Reinhard Mauermann

Numerical and Experimental Analysis of Self Piercing Riveting Process with Carbon Fiber-Reinforced Plastic and Aluminium Sheets

In this study a numerical simulation model was designed for representing the joining process of carbon fiber-reinforced plastics (CFRP) and aluminum alloy with semi-tubular self-piercing rivet. The first step towards this goal is to analyze the piercing process of CFRP numerical and experimental. Thereby the essential process parameters, tool geometries and material characteristics are determined and in finite element model represented. Subsequently the finite element model will be verified and calibrated by experimental studies. The next step is the integration of the calibrated model parameters from the piercing process in the extensive simulation model of self-piercing rivet process. The comparison between the measured and computed values, e.g. process parameters and the geometrical connection characteristics, shows the reached quality of the process model. The presented method provides an experimental reliable characterization of the damage of the composite material and an evaluation of the connection performances, regarding the anisotropic property of CFRP.

The effect of cushion-ram pulsation on hot stamping

Hot stamping is an important technology for manufacturing high-strength components. This technology offers the possibility to achieve significant weight reductions. In this study, cushion-ram pulsation (CRP), a new technology for hot stamping on servo-screw presses, was investigated and applied for hot stamping. Compared to a conventional process, the tests yielded a significantly higher drawing depth. In this paper, the CRP technology and the first test results with hot stamping were described in comparison to the conventional process.

Improving the Ramp-Up Process of a Body-Construction Line by Numerical Supported Design of Clamping Devices and FEM Based Tolerance Prognosis

The use of FEM (finite element method) to assistant in ramp-up processes of car body construction lines is increasing, thanks to developments in recent years [1-3]. Car body manufacturing begins with sheet metal forming, while in subsequent steps the inner structures of the vehicle are assembled and connected to the outer skin by hemming. With reference to the current state of the art, there is no methodology which can reliably predict the dimensional accuracy of body parts through metal forming [4].Additionally, several methods to predict the distortion of joining and the dimensional effect of clamping during the assembly process were presented and validated [4-11]. Dimensional effects of the clamping process are basically the result of a deliberate alignment, other than the given values of construction to compensate dimensional inaccuracy of single parts from the body shop. These deliberate alignments are generally effected through a translation of clamps and pins in the clamping device. Until now, most of the methods of clamping and joining simulation presented have been verified using academic samples.In this report, the quality of forecasting in real problems during a ramp-up process will be verified and expanded. As part of a national project, co-funded by Sächsische Aufbaubank (SAB), the potential of FEM to assist in the ramp-up process were reviewed in a cooperative effort between Porsche Leipzig GmbH and Fraunhofer Institute for Machine Tools and Forming Technology (IWU). Furthermore, it will be shown that developed methods are able to represent the influence of deliberate positioning of clamps in complex samples. For the first time the quality of forecasting through the translation of locating pins is numerically and experimentally qualified.

Tolerances and Measuring Strategies in the Virtual Process Chain for Spot Welded Structures; Substitute Modeling and Automating the Calibration Procedure

This paper presents a method with whose help it is possible, to quickly and precisely predict the influence that thermal spot-shaped joining processes has on the dimensional stability of complex component structures even in early planning phase. The welding distortion is calculated in the context of reduced computing time, based upon an experimentally calibrated mechanical substitute model. This expands existing approaches of substitute models and defines both an experimental and numerical procedure for creating adequate calibration samples. In turn, this makes use of the potential obt ained for standardizing the experimental basis for calculating and modelling the distortion to automatically carry out painstaking calibration processes in simulations and experiments in future based upon mathematical model functions. Finally, the limits to applying the substitute spot welding model are verified with reference to its predictability using a complex joining situation of a car body construction.

