9th European LS-DYNA Conference

Home » Conference Papers » 9th European LS-DYNA Conference

Manchester, 2013
A contribution to validation of SPH new features

Thomas Beal, Nicolas Van Dorsselaer, Vincent Lapoujade (DynaS+)

Since 1998, the Smoothed Particle Hydrodynamics (SPH) method has been developed by LSTC in LS-DYNA®. This method is called a meshfree method because traditional finite elements are replaced by particles which are not physically connected but mathematically linked. It is an alternative to the classical Lagrangian Finite Elements method and is used to simulate problems where materials are submitted to hydrodynamic deformation modes, such as high velocity impacts. In order to improve the capabilities of SPH in LS-DYNA, LSTC recently developed several new functionalities available in the latest versions of LS-DYNA. The first new feature is the possibility to choose a Lagrangian kernel for SPH particles, which means there are always the same neighbors for one SPH node (more stability in tension). The second option is Hybrid SPH / Solid elements, designed to couple the benefits of both SPH and Lagrangian finite elements. With these new Hybrid elements, it is now possible to switch from SPH to solid elements using certain criteria and to realize a better transition between SPH and finite elements areas (no more tied contacts needed). This paper presents the DynaS+ contribution to the test and the validation of these new SPH options. Simple test cases and more complicated cases representative of industrial problems have been performed to assess the behavior and the interest of these new features. All tests have been conducted using LS-DYNA V970 R6.0.0 released in 2012. A comparison with older modeling ways will show the benefits they already bring, and the ones they will bring in the future, when their development will be completed.

A General Damage Initiation and Evolution Model (DIEM) in LS-DYNA

Thomas Borrvall, Thomas Johansson and Mikael Schill (DYNAmore Nordic AB), Johan Jergéus (Volvo Car Corporation), Kjell Mattiasson (Volvo Car Corporation and Chalmers University of Technology), Paul DuBois

As the automotive industry is reducing the number of physical prototypes in favour of computer simulations during their design processes, a lot of demands is put on the accuracy of the virtual finite element models used for this purpose. In this context the mathematical modelling of fracture is of major importance and has been a field of intense research over the past 50 some years. There are numerous fracture models implemented in LS-DYNA, but as of tradition a fracture model is statically linked to the underlying constitutive model which in practice limits its usage to a single stage of the design process. Recently GISSMO [1,2,3] was introduced in an attempt to remedy this shortcoming by allowing the fracture model to be separated from the constitutive model, thus facilitating results from manufacturing simulations to be transferred to subsequent crash simulations. Behind GISSMO, the Damage Initiation and Evolution Model (DIEM) was developed in the same spirit and with similar capabilities, but this model has to the authors’ knowledge not been used extensively. The intention with this paper is to present this latter fracture model and compare it with GISSMO and, to some extent, with CrachFEM [4] to which it has superficial similarities.

A Quadratic Pipe Element in LS-DYNA®

Tobias Olsson, Daniel Hilding (DYNAmore Nordic AB)

Analysis of long piping structures can be challenging due to the enormous number of shell/solid elements that would be required to model a piping structure accurate. In that context a new beam element has been developed that can, if used correctly, reduce the number of elements used in a pipe simulation. Since it is constructed of 3 nodes it is perfect for describing pipe bends, so called elbows. This documentimplemented inDYNA R7 [1]. is meant as an introduction and modelling techniquesLS-DYNA® R7.0.0 but improvements are implementedfor the elbowin the comingelement. It is update of LS-DYNA R7

Adetailedaircraft tyre finite element modellanding safety assessment for hard

Hua Guo, Christophe Bastien, Mike V. Blundell, Gary Wood (Coventry University, Dunlop Aircraft Tyres Limited)

Tyres have an important role in landing gear systems upon aircraft landing and taxing on the ground. The performances of an aircraft tyre under varied load conditions are vital requirements for aircraft safety certification. This paper describes the development of a detailed finite element (FE) model of an aircraft test tyre in order to investigate its performance and assess its safety criteria. The work follows the findings from previous researches [1] [2] [3] [4] and focuses on the aircraft tyre safety assessment under various loading scenarios that were not yet studied. Initially, tyre inflation and static load simulations have been analyzed based on a full-scaled 3D detail LS-Dyna® FE model, replicating the actual geometry and the correlated material properties from industrial experimental data. The dynamic simulations that aim to duplicate tyre load upon aircraft landing scenarios have also been analyzed. Following the comments from aircraft tyre industrial data and guidelines[5] [6], the dynamic simulations have covered the tyre loading scenario from on-road taxing, normal (soft) landing, hard landing to crash landing under different aircraft landing weights. The stresses on tyre/wheel contact areas and on bead cords have been chosen as the safety criteria. The simulation results, analysis and comments have been discussed in great details. The modelling and simulations described in this paper aim to demonstrate the effective use of FE models for aircraft safety assessment, by studying the criteria of the tyre for load cases corresponding with testing and operational scenarios. The development of such predictive model would allow the manufacturers to assess tyre availability during the design process, and also add to the general drive towards the use of more virtual prototypes in an area traditionally reliant on experimental testing.

Advances in Simulating the Processing of Composite Materials by Electromagnetic Induction

M. Duhovic, P. Mitschang, M. Maier (Institut für Verbundwerkstoffe GmbH, Germany), I. Caldichoury, P. L’Eplattenier (LSTC)

In the previous installment of this work, a flat spiral “pancake” type coil geometry and two different plate material types with large differences in electrical and thermal conductivity (structural steel and carbon fiber reinforced polymer composite) were used to perform static plate induction heating characterization experiments for the purposes of characterizing the heating behavior of both materials in preparation for continuous induction welding simulations. The static plate heating tests were validated experimentally in two Finite Element Analysis (FEA) software codes including LSTC’s LS- DYNA® 980 (R7) solver. Following on from this initial work, a simulation test-bed has been created in order to study the continuous induction welding of two joining partners. The simulation test-bed mimics an experimental setup developed at the Institut für Verbundwerkstoffe (IVW) GmbH which considers a two-dimensional joining setup (two flat overlapping plates) and allows a more complete investigation of the thermal behavior that occurs during a continuous induction welding process. 3D surface plots of the top surface temperatures which are generated across the entire width of the joint as well as along its length can be investigated for different welding speeds and induction welding processing parameters. More importantly, the same types of surface plots can also be generated at the joining interface providing a complete view of the temperature profile that occurs during the process at this important location. This information can be used to decide on the optimum processing parameters to ensure that the material anywhere at the joining interface always remains within its prescribed upper and lower processing temperature limits. With its three-way physics coupling, the simulation test-bed also allows the consideration of further processing parameters including the influence of roller contact and additional top surface cooling via a moving air-jet nozzle for different induction welding speeds.

Ageing Effect on Crashworthiness of Bus Rollover

A.H. Iskandar, Q.M. Li (University of Manchester, UK)

International standards related to protection of occupants in rollover accidents are generally applied on new vehicles and regulated for design approval. However, there is no standard regulation governing rollover crash performance of aged vehicles and no proper systematic test to monitor it. It is known that vehicle structural integrity deteriorates after years of operation undergoing various service environments and conditions. The scenario is worsen if the vehicle is poorly maintained and still in operation beyond its lifespan. These factors directly influence the vehicle crashworthiness level and affects its safety. The present paper attempts to show how ageing factors like corrosion and deterioration of mechanical properties may change the overall vehicle structural strength and hence its crashworthiness performance. As a case study, rollover analysis according to European standards (UNECE R66) was set up on a finite element bus model using LS-Prepost and simulation was performed using LS-DYNA . Different levels of condition due to ageing effect were included in the model to examine the changes of rollover performance of the bus. Aged bus rollover studies from real world crashworthiness investigation in Malaysia were referred for comparison to support the argument. Result shows that ageing factors directly influence the performance of vehicle crashworthiness.

AirbagApplication for Structural Racing Car Component

Alessia Prato, Marco Anghileri, Andrea Milanese (Dipartimento di Scienze e Tecnologie Aerospaziali, Italy), Chiara Marozzi (CRM Group)

In motorsports, there is a deep research in order to make cars more and more competitive, throughout an accurate study on aerodynamics, powerful and advanced engines, structural component design, materials and, of course, on racing strategy. That’s even more true when racing conditions are extremely severe, like in endurance races, such as 24 Hours of Le Mans, where changes in drivers and in racing conditions make the race more exciting. To improve performances, weight is one of the most strict design conditions: a lighter car is faster, more efficient and more competitive so that performances can be improved, with more controllability in acceleration and braking phases, reducing emissions and consumptions. Evolution in studies upon materials and structural applications has got a key role to reduce weight and the use of innovative structures is an open field for experimental tests. Many research works were performed upon materials used in motorsports, especially composite fiber reinforced ones. In [1], Adam presented a general overview on composite mechanical characteristics, production methods, and design principles for racing cars composite applications. In [2], a similar study was presented by Cole and Sherman referring to different light-metals and their use for automotive structural components. Innovative structural applications on car components were also studied in order to improve strength and reduce, at the same time, weight. In the fifties experimental tests carried out at Langley Aeronautical Laboratory of National Advisory Committee for Aeronautics (NACA) revealed that a small amount of pressure inside cylindrical components could increase strength. The first rigorous study upon this topic was performed by Lo, Crate and Schwartz [3]. From the comparison of experimental tests and analytical studies, they demonstrated that internal pressure can positively influence the resistance of cylindrical structures under torsional and compressive loads. Later some studies were carried out on specific cases, like silos and pipes, and Mathon and Limam [4] showed that the effects of internal pressure on closed cylinders under pure bending are dependent on the geometry and, particularly, on the ratio radius/thickness (R/t) and length/radius (L/R). Other studies [5] presented the effects on composite structures with different loads and with or without internal pressure. In this work, the effects of internal pressure on closed cylinders subjected to different loads were ® numerically investigated. Internal pressure was created throughout the use of LS-DYNA 971 airbag models and increasing levels of pressure were tested to evaluate the effect on the analyzed structure. Several FE models of different cylindrical geometries and loads were created and a comparison between metals and composite materials was numerically performed. A focused research on an innovative shape, that can take advantages from this kind of application, was also performed and its results compared with previous model ones.

ALE Incompressible Fluid in LS-DYNA

Nicolas Aquelet, Mhamed Souli (LSTC)

The computation of fluid forces acting on a rigid or deformable structure constitutes a major problem in fluid-structure interaction. However, the majority of numerical tests consists in using two different codes to separately solve pressure of the fluid and structural displacements. In this paper, a monolithic with an ALE formulation approach is used to implicitly calculate the pressure of an incompressible fluid applied to the structure. The projection method proposed by Gresho is used to decouple the velocity and pressure.

An automated belt fitting tool for 6 and 10 year-old child crash dummies

Gillian Mara, Christian Dalton (Semcon)

Modelling child occupant safety through computer aided engineering (CAE) within a vehicle is an area that is constantly developing. In 2003, Euro NCAP introduced a child occupant protection rating to inform consumers the results of vehicle safety assessments [1]. These assessments used 18 month and 3 year old dummies in frontal and side crash. Developments in technology and crash dummies have led to the introduction of new larger child dummies in the Q series, those being the 6 and 10 year old. In 2016-18 it is proposed these will be used in frontal and side EuroNCAP consumer crash ratings, where the dummy will be sat in a child seat. In the competitive market there is a demand on manufacturers to produce cars in a shorter time frame from idea to production, therefore making the overall process more efficient and cost effective. One area that this places emphasis on is predicting how cars perform where it has the potential to reduce the need to produce as many prototype vehicles. To ensure the results are accurate the predictive method of finite element analysis (FEA) is very dependent on the inputs, where those are primarily grouped into materials, geometry and boundary conditions. In terms of the child occupant prediction this means obtaining realistic inputs to gain realistic output kinematics and injury criteria during the frontal or side crash. The belt restraint system is one of the most predominant in affecting the child kinematics and one that needs to be modelled accurately. This paper will describe an automated seat belt fitting method using the LS- DYNA® package that is efficient and effective in creating a belt geometry and path for a 6 and 10 year old dummy in a child seat.

