS *SURFACE, NAME=DOMER DOMER,SPOS ** ** SPRING AND DASHPOT FOR FORMING STAGE 2 ** *ELEMENT,TYPE=SPRINGA,ELSET=SPRING

1301,1301,1302 *NSET,NSET=SPEND 1302, *SPRING,ELSET=SPRING 6.3E6, *ELEMENT,TYPE=DASHPOTA,ELSET=DASHPOT 2301,1301,1302 *DASHPOT,ELSET=DASHPOT 1.E4, *NSET,NSET=REF REFP2,REFD2,REFDOMER,SPEND ** ** MASS FOR SPRING IN FORMING STAGE 2 ** *ELEMENT,TYPE=MASS,ELSET=MASS 1302,1301 *MASS,ELSET=MASS 10., ** ** AMPLITUDE DEFINITIONS ** *AMPLITUDE,NAME=RIGID,DEFINITION=SMOOTH STEP 0.,0.,1.E-6,1. *AMPLITUDE,NAME=FORM2,DEFINITION=SMOOTH STEP 0.,0.,.033,1. ** ** STEP 4 -- POSITION RIGID BODIES FOR SECOND PUNCHING STAGE ** *STEP *DYNAMIC,EXPLICIT ,1.E-6 *RESTART,WRITE,NUM=1 *BOUNDARY,TYPE=VELOCITY REFDOMER,1,1,0. REFDOMER,6,6,0. REFD2,1,1,0. REFD2,6,6,0. SPEND,1,1,0. SPEND,3,3,0. REFP2,1,1,0. REFP2,6,6,0. *BOUNDARY,AMPLITUDE=RIGID REFDOMER,2,2,-.02355 REFD2,2,2,-.0153 SPEND,2,2,-.0153 REFP2,2,2,-.01465 *BOUNDARY BLANK, ENCASTRE *DIAGNOSTICS,CONTACT INITIAL OVERCLOSURE=DETAIL,DEFORMATION SPEED=DETAIL *OUTPUT, HISTORY, TIME INTERVAL=1.E-5 *NODE OUTPUT, NSET=NHIST U,V *NODE OUTPUT, NSET=REF U, RF *ENERGY OUTPUT

ALLKE, ALLIE, ALLWK, ALLVD, ALLAE, ALLPD, ETOTAL *OUTPUT, FIELD, NUMBER=4 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U, *END STEP ** ** STEP 5 -- ANNEAL ** *STEP *ANNEAL *END STEP ** ** STEP 6 -- SECOND PUNCHING STAGE ** *STEP *DYNAMIC,EXPLICIT ,.033 *FIXED MASS SCALING, FACTOR=10, ELSET=BLANK *DIAGNOSTICS,CONTACT INITIAL OVERCLOSURE=DETAIL,DEFORMATION SPEED=DETAIL *CONTACT PAIR,INTERACTION=INTER PUNCH2,BLANK_TOP DIE2,BLANK_BOT DOMER,BLANK_BOT *SURFACE INTERACTION,NAME=INTER *FRICTION .1, *BOUNDARY,TYPE=DISPLACEMENT,AMPLITUDE=FORM2,OP=NEW REFP2,2,2,-0.010 *BOUNDARY,TYPE=VELOCITY,OP=NEW REFP2,1,1,0. REFP2,6,6,0. REFD2,1,1,0. REFD2,6,6,0. SPEND,1,3,0. REFDOMER,1,6,0. BSYM,1,1,0. BSYM,6,6,0. *MONITOR,NODE=1301,DOF=2 *RESTART,WRITE,NUMBER INTERVAL=10 ** ** Output for HKS QA purposes ** *FILE OUTPUT,NUMBER INTERVAL=1 *ENERGY FILE *NODE FILE U,RF *EL FILE S, *END STEP draw_spring2_std.inp *HEADING Deep drawing of a can bottom -- second springback

ABAQUS/Standard springback following draw_stage2.inp SI units (kg, m, s, N) *IMPORT, STATE=YES, UPDATE=NO BLANK, *IMPORT ELSET BLANK, *IMPORT NSET BSYM, *BOUNDARY BSYM,1 BSYM,6 *RESTART,WRITE,FREQ=999 *STEP,NLGEOM *STATIC .1,1. *BOUNDARY,FIXED BSYM,2,2 *OUTPUT, FIELD, FREQUENCY=10 *ELEMENT OUTPUT, ELSET=BLANK STH, SF, S *NODE OUTPUT U, ** ** Output for HKS QA purposes ** *NODE FILE,FREQ=999 U, *EL FILE,FREQ=999 S, *END STEP