NIO are a global automotive startup producing electric vehicles for the China market. Our second vehicle, the ES6, was unveiled in December 2018 in Shanghai. It features a lightweight carbon fibre floor body structure, which will become the first high volume CFRP production part in ASIA. This presentation describes the CAE activities undertaken to develop the composite body structure. It explains the approach that was taken to construct the DYNA material cards and the various material tests used to validate them. It explores the various CAE activities used to develop and optimise the design of the parts and the layups of composite layers, and then the successful validation of the parts.
During the development process of a new platform or car model, each design iterate is subjected to a large number of load cases, both dynamic as well as static. At Volvo Car Corporation, this process is almost entirely carried out using virtual testing by finite element analysis. The amount of physical prototypes is reduced to a minimum, and in many cases physical testing is limited to the component or sub-assembly level. Still, the final design must pass a number of physical tests and legal requirements, where roof crush is an important test of the structural integrity of the cab. The purpose of the FMVSS roof crush test is to “reduce deaths and injuries due to the crushing of the roof into the passenger compartment in rollover accidents” [6]. At Volvo Car Corporation, occupant safety is a fundamental element in all development projects since the start of the company, and the Volvo XC40 received a 5-star rating when tested by Euro NCAP [7]. The roof crush resistance is important with relation to safety in case of a roll-over accident, since the structural integrity of the car body makes the final line of defense, but many safety systems will interact in this case, from driver assist systems to electronic stability systems and restraint systems. The roof crush test will induce high stresses in many structural parts of the car body, for example the A-, B- and C-pillars, window frame and roof. This means that the analysis must be carried out meticulously, since the roof strength requirement may set design limits for many structural parts. Also new design concepts, such as composite roof panels or panorama glass roofs, imply new challenges for the roof crush analysis. The testing procedure according to FMVSS 216 [6] is specified as quasi-static (the time to complete the test is minimum 10, maximum 120 seconds), but has traditionally been run in explicit LS-DYNA in only a fraction of this time. From this viewpoint the roof-crush load case would be a typical application of implicit analysis, allowing the simulation of the test to get closer to the real test procedure. As a part of the ongoing method development work, it was decided to evaluate also the implicit technique, using the Volvo XC40 as a benchmark case. A previous study [5] indicated that it is possible to re-use FE-models originally created for crash load cases also for quasi-static load cases using the implicit solver in LS-DYNA with a reasonable modification effort. A previous study on implicit roof-crush analyses in LS-DYNA [1] indicated good correlation to explicit analyses, as well as reasonable performance with respect to solution time. Also, the publicly available examples [2][3] of implicit roof-crush analyses served as great inspiration in the present work.
The mesh behaviour and convergence rate of six hexahedral element formulations in LS-DYNA were investigated by means of crashworthiness simulation. The element formulations are: constant stress solid element (ELFORM 1), fully integrated S/R (Selective Reduced) solid element (ELFORM 2), fully integrated S/R solid element with reduced transverse shear locking (ELFORM -1 and ELFORM -2), 20-noded serendipity element (ELFORM 23), and 27-noded fully integrated S/R quadratic solid element (ELFORM 24). FE-simulations of the axial crushing of aluminium profiles were set up with these element formulations. The convergence rate of each element formulation was investigated by varying the mesh resolution. For validating the simulation results, four extruded profiles with rectangular hollow cross-sections were experimentally tested under quasi-static axial crushing load. On that basis, the performance of each element formulation was investigated in terms of their convergence rate, accuracy, and computational cost to elaborate an approach for future tasks. Finally, various aspects which should be considered while using these element formulations for this class of problem are discussed.
Playing with LEGO® bricks is something many engineers might have enjoyed during their childhood. Building any kind of mechanical construction allows creativity and complexity to an extent which probably contributed to their fascination and finally to their decision of becoming engineers. It’s interesting to see how many of them are still fascinated by LEGO® even in their adult life. Especially for children, or for those with an active inner child, crashing these models into each other is even more fun, because seeing all those bricks fly all over the place is just fascinating, beyond any scientific or professional aspect.
The improvement of tractive performance on ice is one of the most challenging aspect in the nowadays tire industry. For this reason, a model which can predict the friction coefficient on ice can be useful in the winter-tire design. However, the highly multiphysics nature of the interaction between rubber and ice [1-4], as well as the magnitude of the dimensions involved make the development of a numerical model a quite complex issue. In this work, a first step for the prediction of the friction coefficient on ice is proposed using the finite element method. The subject of this analysis is the transient phase of the sliding interaction between a rubber block and an ice surface. The User Define Friction module of LS-DYNA has allowed to implement the suitable friction law for contacts with ice, widely used in the literature [3, 5], which follows a microscopic approach and it is based on a viscous formulation. Considering again the dimensions involved and the duration of the transient phase, it is impossible to directly validate the model through experimental detection [3]. However, in the subsequent steady-state phase, which involves higher amounts of water and longer time, the experimental measurements are easier. To compare the results, in the literature an indirect procedure was used in order to provide a qualitative validation, using the strict link between the transient phase and the steady-state one. The final comparison between the LS-DYNA results and the literature results has shown a good correlation level.
Development of Carbon Fibre Floor Structure for Premium Electric SUV