An Automated Process for Interior Head Impact Analysis

Thanassis Fokilidis, Antonis Perifanis (BETA CAE Systems SA), Anneli Hoegberg (VOLVO CAR Corporation)

During the last decade, the Occupant Protection legislation requirements on interior safety properties of automobiles have become considerably more demanding. The US standard laboratory test procedure FMVSS201U regulates the tests for the head protection against the event of its impact on the upper interior roof. The procedure involves the identification of the most critical target zones and headform target angles. To minimize the need for physical tests, the product performance evaluations are, as far as possible, performed with numerical simulations involving CAE software. However, covering all possible load cases would require an infinite number of simulations. In order to automate the generation of input data and to assist in finding the critical load parameters, Volvo Cars Safety Centre and BETA CAE Systems employed a process that involves ANSA pre- SURFHVVRU DQG (7$ SRVW-processor, extended with special tools. This automated process starts with the model file input to ANSA and concludes with the assessment of the LS-‘<1$ UHVXOWV ZLWKLQ ü ù requiring minimum human interaction. An advanced algorithm, for the automatic headform positioning, which identifies the areas where the maximum vertical angle can be reached, has been developed. The process also involves a robustness analysis for each target point, and thus reduces the uncertainty of the problem. This has been successfully deployed within Volvo Cars Safety Centre, leading to great reduction of modelling time

Applying Digital Image Correlation Methods to SAMP-1 Characterization

Hubert Lobo, Brian Croop, Dan Roy (DatapointLabs, USA)

SAMP-1 is a complex material model designed to capture non-Mises yield and localization behavior in plastics. To perform well, it is highly dependent on accurate post-yield material data. A number of assumptions and approximations are currently used to translate measured stress-strain data into the material parameters related to these inputs. In this paper, we look at the use of direct localized strain measurements using digital image correlation (DIC) as a way to more directly extract the required data needed for SAMP-1

Assessment Of Mach Stem Pressures: Comparison Of Experiments With Engineering And Eulerian Models

Len Schwer (Schwer Engineering & Consulting Services)

Laboratory scale tests conducted by Kisters and Kuder (2012) provide reflected pressure histories for a 1kg TNT charge detonated at 0.32m above a rigid surface. The pressure histories are measured 1m from the charge using a vertical array of pressure transducers to infer the height of the Mach Stem at this range. Two such vertical arrays, each with 9 pressure transducers at varying heights above the surface, were used, and the test was repeated three times. Four additional configurations with heights- of-burst (HOB) varying from 0.1 to 0.32m and standoff ranges of 0.9 to 1.2m were conducted, but not reported in the present reference. Examination of the measured wave forms allowed Kisters and Kuder to estimate the height of the Mach Stem, a.k.a. triple point, to within 60mm (vertical gauge spacing). The pressure histories from the vertical array measurements provide an opportunity to assess the accuracy of the LS-DYNA air blast engineering model referred to as LOAD_BLAST_ENHANCED, and results from simulations using the LS-DYNA Eulerian solver usually referred to as Multi-Material Lagrangian Eulerian (MM-ALE). In addition to comparisons of maximum pressure, and time-of-arrival (TOA), the time integrated pressure histories provide maximum impulses for comparison.

BEM Methods For acoustic and vibroacoustic problems in LSDYNA

Souli Mhamed, Yuang Yun, Zhe Cui, Zeguar Tayeb (LSTC)

The present work concerns the new capability of LS-DYNA in solving acoustic and vibroacoustic problems in time domain and frequency domain, using the following LSDYNA Cards: *MAT_ACOUSTIC *FREQUENCY_DOMAIN_ACOUSTIC_BEM *FREQUENCY_DOMAIN_ACOUSTIC_FEM *FREQUENCY_DOMAIN_SSD In vibroacoustic problems, which are assumed to be weak acoustic-structure interactions, the transient structural response is computed first. By applying the FFT, it is transformed into a frequency response. The obtained result is taken as boundary condition for the acoustic part of the vibro acoustic problem. Consequently, the radiated noise at any point into space can be calculated. The new developed LS-DYNA keyword is based on boundary element method (BEM) in which only the surface of the acoustic domain needs to be discretized. Besides BEM that solves the Helmholtz equation as a linear system, the new card allows, also, to use two other approximate Rayleigh and Kirchhoff methods. Both methods do not require a system of equations to be assembled and solved. Consequently, they are faster than BEM. Rayleigh method assumes that the radiating structure is a plane surface clamped into an infinite rigid plane. In Kirchhoff method, BEM is coupled to FEM used for acoustics in LS-DYNA by prescribing non reflecting boundary condition. In this case, at least one fluid layer needs to be merged to the vibrating structure.

Buried Charge Engineering Model: Verification And Validation

Len Schwer (Schwer Engineering & Consulting Services), Todd Slavik (LSTC)

Livermore Software Technology Corporation (LSTC) recently added an empirically-based model for buried mine blast loading via the keyword *INITIAL_IMPULSE_MINE. This is an engineering model based upon experimental results, much like the air blast engineering model provided by *LOAD_BLAST_ENHNACED. The model is based on the work of Westine, et al. (1985) as presented, and extended, by Tremblay (1998). The implementation is applicable to flat (horizontal) and oblique (angled) target plates consisting of either shell or solid elements. Westine, et al. (1985) performed a series of buried charge experiments using thick (non-deforming) target plates with a set of impulse plugs inserted into holes drilled in the plate. The impulse plugs were spaced at 10 different ranges from the plate center, and hence angles from the center of the buried charge. Detonation of the buried charge forces the impulse plugs vertically out of the plate and measurement of their speed provides an estimate of the momentum (impulse) provided by the buried charge.

Closing the Gap between Process- and Crash Simulation for Composite Materials

Ch. Liebold, A. Haufe (DYNAmore GmbH, Germany), H. Finckh (Deutsches Institut für Textil- und Faserforschung Denkendor, Germany), T. Günther(Volume Graphics GmbH, Germany)

The usage of fiber reinforced plastics (FRP) becomes more and more important in the automotive industry and therefore, the requirements for accurate crash analysis results for such materials increase. As already known from sheet metal forming, process simulations are performed to predict fiber orientations for all kinds of short, long or endless fiber reinforced parts, even though most of the available software tools only use geometrical approximations. On the example of carbon reinforced braided tubes, a mapping tool will be introduced which is able to transfer fiber orientations as well as initial stresses and strains gained from braiding simulations using a high resolution beam mesh onto a coarse crash mesh. The mapping results will be compared with CT–scans taken from such tubes to confirm their accuracy. In the near future, the developed mapping algorithm shall support all kinds of process simulations such as draping or weaving for different element types as well as for different composite materials.

Development of special new versions of the FAT/PDB Dummy models for quick analysis response. The Rapid Analysis Models (R.A.M.)

Reuben D’Souza, Yupeng Huang, Sebastian Stahlschmidt (DYNAmore GmbH, Germany)

The finite element model of the FAT and PDB Dummy models have been developed in co-operation with the German Automotive lndustry for the last few years, One of the major goals during the development of the models was to achieve a high degree of accuracy of the model as well as to have a robust model without numerical instabilities. During the course of the model development, the geometry of various components has evolved considerably and finite element meshes for the various components have gotten finer. New material tests were also carried out for many components resulting in the use of advanced material models. This has naturally resulted in an increase in the computation times for the model. Increased computation times are definitely a hindrance when it comes to carrying out a studywith a large sample set for example. Keeping this in mind, we at DYNAmore have developed a simplified version of the ES-2IES-2re model meant for rapid prototming. This simplified model requires considerably lesser computation time and delivers extremely good results in the component and barrier tests, considering the degree of simplicity of the model. The geometry of the model has remained unchanged. Simplified and “quicker” materid models have been used wherever possible and all the certification and component tests were carried out for the model. Development of a R.A.M. model of the WorldSID 500b has also been started. As a final step, the soffware CORA [1J (CORrelation and Analysis) waas used to obtain an objective comparison of the simplified model to the original model.

Drilling process modelling using SPH

Virginija Gyliene, Vytautas Ostasevicius, Martynas Ubartas (Kaunas University of Technology, Lithuania)

Numerical methods became a powerful tool to understand such complicated process as cutting processes. Based on finite element techniques, firstly a big challenge – geometric modeling of bodies coming in contact interaction and mesh size/zone choice and creation. Secondly, today’s we now that finite element modeling is sure tool in the research of cutting processes after many series of experiments. So, here is the purpose to get reliable solution more quickly. ® The latest versions of LS-DYNA present the possibilities of Smooth Particle Hydrodynamics (SPH) method. This method was used in 3D drilling process modeling. Solution was done assuming high impact contact interaction, when cutting tool was assumed as non-deformable rigid tool, made of solid elements and work-piece consisted of particles instead of elements. SPH model was created by SPH generation interface from solid nodes. The papers will presents some aspects of multi edge cutting process modeling (drilling) using SPH method, as contact definition and SPH control. Either work-piece failure by tool motion (feed and rotation) was generated. So, two movement laws were defined. Firstly, the understanding of SPH method also is desirable for its appreciation that this method is “natural” considering the chip separation process. Secondly, this method is so-called as “fast” method as concerns Kernel function. These mentioned aspects were reason for the choice of SPH method for drilling process modelling with the purpose in further analysis to apply ultrasonic excitation. Firstly, the paper presents drilling experimental setup. Also the paper describes the techniques of geometrical model generation using SPH particles. The techniques of modelling of high impact contact interaction are presented.

Drilling rotation constraint for shell elements in implicit and explicit analyses

Tobias Erhart (DYNAmore GmbH, Germany), Thomas Borrvall (DYNAmore Nordic AB, Sweden)

A subordinate but interesting detail in theory and application of shell elements is investigated in this study, namely the drilling rotation constraint approach. Standard shell elements exhibit 3 translational and 3 rotational degrees-of-freedom at each node. While two nodal rotations are directly associated with bending and twisting modes, the third rotation about the shell normal (also known as the drilling rotation) does not provide any resisting force or stiffness by itself. This fact leads to zero valued components in the stiffness matrix for implicit analyses, which in turn results in a system of equations that cannot be solved. Therefore a small amount of stiffness in form of a torsional spring is artificially added just to remedy the singularity but not to affect the solution too much. This is absolutely necessary to deal with implicit analysis, otherwise no results could be obtained. On the other hand, it might be helpful to have this option also available in explicit analyses to improve results in special situations. It is the intention of this paper to present the theoretical background of this phenomenon and to illustrate the influence of the constraint method in several numerical examples.

Evaluation of a dummy design by using a human body model

Christian Gehre, Norbert Praxl (PDB), Sebastian Stahlschmidt, Dirk Freßmann (DYNAmore GmbH)

Human body models for occupant protection became popular in the last years. They started to turn from high sophisticated research tools to reasonably applicable tools to support some specific areas of occupant safety. This study is focused on the evaluation of the BioRID-II shoulder design by using the THUMS human body model. After the introduction of the BioRID-II into several test protocols to assess whiplash associated disorder, some serious concerns about the dummy’s performance emerged. Various simulations and tests indicated that the dummy’s shoulder design may cause unrealistic loads. Simulation runs with the THUMS in the same test environment were used to verify this assumption. The overall kinematics and therefore the accelerations of human body model and dummy model correlate well. A comparison of forces and moments between both is difficult because of the completely different internal structure of human and dummy. However, in-depth analyses showed that the simple dummy shoulder causes direct neck loads, while a human shoulder distributes the load through clavicles and scapulas to the whole rib cage. The same artefact was observed at the recently developed model of a female rear impact dummy (EvaRID) that is based on the BioRID-II design. The THUMS was scaled down to the size of the EvaRID by keeping almost same relative differences in size as observed between BioRID-II and THUMS. The dummy artefact could be verified with the downscaled THUMS, too. In summary, a human body model is a complex, not easy to handle but helpful tool to evaluate the performance of crash test dummies and to identify dummy related artefacts. While the overall kinematics between dummy and human model are somehow comparable, forces and moments may differ because of the different internal designs of dummy and human body model.

Experimental Investigation and FE Modeling of the High Temperature Dynamic Properties of Metals in the Kolsky method

Pavel A. Mossakovsky, Alexander V. Inyuhin, Fedor K. Antonov (Reaserch Institute of Mechanics of Lomonosov Moscow State University, Russia), Liliya A. Kostyreva (FSUE “GTERPC ‘Salut”, Russia)

In a wide range of practical problem on the dynamic strength associated with the impact and penetration an adequate definition of the material properties as a strain rate and temperature function has a great importance. One of the most effective and reliable methods for determining materials properties at high strain rates is the Kolsky method using the split Hopkinson bar [1]. This method is based on the one-dimensional theory of elastic wave propagation and the assumption of homogeneity of strain in the sample. It allows to obtain the deformation diagram for the processes of compression and tension in the range of strain rates 200-10000 s-1. Two basic schemes of the initial heating of the sample exist to study the materials properties at elevated temperature. In the one case, the heater is located directly on the axis of the rods (Fig.1) and heat-exposed not only the sample but also the adjacent parts of the rods. In the second scheme, the furnace is at the distance from the rod (Fig. 2), heat is directly exposed to the sample. Thus, researcher should take into account a non-uniform temperature field produced by rapid cooling of the sample while delivering from the oven to the bars. In all cases it is necessary to clarify the standard scheme of the Kolsky method. The paper is concerned with a numerical and experimental investigation of the temperature and the stress-strain state in the specimen in compression tests with split Hopkinson bar with a remote oven. The goal was to determine the degree of heterogeneity of the temperature field in the sample during the test and to discuss the approaches to the refinement of a standard methodology.

FE Fracture Prediction on Edges of Punched Holes

Dr Janka Cafolla, David P. Norman (Tata Steel UK Limited)

Many industry sources identified fracture from holes and edges as a key knowledge gap for fracture prediction of AHSS and UHSS grades. Ford Aachen with MATFEM cooperation [1] has shown an importance of including fracture failure models into their finite element (FE) simulations especially when designing with UHSS such as Boron Steel. Hole expansion coefficient (HEC) is one of the ways to characterise the edge cracking of metals. Many studies investigated the best way of predicting the fracture using HEC in their experimental work [2], [3] combined with FE simulation [4]. In the later one a simplification for an FE simulation is done such that only 2D model is utilized. Since the study examines the area around the hole in quite a detail, the mesh density is too high for general use in crash FE model of a full vehicle. In forming simulations HEC and forming limit curve (FLC) are often the criteria for inspecting the possibility of material failure. In crash simulations neither HEC nor FLC are generally used. More suited are the phenomenological fracture models which are more applicable for a wider range of engineering problems. There are now a number of these fracture models available. Examples of some of these fracture models are from Gurson [5] and various extensions to Gurson, Dell and Gese [6], Wierzbicki [7] and Wilkins [8]. Dell and Gese use CrachFEM material model developed by MATFEM [9]. Tata Steel adopted the use of CrachFEM material in past few years. Horn [10] and his colleagues at Tata Steel Research and Development have investigated several methods for measuring and analysing the data needed as input variables in CrachFEM material cards. CrachFEM material data are now available for the customers in AURORA material database. Norman and Buckley [11] have successfully used this data in their automotive virtual development process. Their test-FE correlation gave very promising results. CrachFEM material model has therefore become a winning tool for developing a simple method of fracture prediction in a crash event. A simple rectangular test specimen with a hole in the middle was designed such that it was possible to use the optical strain measurement system – ARAMIS. This gave an opportunity to correlate not only force-displacement curves between the test and FE simulations but also strain development around the hole.

Finite Element Modeling of Aluminium Honeycomb with Variable Crush Strength and Its Application in AE-MDB Model

M. Asadi (Anglia Ruskin University), B. Walker, H. Mebrahtu (ARUP Campus (UK)), M. Ashmead (Cellbond (A Division of Encocam Ltd.))

Aluminium honeycomb blocks are often to gain differentiated crush strength pattern to represent variable behavior while subjected to static/dynamic deformation. Current article demonstrates the methodology to validate modeling techniques and implementing in a finite element model for the Advanced European Mobile Deformable Barrier (AE-MDB). AE-MDB v3.9 side impact barrier has been investigated in present paper. The FE model is then examined using experimental data from a set of full-scale tests. Component tests have been designed and performed to establish the material characteristics for the FE model to maintain the crush strength pattern within the specified design corridors. The model then has been analysed using LS-DYNA© under certain boundary conditions according to the test specifications and the results have been compared to the physical test data. The barrier has been subjected to the Flat-Wall and Pole tests while the obstacles were blocked against the barrier on a mobile trolley. The methodology is then certified through comparison of the deformation pattern and numerical information with the experiments.

Finite element modelling and validation of the honeycombs for automobile crash MDB and ODB

Hui Yang, Qianbo Liu, Liangfu Yu (Shanghai Motor Vehicle Inspection Centre), Lin Li (Automotive College of Tongji University)

Honeycomb materials are widely used in automotive crash tests. Typically, it is the main components of the ODB (Offset Deformable Barrier) and MDB (Mobile Deformable Barrier) stipulated in ECE Regulation No.94 and No.95 on automotive crash test. These two kinds of honeycombs or barriers are also adopted by Chinese regulations. The accuracy and efficiency are most important for the CAE analysis of automotive crash simulation. In the earlier the solid elements is mainly employed for honeycomb modelling due to the limitation of computer calculation. The challenge of the solid element modelling is to overcome the hourglass energy, computational stability and local deformation simulation, etc. Recently, with the rapid improvement in computer hardwires, the shell elements are [1][2][3] more and more used for modelling the honeycomb . The shell model for honeycomb has some advantages such as high computational stability, lower hourglass energy and good simulation for detailed local deformation. The shell models of honeycomb can be found from the LSTC Inc. and [3] Wang . In China, the majorities of auto manufacturers still use the solid model of honeycomb from overseas commercial models such as ARUP and ESI honeycomb models. The shell model of honeycomb hasn’t yet been widely used because it is in grow-up stage and needs more validations for its accuracy. Meanwhile, some data or parameters in these commercial models is invisible and cannot be handled. Furthermore, some problems were found in the actual CAE applications, such as too strong glue, excessively hard character of the whole honeycomb and abnormal energy, and so on. On the other hand, many auto manufacturers still insist on developing own honeycomb models so that they can grasp the whole analysis simulation. Based on the experimental data and other literatures, this paper presents the FE models of ECE ODB and MDB developed by LS-DYNA®. By means of whole vehicle crash validations, these two models can give a good accuracy and computation stability. All the codes of these models will be opened to the public so that it will be helpful for auto engineers to comprehend the details of the honeycomb model and to improve the models.

Fluid-Structure Interaction involving Close-in Detonation Effects on Column using LBE MM-ALE Method

Swee Hong Tan, Shih Kwang Tay, Jiing Koon Poon, David Chng (ministry Of Home Affairs, Singapore)

This paper shares the experiences gathered from studies conducted on the use of *Load_Blast_Enhanced (LBE) keyword to couple empirical blast loads to air domain in Multi-Material Arbitrary Lagrangian-Euler (MM-ALE) environment and on Fluid-Structure Interaction (FSI) computations relating to various aspects of coupling technique in LS-DYNA® via *Constrained_Lagrange_in_Solid keyword for structures composing of mainly solid elements. This paper also presents a case-study in which results from the LBE MM-ALE FSI simulation were compared to experimental data from full-scale blast trials, as well as results from associated pre-test simulations. The pre- test simulations were done using a 2-stage numerical approach which involved applying segmental pressure loadings derived from Computational Fluid Dynamics (CFD) calculations on LS-DYNA Lagrangian models to predict structural response.

Following Nature’s Lead for Ultimate Design Efficiency The ACP Process as Applied to FSV

Akbar Farahani (ETA Inc.), Jody Shaw (U.S. Steel)

The shapes and configurations of nature are wildly complicated, non-intuitive and completely amazing. The shapes and forms found in nature in the structure of a tree, a human skeleton, insects and animals are truly the most efficient designs imaginable. By mimicking the flawless balance between structure and strength of nature’s most efficient shapes, engineers can learn how to incorporate similar balance to product structural design for automobiles, aircraft and other systems. The Accelerated Concept to Product (ACP) ProcessTM is a methodology which enables the structure of a product, such as the vehicle’s body-in-white, to mimic “Nature’s Way” [13]. Doing so creates the ultimate design efficiency, where structure and strength are perfectly balanced for the intended function. ACP is a proprietary, performance-driven, holistic product design development method based on design optimization and incorporates the use of multiple CAE tools in a systematic process to generate the optimal design solution. This methodology provides solutions, which address the challenges facing the modern product development environment. It achieves this by synchronizing the individual facets of the product development process, resulting in an overall reduction in development costs and time to market.

GasDynamic Simulation Capability for Side Impact Pressure Sensing Calibration

Bill Feng, Amarjit Grewal (Body CAE, Jaguar Land Rover Ltd), Steve Hickman, Dave Moore, Alexandro Badea-Romero (Vehicle Safety, Jaguar Land Rover Ltd), Christian Dalton (Semcon Engineering)

A method has been developed that uses crash simulation models to provide side impact door pressure sensor data to the sensor calibration engineers much earlier than was hitherto possible, thereby affording the opportunity to reduce the time period required for physical sensor calibration in the vehicle development programme.

Head of Leveraging LS-DYNA Explicit, Implicit, Hybrid technologies with SGI hardware, Cyclone Cloud Bursting and d3VIEW

Olivier Schreiber, Tony DeVarco, Scott Shaw (SGI), Suri Bala (LSTC)

LSTC Explicit, Implicit solver technologies are closely integrated following LSTC’s single executable strategy. Seamless switching from large time steps transient dynamics to linear statics and normal modes analysis can thus consistently exploit latest algorithm improvements in Shared Memory Parallelism (SMP), Distributed Memory Parallelism (DMP) and their combination (Hybrid Mode) and leverage SGI computer architectures using SGI’s software stack, establishing `topcrunch’ world records since 2007. 7KLV SDSHU ZLOO VKRZ KRZ WKLV LV DFFRPSOLVKHG RQ 6*,¶V PXOWi-node Distributed Memory Processor ® ® clusters such as SGI Rackable and SGI ICETM X up to Shared Memory Processor servers such as SGI® UV 2000 all available in SGI® CycloneŒ. Cyclone is a LS-DYNA® cloud computing service provided in partnership with LSTC. This paper will discuss how customers are using SGI’s compute and storage infrastructure to run LS-DYNA simulations in a massively scalable environment. SGI’s front-end to Cyclone is powered by d3VIEW Œ, a web portal based software used to submit, monitor and view results without the need to download large files. d3VIEW’s SimlyzerŒ technology performs post-simulation analysis and visualization that is proven to eliminate over 80% of the LS- DYNA post processing repetitive tasks with no necessary scripting.

Influence of Contact Parameters on Short-Event Crash Simulation Results

Nick Kalargeros, Jack Perry, Antonio Peralta, Lee Pearce (Jaguar Land Rover Limited)

The aim of a vehicle crash simulation is to characterise and quantify the performance of specific regions in terms of energy dissipation, distribution and intensity. Such detailed understanding of the crash event will enable the analysis and prediction of occupant and / or pedestrian injuries. To achieve this, an interacting chain of individual components and systems need to be studied in terms of its energy management and absorption capacity. A study should consider the unique contact interactions between the key components and systems involved. More so, these unique contact interactions have to be numerically captured and formulated in a manner that is faithful to the actual physical event. The premise of this paper is to report initial findings from a study of the sensitivity of short crash events to different contact parameters and conditions. The relevant CAE modelling representations which lead to better agreement between virtual and physical results are explained. The aim is to increase the predictive capability not only of the nominal accuracy of the CAE predictions, but also to fully capture the chronological sequence of events and behaviours during the real short-event crash simulation.

Influence of Discretisation on Stiffness and Failure Prediction in Crashworthiness Simulation of Automotive High Pressure Die Cast Components

Felix Brenner, Helmut Gese, Gernot Oberhofer (MATFEM Partnerschaft Dr. Gese & Oberhofer), Michael Buckley (Jaguar Land Rover Limited)

Castings are widely used as part of the car chassis in automobile manufacture because of their light weight and the flexibility of the design process. Due to the comparable low ductility of castings, it is essential for crash simulations to gain dependable analyses. However, modelling casting parts correctly for finite element analyses is an issue for several reasons. In order to represent the elastoplastic stiffness correctly and thus to obtain reliable failure predictions, an accurate prediction of plastic strains and the corresponding stress states is required. To meet these conditions an adequate material and failure model is needed. Besides the characterisation and modelling of the material, the geometric discretisation is a trade-off between computational costs, meshing effort and the quality of the results that can be achieved in simulations. Typically, no general guidance is provided on the appropriate element formulation or the impact this choice may have on the results. Lastly, in industrial environments economic competition usually does not allow for extensive basic research. Conservative methodologies in development and simulation of castings are the norm since new methods carry the risk of failure. Thus avenues of improving accuracy and reducing costs of simulations remain to be explored.

InteractionPossibilities of Bonded and Loose Particles in LS-DYNA®

Nils Karajan (DYNAmore GmbH, Germany), Zhidong Han, Hailong Ten, Jason Wang (LSTC)

The goal of this presentation is to outline the current development status of LS-DYNA® with respect to simulations using the discrete-element method (DEM), which is based on Cundall & Strack [1]. Starting with assemblies of loose discrete spherical particles, different types of granular media can be discretized to predict their behavior, for instance, during mixing processes, storage and discharge in silos or transportation on belts. Following this, the interaction of the discrete particles with themselves as well as their surrounding deformable or rigid structures can be taken into account. Herein, friction coefficients as well as spring and damping constants can be defined in normal and tangential direction. Wet particles can be estimated with the aid of a capillary force model. Even though the geometric shape of the particles is always spherical, a certain roughness of the grains can be achieved by introducing a rolling friction or by defining clustered particles using bonds. Moreover, with the introduction of bonded particles, linear-elastic solid material behavior can be modeled. Herein, the mechanical behavior of the bonds may either be prescribed manually or computed internally by LS-DYNA in an automated fashion using the elastic constants given in a material card. With the definition of a fracture energy release rate of the bonds, fracture mechanics of brittle materials can be studied. Herein, the number of bonds of a particle to the neighboring particles can be defined with a bond radius. Note that the breakage of single bonds can be interpreted as micro cracks that eventually evolve to macro cracks. This presentation will give an overview of the involved material cards and provides information on how the cards are used. For a better understanding of the involved parameters, simple examples will be presented addressing particle-particle as well as particle-structure interaction.

JFOLD – Introducing A New Simulation-Based Airbag Folding System for LS-DYNA®

Shinya Hayashi (JSOL Corporation)

Computer simulation is playing an increasingly important role in the design, development and application of airbag safety systems. As folding patterns and airbag structures become more and more complex, users are turning to simulation based folding solutions to generate accurately folded models in a short space of time. To meet this demand, a new software tool called JFOLD has been developed by JSOL Corporation to enable successful airbag folding using LS-DYNA. JFOLD’s intuitive and interactive system guides the user through the folding steps using flow-chart graphics, interactive tool positioning/resizing, tool motion control, animation preview and so on. JFOLD runs inside the powerful and popular pre-processor Primer.

LS-DYNA® R7: Conjugate heat transfer problems and coupling between the Incompressible CFD (ICFD) solver and the thermal solver, applications, results and examples.

Iñaki Çaldichoury, Facundo Del Pin (LSTC)

LS-DYNA version R7 includes CFD solvers for both compressible and incompressible flows. The incompressible CFD solver (ICFD) may run as a stand alone CFD solver for pure thermal fluid problems or it can be strongly coupled using a monolithical approach with the LS-DYNA solid thermal solver in order to solve the complete conjugate heat transfer problem. This paper will focus on the thermal part of the ICFDthermal and conjugate heat transfer problems will beillustration and discussion purposes. solver and its associated features. Several results of presented as well as some industrial applications for

LS-DYNA® R7: Coupled Multiphysics analysis involving Electromagnetism (EM), Incompressible CFD (ICFD) and solid mechanics thermal solver for conjugate heat transfer problem solving

Iñaki Çaldichoury, Pierre L’Eplattenier, Facundo del Pin (LSTC), Miro Duhovic (Institut für Verbundwerkstoffe GmbH)

LS DYNA R7’s new modules and capabilities include: two fluid mechanics (CFD) solvers for incompressible (ICFD) and compressible flows (CESE) and an Electromagnetism solver (EM). The objective of these solvers is not only to solve for their particular domain of physics but to make full use of LS-DYNA capabilities and material library in order to solve coupled multiphysics. This paper will present how the EM solver can solve inductive heating problems, the problematic that arises when cooling the heated materials and/or coils is needed and how the ICFD solver can be used in conjunction in order to solve the complete EM-conjugate heat transfer problem. For illustration purposes, an industrial application studied at the Institut für Verbundwerkstoffe (See “Advances in simulating the processing of materials by electromagnetism induction” paper) will be introduced and discussed.

LS-DYNA® R7: Recent developments, application areas and validation results of the compressible fluid solver (CESE) specialized in high speed flows.

Zeng-Chan Zhang, Iñaki Çaldichoury (LSTC)

LS-DYNA version R7 includes CFD solvers for both compressible and incompressible flows. The compressible flow solver is based on the CESE method, a novel numerical method for solving conservation laws. It has many nontraditional features such as space-time conservation, second order accuracy for flow variables and a powerful shock wave capturing strategy. This paper will focus on some advanced features of the solver namely its FSI capabilities. Several potential industrial applications will be presented such as airbag openings, piston type applications and turbomachines. Some results on high speed supersonic flows will also be presented for illustration and discussion purposes.

LS-DYNA® R7: Strong Fluid Structure Interaction (FSI) capabilities and associated meshing tools for the incompressible CFD solver (ICFD), applications and examples.

Facundo Del Pin, Iñaki Çaldichoury (LSTC)

LS-DYNA version R7 includes CFD solvers for both compressible and incompressible flows. The solvers may run as standalone CFD solvers or they could be coupled to the LS-DYNA solid mechanics and thermal solvers for fluid structure interaction (FSI) and conjugate heat transfer problems. This paper will focus on the Incompressible CFD solver in LS-DYNA (ICFD) and its Fluid-solid interaction capabilities (FSI). Fluid structure interaction problems occur in physics whenever the flow over a structure causes deformation or displacement which in turn may influence the way how the fluid behaves. One of WKH VROYHU¶V PDLQ IHDWXUHV LV WKH LPSOHPHQWDWLRQ RI D UREXVW VWURQJ )6, FRXSOLQJ ZKLFK RSHQV D ZLGH QHZ range of applications in the range of aerodynamics, hydrodynamics, hemodynamics and so forth. Several examples will be provided for illustration and discussion. The ICFD solver is the first in LS-DYNA to make use of a new volume mesher that takes surface meshes bounding the fluid domain as input. For FSI problems that involve big displacements, the volume mesher algorithms need to be robust and flexible. Some of the latest developments and mesh control tools that are made available for the user will therefore also be introduced.

LS-DYNA®R7:Update On The Electromagnetism Module (EM)

Pierre L’Eplattenier, Iñaki Çaldichoury, Julie Anton (LSTC)

An electromagnetism module is being developed in LS-DYNA version R7 double precision for coupled mechanical/thermal/electromagnetic simulations. The physics, numerical methods and capabilities of this module will be introduced. Some examples of industrial applications will be presented. These include magnetic metal forming, bending and welding in different configurations, high pressure generation for equation of state studies and material characterization, induction heating, resistive heating, short circuits due to crashes, electromagnetic launchers, ring expansions, magnetic levitation and so forth. Additionally, magnetic material capabilities are currently available for beta testing and will also be discussed in this paper.

LS-OPT Parameters Identification on Concrete Sample Tests for an Impact Simulation on Concrete Slab

Nicolas Van Dorsselaer, Vincent Lapoujade (DynaS+), Georges Nahas, Bertrand Ciree, François Tarallo, Jean-Mathieu Rambach (Institut de Radioprotection et de Sûreté Nucléaire)

The dynamic behavior of Concrete is one of the most common and difficult problem of simulation in Nuclear, Defense and Civil fields. In most cases, the data available for modeling problems is much reduced; engineers are obliged to predict the behavior with non sufficient information. Due to this lack of experimental sample based input parameters, the result of simulation becomes “engineer dependent”, leading to much different results than people doing the same modeling problem. In previous paper ([5], [6]) presented during last LS-DYNA Conferences, we showed that a probabilistic approach for concrete modeling can be used to reduce these differences due to the modeling choices. But one of the main conclusions of these papers was that all these modeling techniques never replace experimental concrete sample tests to obtain the right material behavior before simulation. This paper is based on a work realized for an international OECD benchmark initiated by IRSN and CNSC. The main goal of IRIS_2012 Benchmark was to evaluate the ability of simulation to reproduce experimental tests of impacts on concrete slabs. Contrary to the earlier benchmark (IRIS_2010), experimental results of concrete sample tests was this time available in order to calibrate numerical constitutive laws before simulations on real tests. This paper, as the rest of our previous papers about IRIS_2010, will present the use of LSTC products capabilities in this kind of approach. In a first time, a complete LS-DYNA concrete model based on compressive strength will be created using automatic parameters generation capabilities of LS-DYNA. Then this model will be compared to experimental sample results of several cylindrical sample tests (simple compression and confined compressions at several confinement pressures). After sensitivity analysis to identify which parameters of the concrete model can be used to fit experimental results, LS-OPT parameters identification will be performed simultaneously on all cases. Based on the VTT Punching test simulation of IRIS_2012, we will compare the results between simulation with parameters automatically generated, simulation with fitted parameters and experiment. This comparison will be focused on missile velocity after impact and slab concrete damage. We precise that all the calculations presented here are performed with LS-DYNA solver, coupled with LS-OPT software for the probabilistic part of the studies (DoE studies, Monte Carlo Analysis, Robustness and Optimizations).

Modeling of Punctual Joints for Carbon Fiber Reinforced Plastics (CFRP) with *MAT_054

Christian Liebold (DYNAmore GmbH, Germany), David Moncayo (Daimler AG, Germany)

The increasing amount of carbon fiber reinforced plastic (CFRP) components used in the automotive industry opens new questions concerning the numerical modeling of different joining techniques for such materials. Design engineers already know that the beneficial properties of carbon fibers can be best used when considering the necessary joining techniques from the very first step in the preliminary design of a CFRP part. The anisotropical behavior of composites has to be fully considered in numerical simulations as well, especially when failure and partial damage occurs. Using experimental data from bearing tests loaded in different directions, different modeling techniques for structural joints are tested. The achieved correlation between simulation and experimental results will be discussed for a discretization relevant for industrial applications in a vehicle crash environment. Besides a qualitative and quantitative evaluation, the proposed modeling techniques are evaluated in terms of an appropriate representation of failure compared to the failure patterns observed during the experiments.

Modelling internal gas flows in a single stage gas gun using Eulerian/Lagrangian coupling in LS-DYNA

Marina Seidl, Kevin Hughes, Tom De Vuyst (Cranfield University)

Most research on gas guns for impact testing investigates the velocity reached by the projectile or sabot when it hits the target. In this research attention is paid to the effect of initial loading conditions on the velocity reached by the projectile upon exit from the barrel. The work is focussed on the single stage nitrogen gas gun at the Department of Applied Mechanics in the School of Engineering at Cranfield University of this research. This gas gun was used to generate test data for a range of initial pressures in the pressure vessel. An LS-DYNA model of the gas gun which uses the Eulerian/Lagrangian coupling feature is described. The LS-DYNA model results, as well as results from an analytical model, are compared to the test results. The results indicate that, while the analytical model over predicts the projectile velocity, the LS-DYNA model is capable of accurately predicting the projectile velocity as a function of the initial pressure in the pressure vessel. The results also indicate that the opening time of the valve affects the projectile velocity at higher initial pressures.

Modelling Spotweld Fracture Using CrachFEM

D. P. Norman (TATA Steel UK Limited)

In recent years, a number of research institutes have concentrated on trying to develop fracture models that are generally applicable to a wide range of engineering problems. Examples of some of these fracture models are from Gurson and various extensions to Gurson[1], Dell and Gese (CrachFEM)[2], Xue-Wierzbicki[3], Wilkins (EWK model)[4], and du Bois (*MAT_GISSMO)[5]. The key features of these models are a dependency of the fracture strain on the stress triaxiality and a means of accounting for void growth or instability due to necking. Some of these models also incorporate non-linear strain accumulation, kinematic hardening and sophisticated plasticity models, which may be necessary for modelling certain types of materials. The Dell and Gese (CrachFEM) material model is a popular choice in the European automotive industry and has been used in this study. One of the application areas of concern is at or near to spot welds, where the material properties of the weld and Heat Affected Zone (HAZ) are very different to the sheet and fracture predictions can be signicantly affected by this. This work investigates the potential for developing an accurate 3D weld model to describe the lap shear and cross tension plug fracture modes observed in DP600. Obtaining stress strain curves for the weld nugget and HAZ is a challenge. The standard approach is to use heat treated test coupons to perform a range of non-standard material coupon tests, with test coupons having the same micro-structures as the weld nugget and HAZ. To prepare the heat treated test samples requires spot welding simulation to determine the required temperature time curves observed during spot welding followed by Gleeble testing to reproduce the required temperature time cycles on test coupons. This is difficult to achieve in practice and several iterations may be needed to achieve the required micro-structures. This set of tasks is a significant undertaking and has been the subject of a number of university PhD and post-doctorate research studies. Dancette [6] and Sommer [7] are very good examples of this research. In this study, a simplified approach has been adopted using weld micrographs and micro-hardness indentation tests to infer the geometry and stress strain properties and to assume the fracture properties are the same as the sheet. This approach lacks the rigour of material testing coupons with tailored heat treatments but does provide a simpler approach that can be implemented more easily in industry.

Models for strain path independent necking prediction in LS-DYNA

Kjell Mattiasson (Chalmers University of Technology, Volvo Cars Safety Centre) , Johan Jergeus (Volvo Cars Safety Centre), Paul Dubois (Hermes Engineering NV)

Failure in sheet metal can be caused by one of, or a combination of, the following mechanisms: ductille fracture, shear fracture and plastic instability (necking). Ductile fracture is causrd by the initiation, growth and coalescence of voids in the material during plastic straining, commonly referred to as damage growth. Micro defects can also lead to through-thickness shear fracture in the sheet metal. There are several difIerent scenarios involving the a_ove mechanisms leading to material fracture. In ductile materials fracture is normally preceded by the formation of a neck in which the strains localizes after further loading. This neck has the width of the order of the sheet thickness. After the initiation of the neck the strains and the damage inside the neck start to grow rapidly. and finally the material breaks accoding to one of the mechanisms described above. The incipient necking is in this case also referred to as “plastic instability” as the phenomenon is solely dependent on the plastic properties of the material. It should be observed that the incipient necking phenomenon can be captured by a plane stress shell FE model if the elements are small enough, i.e. of the order of the sheet thickness. After the formation of the neck the stress state in the neck turns into a 3D one, and a highly refined 3D FE- model is rEquired to be able to simulate the post-necking behaviour. Another possible scenario is that damage starts to grow before the formation of a neck. The softening of the material can than speed up the neck formation. In some less ductile materials, e.g. some aluminium alloys and austenitic stainless steels, crack formation can even take place before strain Iocalization. Experience has, however, revealed that incipient necking is the by far the most common failure mode in sheet metal forming applications, as well as in automotive crash applications. The current paper will therefore be focused on the prediction of necking, and especially on methods for handling the strong strain path dependence of this phenomenon. The LS-DYNA code is used by Volvo Can for pertorming car crash simulations. Currently the third- party module CrachFEM from the MaHem company, which is linked to the LS-DYNA code. is used for performing material failure analmes. However. Iately several models for perIorming necking as well as fracture simulations have been implemented in LS-DYNA. It is the object of the current paper to give an overview of these options and to present results from some evaluations of these models.

Multi-disciplinary Topology Optimization for Vehicle Bonnet Design

David Salway (GRM Consulting Ltd), Tayeb Zeguer (Jaguar Land Rover Ltd)

Bonnet Pedestrian Head lmpact and Structural Stiffness and Strength targets have conflicting design requirements which currently result in design compromises, and the current CAE methods use different models and solvers. This paper highlights a new CAE capability to provide Multi-Disciplinary Optimization of bonnet geomety to achieve the conflicting Pedestrian Head Impact and structural stiffness/strength targets at lowest weight and cost. The aim has been to combine all bonnet load cases using one code “LS-DYNA” and cary out trade-oft and optimize weight using LS-OPT. A new developed topology process employing VR&D Genesis for HIC optimisation is presented and compared mih LS-TASC tools for a generic bonnet design.

New material modeling approaches for thermoplastics, composites and organic sheet

Matthias Vogler (Consulting engineer, Germany)

In this paper, new anisotropic elastic-viscoplastic constitutive models for simulating thermoplastic materials, endless fiber reinforced composites and organic sheets are presented. The anisotropic material models address the same main features as the isotropic SAMP model (MAT_187 in LS-DYNA). These are in particular pressure dependent yielding allowing for different yielding in tension, compression, shear and biaxial loadings, tabulated input of hardening data for each stress state and a non-associated flow rule for a correct prediction of the volumetric plastic strains. Hence, the anisotropic material models represent a consistent further development of the isotropic SAMP material model (SAMP-1 or MAT_187 in LS-DYNA 971). ) The anisotropy is incorporated by an invariant formulation using structural tensors. This provides interesting modeling techniques for short fiber reinforced thermoplastics and for organic sheets. When modeling short fiber reinforced thermoplastics, the fiber orientation tensor is directly integrated into the constitutive equations and an automated homogenization is performed. That is, the fiber orientation tensor “weights” the structural tensors representing the preferred directions and in the limiting case “all fiber directions equally distributed in all directions”, the isotropic SAMP model is recovered as a special case. When modeling organic sheets, the finite fiber rotations observed under certain loading conditions can be simulated. That is, an initial misalignment of the yarns due to the draping process and also a loading induced misalignment of the yarns due to the forming process can be incorporated easily, letting the structural tensors rotate against each other. The applicability of the anisotropic material models will be shown with three examples. First, simulation results of a short fiber reinforced polymer PA6GF60 are presented. These are in particular the material characterization tests (tensile, compression and shear tests) and quasi-static and dynamic 4a-Impetus bending tests. Secondly, simulation results of quasi-static and dynamic off-axis compression tests of a carbon epoxy IM7-8552 are presented, predicting the experimentally observed pre-failure nonlinearities. Finally, the applicability of the anisotropic model to organic sheets is discussed. The experimentally observed highly non-linear behavior under shear dominated loadings due to the finite fiber rotations and the quasi brittle behavior in uniaxial tension and compression in the main directions can be predicted. In future developments, the whole process chain drape simulation, forming simulation and crash simulation will be addressed.

Node to node contacts for SPH applied to multiple fluids with large density ratio

Jingxiao Xu, Jason Wang (LSTC)

The interesting and complex behavior of fluids emerges mainly from interaction processes. SPH has shown to be a simple, yet flexible method to cope with many fluid simulation problems in a robust way. However in SPH, particles have a spatial distance (smoothing length) over which their properties are smoothed by a kernel function. Smoothed quantities of a particles show falsified values when densities and masses of neighboring particles vary largely within the smoothing length. The erroneous quantities lead to undesirable effects, reaching from unphysical density and pressure variations to spurious and unnatural interface tensions, and even to severe numerical instabilities. In this paper, instead of using the traditional interaction between SPH parts through SPH interpolation, we present a node to node contact between different SPH parts to avoid the instabilities due to large density ratios at the interfaces. The methods allow the users to select the desired amount of contact force between two SPH parts by choosing the desired penalty scale factors according to the simulation problem at hand. Some examples are tested to show that the method was successfully used to stably simulate multiple fluids with large density contrasts without the above described artifacts apparent in standard SPH simulation.

Non-Linear Time History Analysis of Tall Steel Moment Frame Buildings in LS-DYNA

Carlos Molina Hutt (UCL)

® Non-linear time history analyses were carried out in LS-DYNA (LSTC) in order to assess the seismic performance of existing tall steel moment resisting framed buildings. Ground motion earthquake records representative of the Maximum Considered Earthquake (MCE) hazard level defined in current building codes were used in the analysis. This paper focuses on the different component models utilized to capture the complex non-linear elements of the structure: beams, columns, panel zones, splices and moment connections. Both beam and column elements were modelled using the Belytschko-Schwer element formulation with lumped plasticity at both ends of the resultant beam. Columns elements captured interaction between bi-axial bending moment and axial force, buckling in compression and degradation parameters for response under cyclic loads calibrated to match experimental tests results. Beams elements captured implicit degradation in bending and random fracture at the connections. The random fracture was modelled such that plastic rotation at fracture occurred as a random variable characterized by a truncated normal distribution following results from experimental testing. Panel zones and column splices were modelled with discrete elements and general nonlinear translational and rotational springs. Panel zones were modelled using the Krawinkler model by means of an assembly of rigid links and rotational springs to capture the tri-linear shear force-deformation relationship of the joint. Column splices were modelled as non-linear springs capable of reaching their nominal capacity with a sudden brittle failure in axial tension and/or bending and full capacity in compression as observed in experiments. The paper briefly discusses the limitations of complex analytical models in trying to capture the non-linear dynamic response of structural systems and components.

Numerical Determination of Permeability Tensor Components for 3D-braided Composites Using RVC Approach

Fedor K. Antonov (Research Institute of Mechanics of Lomonosov Moscow State University)

Today, one of the most promising trends in the design and manufacturing of composite structures is the use of 3D-reinforced thermoplastic composites. The present paper is concerned with the problem of resin transfer molding (RTM) process modeling, which is an important stage of 3D thermoplastic composites design. It is known that during the impregnation of woven preform local starved spots may occur, the textile pattern may distort and as a result the final structure will differ in mechanical properties form the initial design. The proper selection of RTM process parameters, such as injection holes placement, pressure profile and flow rate is a challenge for designers and process engineers. Nowadays, specialized software is developed for the solution of that problem, but the RTM process modeling in these environments is associated with considerable difficulties, caused by the need to set the permeability tensor, which components should be determined experimentally for each fiber material and for each weaving type. However, the permeability parameters can be determined in virtual experiments using the representative volume cell (RVC) approach by simulating a coupled-field problem of a viscous incompressible fluid flow through a porous medium. The paper demonstrates such an approach for permeability tensor components determination for the chosen 3D textile pattern using LS-DYNA with ALE computational method. Resin flow interaction with the fibers is modeled using FSI approach, while the flow through the fiber material is described by Darcy’s law. As a result, resin pressure drop curves along three RVC directions were determined, on the basis of which the permeability tensor components were obtained.

Numerical Evaluation of an Add-On Vehicle Protection System

Geneviève Toussaint, Amal Bouamoul, Robert Durocher, Benoît St-Jean (Defence Research and Development Canada), Jacob Bélanger (Numerica Technologies Inc)

Defence Research and Development Canada (DRDC) has been involved in programs to reduce the vulnerability of vehicles to mines and improvised explosive devices for many years. In this work, DRDC was mandated to design and optimize an add-on vehicle protection system. Testing an entire vehicle against improvised explosive devices or blast landmines is both time-consuming and “expensive; therefore, to reduce costs, numerical simulations using the LS-DYNA hydrocode [1] were performed to support the development phase of the protection system. A finite element model of a simplified full-scale model of the vehicle called a mock-up was developed for this purpose. This model included the critical sections that were likely to be affected by a threat. Finite element studies were then performed to compare the performance of several protection designs and to orient on the choice of the final one. This paper presents the loading model that was used to simulate the effect of a landmine on the structure, the finite element (FE) models of the baseline and of three concepts and finally, a comparison of the relative performance of the different protection system designs.

Numerical Simulation of Elastic-Plastic Deformation of Aircraft Fuel Tank Access Cover Impacted by Tyre Fragment

Alexander A. Ryabov, Vladimir I. Romanov, Sergey S. Kukanov, Anatoliy K. Botvinkin (Sarov Engineering Center)

All new-designed passenger aircrafts have to meet strict national and international safety requirements in accidents. One of the possible accidents is pneumatic tire damage, which can become the reason of the tire tread fragmenting. Some fragments of the tire tread with different mass and velocity can impact and break different plane elements. According to the safety requirements the designing company has to prove that the new plane withstands the tire fragment impact. The development tests to meet this requirement are obviously difficult and costly experiments, so it is very cost-effective to use a numerical simulation in the design of a plane to solve the problem. In this case the simulations results should be verified by the model experiments. The results of LS-DYNA£ numerical simulation of an aircraft fuel tank access cover elastic-plastic deformation are presented in the paper. The results are compared to experimental data, obtained for the covers subjected to tire fragment impacts at different angles with the speed of 110 m/s. Comparison of the cover residual deflection and strain time-histories shows that the simulation results are in a good agreement with the experimental data.

NVH and Random Vibration Fatigue Analysis of a Landing Gears’s Leg for an Un-Manned Aerial Vehicle Using LS-DYNA®

Al-Bahkali Essam, Elkenani Hisham (King Saud University), Souli Mhamed (Universite de Lille)

The present work concerns the new capability of LS-DYNA in solving Noise, Vibration and Harshness (NVH) and fatigue based on Power Spectrum Density (PSD) analysis. These new capability includes random vibration and high-cycle fatigue analysis in a random vibration environment. In this analysis, the cumultative damage ratio of a landing gear’s leg for an Un-Manned Aerial Vehicle (UAV) is computed using material S-N (Stress-Number of cycles) fatigue curve. Dirlik method is used for the analysis of life time as it is proven to provide accurate results for large number of applications, both in automotive and aerospace industry. It is also compared to other methods that have been developed in LS-DYNA as well. The input acceleration PSD data are provided through measurements.

On The Feng Failure Criterion For Composites

William W. Feng, John O. Hallquist, David J. Benson (LSTC)

In this paper, the criterion is evaluated with experimental data for a boron/epoxy, symmetrically balanced, angle-ply laminate. The results prove that the failure states obtained by the criterion agree with the experimental data. Furthermore, the criterion also predicts whether the failure of a composite is due to matrix or fiber. The advantages for Feng failure criterion over other criteria are summarized, and the criterion is implemented into LS-DYNA for composite materials.

Performance Evaluation Using LS-DYNA Hybrid Version on the K computer

Kenshiro Kondo, Hiroyuki Kanazawa (Fujitsu Limited), Kazuo Minami, Yukihiro Hasegawa (RIKEN), Hiroyuki Umetani (Japan Automobile Manufacturers Association, Inc.), Yu Setoyama, Takafumi Horita (Fujitsu Kyushu Systems Limited)

In order to improve the accuracy of the car crash analysis, the number of elements in the analytical model has been increasing rapidly. A large-scale analysis using model with 10 million elements is slowly becoming popular. Some Companies actually has been adapting large models in their crash analysis nowadays. ́7KH K computer ́> @ D KLJKO\ SDUDOOHO V\VWHP FDQ FDUU\ RXW D FDU FUDVK analysis in several hours which other supercomputers would have taken several days to complete the analysis in the past. However in order to achieve such efficiency, the analytical jobs have to meet following conditions: use LS-DYNA Hybrid version; deploy Groupable contact function of LS-DYNA; and reduce contact definitions as much as possible. In this paper, we investigated the performance and behaviour of LS-DYNA Hybrid version using VHYHUDO WKRXVDQG SURFHVVHV RQ 37KH K computer ́. More specifically, due to the critical role of contact on the performance in a highly parallel system, we mainly focus on the following two aspects of contact calculation part throughout the discussion: the relationship of computational time with the number of contact definitions; and the effectiveness of Groupable contact. This paper gives brief descriptions DERXW 37KH K computer ́ RI 5LNHQ DQG /6-DYNA Hybrid version used in this study in Sections 2 and 3. Several issues that have been encountered when carrying out crash analysis on such highly parallel computing environment are discussed in section 4. The performance bottleneck and factors that hurt scalability are investigated in the next section using simplified model to reveal the effect on the performance with respect to various contact patterns and different number of contact definitions. The effects of Groupable contact function of LS-DYNA versus different contact patterns are studied in this section, too. In the end it is concluded that Groupable contact functions of LS-DYNA help to improve performance of crash analysis on highly parallel environment. A modification of the model to reduce number of contact definitions can boost the performance of LS-DYNA on the K computer.

Performance Optimizations via Connect-IB and Dynamically Connected Transport Service for Maximum Performance on LS-DYNA®

Pak Lui, Gilad Shainer, Brian Klaff (Mellanox Technologies)

From concept to engineering, and from design to test and manufacturing, the automotive industry relies on powerful virtual development solutions. CFD and crash simulations are performed in an effort to secure quality and accelerate the development process. LS-DYNA® relies on Message Passing Interface (MPI) for cluster or node-to- node communications, the de-facto messaging library for high performance clusters. MPI relies on fast server and storage interconnect in order to provide low latency and high messaging rate. The more complex simulation being performed to better simulate the physical model behavior, the higher the performance demands from the cluster interconnect are. The recently launched Mellanox Connect-IBTM InfiniBand adapter introduced a novel high-performance and scalable architecture for high-performance clusters. The architecture was designed from the ground up to provide high performance and scalability for the largest supercomputers in the world, today and in the future. The device includes a new network transport mechanism called Dynamically Connected TransportTM Service (DCT), which was invented to provide a Reliable Connection Transport mechanism — the service that provides many of InfiniBand’s advanced capabilities such as RDMA, large message sends, and low latency kernel bypass — at an unlimited cluster size. The paper will review the novel Connect-IB architecture, the new transport service, and their performance effect on LS-DYNA simulations.

Plastic barrier for industrial applications – Characterization under impact loading using testing and Ls-Dyna®

C. Goubel (LIER SA), E. Di Pasquale (Sim Tech), J. Pascual (BOPLAN)

Safety structures are needed in industrial areas to protect people, machines or goods from forklift circulation. Steel roadside safety barriers are often chosen but the accuracy of these structures designed to restraint a light vehicle under a low angle but at high speed is not obvious. After several (even soft) impacts or contacts, this kind of devices suffers of a poor aspect and some parts have to be changed quite frequently. Plastic barriers developed by BOPLAN present the interest of elastic behaviour which allows to reduce significantly the number of repairs or maintenance operations on the one hand. One the other hand, the use of plastic structures is not a common choice and without any normative context concerning these products, commercial efforts are required to demonstrate the effectiveness of the structures. In order to help in the selection of the barrier type and in order to have scientific arguments, a meaningful collaboration combining testing, numerical simulation and mathematical models was set- up with the final aim to obtain a user friendly interface for structure selection as a function of conditions of use (Mass of forklift, customer regulation speed, etc…).

Prediction of laser welding failure on seat mechanisms simulation

M. Chauffray, G. Delattre, L. Guerin (Faurecia Automotive seating), C. Pouvreau (LIMATB)

For some years, FAUREClA has chosen to industrialize LASER welding technology (Fig. 1 ) to weld it seat components, This choice is a consequence of global lightweight policy in industrial automotive world, To fully answer to this requirement Faurecia needs to join thinner parts using higher strength steels. On one seat sevenl tens of welding lines will be used to join parts with thicknesses from O.S mm to 5 mm. Conventional processes such as resistance spot we/ding or metal active gas have reached their limits where laser welding offered again high potential. The LASER process oHen higher flexibility in term of weldable materials, thichness and seams geometry. LASER is also faster than conventional process and does not need filler metal. On new structures, all weldings are done by one process instead of 2 or 3 in the past, For all those reasdns Faurecia, decided to invest in this high potential joining process. In Faurecia, several join types are used, overlap scan welding, T-joins welding, edge to edge… This study will deal only with overlap scan welding, which represents the most impo_ant part of the seat structure welding, but could be extrapolated to others, For this study, we need to distinguish mo hinds of welding rupture. First is the rupture of the joint itself called melted zone rupture. Second one is the rupture of the material at the welding faot. This rupture appears in an area called Heat Affected Zone (see Fig, 27 and represents the main welding failure mode observed during development phase.

Recent Developments in Material Testing for Characterization of Materials (Deformation and Failure) for LS-DYNA Materials Models

Amos Gilat, and Jeremy D. Seidt (Ohio State University)

Several new testing methods that have been recently developed for mechanical characterization (deformation and failure) of materials are presented. The data from these tests is used for the development and calibration of material models (constitutive relations) in LS-DYNA. The first method involves the use of Digital Image Correlation (DIC) in tests that are used for generating data needed for the MAT224 model. In these tests specimens with different geometries are loaded and DIC measurements are used for determining the equivalent failure strain as a function of stress triaxiality and Lode parameter. The second testing method is a shear test for sheet metals. The experiment is done by using a flat notched specimen in a tensile apparatus. The shear strain is measured by using DIC within and on the boundary of the notch. The third development is a dynamic punch test in which the deformation of the specimen is measured continuously with 3-D DIC. The fourth is a high strain rate tensile testing technique for Kevlar cloth and Kevlar yarn in a tensile Split Hopkinson Bar (SHB) apparatus. The Kevlar cloth/yarn is attached by specially designed adaptors that keep the impedance FRQVWDQW ,Q DGGLWLRQ WR WKH WUDGLWLRQDO PHWKRG RI GHWHUPLQLQJ WKH VSHFLPHQ¶V strain from the recorded waves in the bars the strain is also measured with DIC. The fifth development is an apparatus for testing at intermediate strain rates in compression. In this apparatus the specimen can be deformed at strain rates ranging from 20 s-1 to 200 s-1. The apparatus is a combination of hydraulic actuator and a long (40 m) transmitter bar. The stress in the specimen is determined from the stress wave and the strain and strain rate is determined by using DIC. The results show very clean (no ringing) stress strain curves.

Recent Investigations of Side Curtain Airbag Deployment Simulation using CPM

Hisaki Sugaya, Kazuo Imura, Hiroyuki Mae (Honda R&D Co.,Ltd.)

Recently, Honda has shifted to a vi_ual development process to shorten the development term. In addition, the virtual development process has also been applied in the crash safety one. for this reason, it is necessary to build more accurate simulation modelling method. There are some reports, in which, airbag modelling Methods about gas flow has several technical issues [1] [2] [3I, in the crash safety components. Therefore, this research focuses on simulating the airbag folding properties and deployment behaviour. This study introduces the modelling method of the side curtain airbag (SCAB) with its complicated structure, leading to the accurate deployment simulation.

Relating scatter in occupant injury time histories to instability in airbag behaviour

Richard Brown (Jaguar Land Rover), Dominik Borsotto (Fraunhofer Institute SCAI), Clemens-August Thole (SIDACT GmbH)

Dealing with natural variation in input parameters and environmental conditions presents the automotive industry with significant challenges. Lack of consideration of variability in results can lead to unpleasant surprises during testing, with a consequent risk of unplanned cost and delay. In a purely virtual product development world, analysis techniques must lead to designs that are robust with respect to external noise sources, in order to minimise test-to- test and test-to-prediction variation. This paper discusses some of the issues faced in dealing with variability in an occupant restraint system, and presents an analysis approach that is helping to provide insight into causes of scatter, leading to potential design improvements to help reduce it. Conventionally the CAE process has used nominal values for input parameters, and has been satisfied with single, deterministic solutions. In many cases this approach is based on unreasonable assumptions, and a structured consideration of variability is vital. In this context we describe an example where Principal Components Analysis has been used to study scatter in an airbag model. Building on previous experience with the application of this technology to deformed geometries, the technique has been extended to allow a consideration of scatter in curves, as exemplified by the set of chest acceleration time-histories shown in figure 1. The mathematical background to the PCA method, as implemented in Diffcrash, is presented, and its extension to curves is explained. It will be shown how scatter in two crash dummy channels can be related to each other and to airbag deformation behaviours, as an aid to developing design improvements. Virtual techniques have much to offer in understanding and managing scatter in physical systems, and the consideration of variability in the CAE process is slowly becoming more common-place. The PCA approach presented here is a useful addition to the toolset available, giving valuable insight into physical phenomena.

Seismic Response of Baffled Liquid Containment Tanks

Zuhal Ozdemir (University of Sheffield), Yasin Fahjan (Gebze Institute of Technology), Mhamed Souli (Université des Sciences et des Technologies de Lille)

The failure of liquid storage tanks due to earthquake induced sloshing action of the liquid was extensively observed during many past major earthquakes. The destructive effects of sloshing can however be suppressed in a passive manner by introducing additional sub-structures such as baffles into tanks. The main aim of constructing these sub-structures is to alter the period of sloshing action beneficially and to increase hydrodynamic damping ratio. The main aim of this paper is to numerically quantify the effect of baffles on the response of 2D rigid tanks. In this paper, LS-DYNA program is chosen as a numerical analysis tool due to its high degree of flexibility. The numerical model is first verified using an existing numerical study in the literature and a strong correlation between reference solution and numerical results is obtained in terms of sloshing wave height. Following the verification of the numerical model, the hydrodynamic damping ratio of sloshing in a 2D rigid baffled tank is assessed for different baffle positions. Finally, a parametric study is carried out on 2D rigid tall and broad baffled tanks in order to assess the effect of baffle on the sloshing wave height under different earthquake motions.

Simplified modeling of self-piercing riveted joints for crash simulation with a modified version of *CONSTRAINED_INTERPOLATION_SPOTWELD

Matthias Bier, Silke Sommer (Fraunhofer Institute for Mechanics of Materials IWM, Germany)

The requirements for energy efficiency and lightweight construction in automotive engineering rise steadily. Therefore a maximum flexibility of the used materials is necessary and new joining techniques are constantly developed. The resulting large number of joints with different properties leads to the need to provide for each type of joint an appropriate modeling method for crash simulation. In this paper an approach for a simplified model of a self-piercing riveted joint for crash simulation will be discussed. The used simplified model is a modified version of the *CONSTRAINED_ INTERPOLATION_SPOTWELD. Firstly the realized modifications as a changed yield and failure behavior will be explained and illustrated. Therefore simulation results of the default and the modified version of the *CONSTRAINED_INTERPOLATION_SPOTWELD will be compared. Secondly the procedure to identify the appropriate model parameters will be presented and shown in an exemplary manner. At once the advantages and limitations of the model will be demonstrated. At least the quality of the model will be validated using simulations of different loaded T-joint experiments, which represents the connection between the rocker panel and the B-pillar. For this purpose three different characteristics will be taken into account: the global responses like the calculated and measured force vs. displacement curve of the punch, the local failure behavior and the order of failure of the rivet joints, and at last the internal forces in the simplified model.

Simulation of Warm Forming of 5754 Sheet Aluminium

Trevor Dutton (Dutton Simulation Ltd)

In November 2009 a project was set up to implement an innovative metal forming process into the automotive industry with the goal of producing lightweight, high accuracy, complex-shaped automotive aluminium panels using one main forming operation. The project was known as WAFT – Warm Aluminium Forming Technology – and was part-funded by the UK Technology Strategy Board. The opening premise was that increased formability could be achieved with existing aluminium grades when heated to temperatures in the range 200°C to 350°C [1]. At these temperatures the material does not undergo re-crystallization or achieve superplasticity yet still exhibits increased formability – but the optimum settings for blank and tool temperatures, and also forming rate, were not known. The project aim was to industrialise the warm forming concept to align with conventional cold processing in order to develop a manufacturing process that could achieve steel formability with aluminium. This was to be confirmed in an industrial cell running a demonstrator tool, at rates optimised for premium vehicle production. The grade of aluminium chosen for the study was 5754; this is widely used for cold forming automotive body-in-white structural panels, and issues regarding assembly and behaviour in the vehicle are well understood. However, the reduced formability of 5754 compared with steel drives body-in-white design to adopt simpler forms and more numerous parts in sub-assemblies to create the required levels of complexity – all of which has significant cost implications and an impact on the overall carbon footprint of the manufacturing process.

Springback Analysis and Optimization in Sheet Metal Forming

Abdulaziz Alghtani, P.C. Brooks, D.C. Barton, V.V. Toropov (School of Mechanical Engineering University of Leed)

An accurate prediction of springback in sheet metal forming processes requires complex hardening material models. In this research, numerical analysis of the springback in U-bending was conducted using the well-known Yoshida model, available and known as the YU model in LS-DYNA. This model has seven main parameters which describe the behaviour of the material as it undergoes metal forming processes. Initially, mesh sensitivity studies were conducted to derive a suitable mesh that represents an appropriate compromise between accuracy and computer time. Secondly design of experiment (DoE) was employed to make 30 combinations of two design variables (die radius and clearance) uniformly through a design space. Parametric optimisation studies were also conducted to investigate the influence of these variables and to make recommendations to minimise the springback. The results show that the blank element mesh density has a significant effect on the springback prediction. Additionally the results demonstrate that certain geometrical parameters have a significant impact in controlling the springback but that optimised values can be identified to minimise the effect.

Stochastic Simulation of Aircraft Fuselage Assembly Considering Manufacturing Uncertainties

Dietmar C. Vogt, Sönke Klostermann (EADS)

In aircraft production the use of rivets as permanent mechanical fastener to assemble lightweight sheet metal structures is very common. At assembly the rivet is placed in a through boring and the buck-tail is plastically deformed to create a second head. Thus rivets are positive locking and can carry axial tension loads. However, rivets are mainly used to transfer shear loads via the seating stress of their cylindrical shaft. The remaining pre-stress in the rivet and its local area after the riveting process is subject to immanent manufacturing scatter. When assembling the fuselage of commercial aircrafts additional inherent uncertainties are impacting the riveting process. The cylindrical barrels of a fuselage are typically manufactured from large thin walled shell structures underlying geometric tolerances and variations of the boundary conditions. Managing these uncertainties has a significant impact on the geometrical and structural product quality. In this paper the resulting variations of the three-dimensional residual stress condition will be analysed by simulation. To simulate the fuselage assembly process the model must be able to predict the influence of manufacturing uncertainties appropriately. Therefore these uncertainties will be considered already during the modelling by stochastic parameters and random fields. While most application examples in the literature are quite simple [1], the present paper aims to apply random ® fields in an industrial application using LS-DYNA . By utilizing non-invasive methods the approach can be adapted to different FE-Solvers without too much effort.

Stochastic Spray and Chemically Reacting Flow in LS-DYNA

Kyoung-Su Im, Zen-Chan Zhang, Jr. Grant O. Cook (LSTC)

The injection of fuel sprays into an automotive engine and liquid jets into a high-speed flow stream is an important process in modern automotive gasoline and diesel engines, and propulsion in gas turbine and supenonic vehicles. ln such applications, the combustion peformance depends strongly on spray atomization, penetration, and the mixing process bemeen the free stream air and the liquid fuel. As a result, the study of liquid spray in such areas has become an important research topic.

Strain-softening in continuum damage models: Investigation of MAT_058

Karla Simone Gemkow, Rade Vignjevic (Cranfield University)

Composite materials are of increasing interest to automotive and aviation industry due to their high strength and stiffness. Therefore they are commonly used to replace metallic materials. However their mechanical behaviour is complex, especially when damage is considered. Composite damage leads to degradation of material properties which results in behaviour known as strain-softening. ® An implementation of strain-softening in numerical codes, such as LS-DYNA , leads to mesh sensitivity of results and therefore those models are not reliable. The user of damage models with strain-softening needs a good understanding of those material models to evaluate results critically. This work aims to provide an insight on strain-softening behaviour in a mathematical sense and its consequences on numerical codes. An analytical solution is derived for a one-dimensional dynamic bar problem which allows a direct comparison with numerical results. It was found that deformation localises in an area which is governed by the chosen element size and therefore causes mesh sensitivity. Strain grows infinitely in the strain-softening area with a simultaneous drop of stress. Outside the strain-softening area the problem unloads elastically. The dissipated energy tends to vanish.

The pressure response in the brain during short duration impacts

C. Pearce, P.G. Young (University of Exeter), L. Cowlam, and B. Walker (Arup)

The mechanisms which lead to brain injury in response to blunt head impacts are investigated using three finite-element models of the human head, which range from low to high biofidelity. The models were developed directly from MRI image data using a technique adapted from the marching cubes approach which automates the generation of meshes and allows for a number of different structures (e.g. skull, scalp, brain) to be meshed simultaneously. Experiments were carried out on the finite- element models to validate an analytical representation of head impact based on full 3D elasticity equations developed by one of the authors, and good agreement was observed. The analytical and numerical models were used in parallel to explore the phenomenon of large transient pressure magnification in the brain. This behaviour, proposed by one of the authors, occurs as a result of low duration low velocity impacts. The implications of these high pressure transients are also discussed. Finally individual case studies demonstrate the relevance of this research to realistic head injury scenarios.

Time and rate dependent constitutive model coupled with nonlocal damage at finite strains for semi-crystalline polymers

Romain Balieu, Franck Lauro, Bruno Bennani (Univ Lille Nord de France), Tsukatada Mastumoto, Ernesto Mottola (Toyota Motor Europe)

Many constitutive models were developed in the literature to model the complex behaviour of polymer materials. These models can be sorted in two categories: the physical based models where the microstructure of the material is taken into account for representing the macroscopic behaviour [1,2] and the phenomenological based models where the material discontinuities, in the microstructural scale, are homogenised in a representative volume element. In this way, elasto-plastic constitutive models based on the 3RYHUVWUHVV ́ FRQFHSW 9%2 > @ Xsing the unified state variable theory were extended for polymeric materials [4,5]. The addition of mineral fillers in the semi-crystalline matrix increases the cavitation phenomenon. In this case, the viscoelastic-viscoplastic deformation of the material is accompanied by damage in the form of nucleation, growth and coalescence of cavities. Many damage model were developed for polymer application in order to represent this phenomenon [6,7,8,9]. The damage present in this kind of material induces a softening behaviour which leads to the localisation of the strain in a narrow zone of the structure accompanied by numerical solutions depending of the finite element mesh. The nonlocal model where introduced in the literature in this way, in order to overcome the mesh dependency phenomenon [10,11]. In this work, a non-associated viscoelastic-viscoplastic model coupled with nonlocal damage is developed in order to model a mineral filled semi-crystalline polymer used in the automotive industry. The constitutive equations of the model are stated under finite strain framework by using a hypoelastic formulation. The interesting properties of the logarithmic tensor linking the work conjugate pair Cauchy stress and Henky strain are used in the proposed model. In order to obtain a mesh independent solution with the material exhibiting softening, an integral-type nonlocal damage is developed in this work.

Topology optimisation method for crashworthiness design using Hybrid Cellular Automata and thin-walled ground structures

Stephan Hunkeler, Milan Rayamajhi (Queen Mary University of London), Fabian Duddeck (Technische Universität München)

Crashworthiness is one of the most demanding design cases for vehicle structures. Until a few years ago, it was mainly addressed using trial and error approaches; but recently, automated structural optimisation for crashworthiness design got more and more popular. So far, most relevant applications use size or shape optimisation. Nevertheless, the ultimate way to achieve significant mass reduction is to use topology optimisation. While topology optimisation for static mechanics is a well established field of research, applications to crashworthiness can be rarely found. Due to high non-linearity of crash simulation, classic topology optimisation methods cannot be applied directly to crashworthiness design. Therefore, alternative methods have been developed. This paper first presents the available methods for topology optimisation in crashworthiness design. A discussion of these methods highlights the opportunity to develop an alternative method which is detailed in the second section of this paper. Then two application examples are presented to showcase the interest and capabilities of this method.

Transitioning From SMP To MPP-DYNA3D For The Simulation Of Large Thermal-structural Implicit Problems

Dr. Gurdip S. Kalsi (AWE)

The LS-DYNA family of codes have been used at AWE for many years. For a long time they were used on our shared memory platforms (SMPs) to carry out implicit structural and coupled thermal- structural analyses, amongst others. Over time processor speeds have continually increased and larger memory has become available at reducing costs. This has led to an increase in the size of models as meshes have been refined for better definition and realism of the problems under investigation. However the simulation of the long-term responses of engineering structures poses special difficulties when large models need to be analysed encompassing non-linear behaviours. These non-linearities can arise through sliding interfaces, and more commonly through the complex constitutive responses of non-traditional fabrication materials, such as foams and explosives, that act as structural, load-bearing components. Unlike explicit analysis, in implicit problems the equations cannot be decoupled from each other, and so implicit simulations immediately make large demands on the amount of memory required to solve the problem in-core, and these requirements increase rapidly as the model is refined. For the most complex analyses the turnaround times can grow from weeks to potentially months, as model size increases. This problem is being addressed by re-writing implicit solvers to run in parallel mode on distributed memory platforms (DMPs). Although these developments have helped these codes to reduce turnaround times, work is required to further enhance their scalability. Shared memory and MPI-versions of LS-DYNA have been used at AWE to investigate the transition from SMPs to DMPs for the solution of large, contact-dominated thermal-structural implicit problems. The hybrid version of these codes was also used in some simulations, but this is early work at AWE. This paper reports our findings. It also examines the influence of code characteristics on computing platform requirements. The significant reduction in turnaround time that was realised using MPI instead of the SMP version for a major test problem will be presented, and the scaling characteristics of the MPI and hybrid versions of LS-DYNA for this problem will be shown.

Transmissionlosssimulation of acoustic elements in LS-DYNA®

Marko Krebelj (Akrapovič d.d.)

This paper presents validated simulations of transmission loss in LS-DYNA for basic acoustic elements in exhaust systems. According to [1] there are several indicators available that describe the acoustic performance of an exhaust system and its components. These mainly include transmission loss (TL), insertion loss (IL) and noise reduction (NR). The TL is a ratio of sound power level between the inlet wave entering and the transmitted wave exiting the element. Acoustic element termination has to be anechoic, since the TL is a property of the acoustic element only. The NR is sound pressure level difference across the element. The IL is the loss of sound power from the insertion of an arbitrary acoustic element. In this paper we will focus on the TL only. There are several applicable methods in use to measure the TL. The most common and popular approach for measuring transmission loss is decomposition method or VRPHWLPHV FDOOHG 3WKUHH-pole PHWKRG ́ The method is based on the decomposition theory. The basic idea of the method is that the sound pressure may be decomposed in its incident and reflected waves. When the pressure wave is decomposed, the TL can be calculated. 6RXQG LV RQH RI WKH NH\ WULJJHU HOHPHQWV WKDW PDNH FXVWRPHU ZDQW WR EX\ DQ $NUDSRYLþ H[KDXVW system. Well-done sound solutions have been so far developed with the testing of different exhaust system prototype configurations. This demands time and costs to build and test every new configuration. Therefore, we had been always looking for faster solutions. The goal in our company was to create measurement-validated probational simulation models in LS-DYNA to examine the acoustic performance of such parts. Acoustic component performance prediction is a good example of the use of simulation software in industrial applications. LS-DYNA is one of the widely used finite element codes for solving complex mechanical problems. One of the recent developments is the addition of a vibro-acoustic solver, which enables users to perform a number of vibro-acoustic analyses in the frequency domain. In order to obtain a numerical solution in our case we have used the recently implemented *FREQUENCY_DOMAIN_ACOUSTIC_BEM keyword in LS-DYNA. This new keyword allows users to run acoustic computations based on boundary element method (BEM).

Ultra High Power applications designed using the LS-DYNA EMAG solver

Gilles Mazars, Gilles Avrillaud, Anne-Claire Jeanson, Jean-Paul Cuq-Lelandais (Bmax)

Bmax offers industrial solutions based on Ultra High Power (UHP) technology and provides manufacturing solutions for Magnetic Pulse (MP) Forming, Welding and Crimping as well as Electro- Hydro-Forming (EHF) processes. The LS-DYNA® Emag solver version 1.7 used in the Eddy’s current approximation allows modelling different parts of the system like coils, workpieces and the dynamic interactions between them. This module is used as an advanced design tool to for example optimize the Lorentz forces generated in the workpiece. To achieve accurate numerical results, many comparisons have been done using Photon Doppler Velocimeter (PDV) measurements implemented in specific test beds. The described case is of a tube crushing. We will also show how LS-OPT® identification and sensitivity capabilities have been used for material identification using the Magnetic Pulse Forming process. The main objective is to characterize the material dynamic behaviour in the strain rate ranges generated by the Bmax processes (around 1000 to 10,000/s). To summarise the LS-DYNA and the LS-PrePost® capabilities greatly aides the engineering team to provide a feasible industrial solutions from tool design to building predictive models to study the feasibility of a project. We can then now respond faster to customer specifications, but also better understand the physical processes and ultimately reduce design costs by limiting the number of physical testing.

Unsteady Aerodynamic Analysis around Oscillating Ahmed body by LS-DYNA ver.980

Hiroyuki Sawamoto Tsuyoshi Yasuki, Hiroshi Tanaka, Kazuyoshi Ishii (Toyota Motor Corporation)

This paper describes an aerodynamic effect of the Ahmed body model in sinusoidal pitching motion. Relationships of pitching angle to lift coefficient were analyzed by computational fluid dynamics (CFD) of LS-DYNA ver.980. Those numerical results showed good agreements with wind tunnel test results reported in the past. The flow field was investigated by using this numerical analysis result and the relationship of bottom surface pressure distribution to the lift coefficient was clarified.

UsingLS-OPT for Parameter Identification and MAT_FABRIC with FORM=-14

David Dubois (Autoliv), Jimmy Forsberg (DYNAmore Nordic AB)

This work was carried out as a methodology development project in a joint venture between Autoliv and DYNAmore Nordic AB. The outset of the project was to obtain a better component behavior due to a more realistic material behavior in the simulation of airbag models. The observation underlying the project was that the current fabric model used in most airbag models is *MAT_FABRIC and FORM=14. In FORM=14 there is no consideration taken to a stiffened response due to a bi-axial stress state in the fabric. To consider the bi-axial stress state, FORM=-14 was implemented some years ago but has, until now, not been used. The objective with this implementation is to increase the stiffness in the fabric when subjected to a bi-axial stress state. This paper presents a resume of the features found in *MAT_FABRIC, a methodology to fit the simulation model to material test data using LS-OPT and finally a comparison between the behavior of the different FORM options.

Validation of Mine Blast Simulations with Field Tests

İlker Kurtoğlu, Berkay Salihoğlu, Y. Caner Taşan, Gökhan Tekin (FNSS Savunma Sistemleri A.S.)

In this work, LS-DYNA® simulation results of mine blast of flat aluminum/steel plates and validation studies using mine blast tests are presented. The buried mine simulations are performed using ALE initial volume fraction and SPH methods. The high strain rate Johnson Cook material models of the plates are obtained through Split Hopkinson Pressure Bar tests, for both high strength steel and aluminum specimen. The soil parameters such as density and humidity ratio are determined by appropriate tests. Structural acceleration and strain data as well as blast pressures are measured during the field mine blast tests. A test setup is designed and manufactured in order to conduct the tests. This setup consists of dead weights for constraining the plate during the explosion. For measuring the plastic and total deformation of the plates, a novel method using thin walled aluminum cones are used. High speed and high bandwidth data acquisition systems are used to capture the highly dynamic behavior of the plates. Moreover, incident and reflected blast pressures from various distances are measured and compared with analytical methods, LS-DYNA simulations and bikini gage measurements. In this way, explosive and blast characteristics are verified. The correlation and similarity of simulation and test results for acceleration and deformation characteristics are presented. The results obtained by LS-DYNA simulations show very good agreement with results obtained in the field tests.

Validation ofModel fora Thermoplastic Composite Material Low Carbon Vehicle Applications

Oliver Tomlin (GRM Consulting Ltd), Neil Reynolds (WMG, University of Warwick)

The methods developed for creating and validating a thermoplastic composite (TPC) material model in LS-DYNA are presented. Included are details of the mechanical characterisation methods and test types required to fulfil the input requirements of MAT058. Model input data are validated through correlation between coupon and sub-system physical tests and simulations with justification of the boundary conditions. The test methods are explored in more detail following the sub-system model validation, giving a clear understanding of the need for integration between CAE and test. Validation of the material model is further explored with results from quasi-static and impact tests. Results show that confidence in the predictive capabilities of MAT058 is high.