Deep Drawing with Local Hardening on Digital Multi-axis Servo Press

The paper discusses a new drawing technology, based on a synchronized movement of ram and cushion with multiple bending operations in alternating directions called “bi-directional deep drawing (BDD).” The goal is to avoid local thinning by strengthening the weak point using local hardening. BDD operations are realized before the conventional deep drawing process. This results in a local strain hardening at the weak point of the workpiece, which is usually located at the bottom punch radius. Two major aspects have to be given attention due to the high number of process parameters. On the one hand, for process design, it is helpful to have a tool by means of which it is possible to simultaneously create both the machine program for the servo press and the initial configuration for the process simulation. From the authors’ point of view, this complexity can only be represented by a numerical analysis method, on the other hand. Consequently, both aspects are given attention in this paper.

Improving Surface Based Clamping Simulations by Measuring the Position of Active Surfaces during the Clamping Process

The manufacturing process of body parts starts with the step of sheet metal forming. The single parts, produced at the press shop, are put into clamping devices in order to align and to fix them. The fixation takes part before further operations like joining can be carried out. In order to simulate the process chain of add-on body parts realistically, the clamping process (closing the clamping device) has to be taken into account. The stationary surfaces of a clamping device are called passive and the moveable surfaces are called active surfaces. If the clamping process is calculated by means of active surfaces, their positions need to be measured in the state of a closed clamping device. While the passive surfaces of a body construction device can be measured with high reproducibility, the measurement of active surfaces in the state of a closed device is impracticable because of the loss of accessibility. Furthermore, if the parts to be clamped or the position of the clamping device differ from their designed position, the assembly works like a flat spring against the clamping device force in all spatial directions. The active surface does not reach the position which was measured before. In order to take these facts in clamping simulations into account, the end position of the active surfaces should be known. A clamping device concept on the basis of a measuring probe for optical measurement systems was developed. It is possible to determine the position of active surfaces with high reproducibility while the parts are clamped. It can be shown, that the presented clamping device concept contributes to significantly better results of clamping simulations. Thus a better starting basis for further simulations along the process chain is offered.

Test Procedures and Simulation Model Calibration Strategies for Nonlinear Stiffness Behavior of Joints in Large-Scale Structures

State-of-art models for mechanical joints in large scale structures typically consider only the linear behavior of the joint zones with lower complex approaches, such as rigid or elastic beams or a merge of opposite sheet metal nodes. In the present study several feasible methods to model nonlinear joint behavior and the connection between sheets and joint are investigated and evaluated. A preferred combination based on nonlinear springs was chosen, which meets the requirements for application in large scale structure models: low computation time, mesh independence and availability in several FEM software packages. For the calibration of the joint zone models a 2-point-tension-specimen was used. Five different joint types and the two sheet material combinations aluminium/aluminium as well as steel/steel were investigated. With the calibrated models a more complex 5-point-tension-specimen was used to consider the local interoperation of the joints. Some deviations were determined especially for highly stressed joint zones. Hence an average function was defined to consider both, the local deformations in the joint zone and additionally the more global sheet deformations. Finally, the simplified joint models were used in a complex specimen model with 22 joints. The comparisons between experimentally and numerically determined results show a good accordance. The nonlinear joint behavior is captured very well. A method is presented, which uses 2-point-specimens to calibrate simplified joint models with nonlinear deformation characteristics. The efficient application in large scale structure models is possible due to simplicity, stability, low computation times and mesh independent implementation.

Predicting Dimensional Accuracy of Laser Welded Aluminum Add-On Body Parts

Laser welding of complex aluminum add-on body parts such as vehicle doors, is a common joining technology in the automotive industry. Besides the many advantages (e.g. high processing speed) laser welding provides, temperature induced distortions are an important task to deal with. In the last twenty years, several simplified FE methods, which predict welding distortion (weld seams, spot welds) of large assemblies, were presented. In order to simulate the distortion of large car body components properly, realistic clamping conditions need to be considered [1, 2, 3]. Furthermore, the calibration process of simplified models has to be examined systematically, to find out their limits and achieve optimal simulation results [4]. In this paper, a new FE model is presented to predict distortion of laser welded structures, based on a shrinkage volume approach. Effective surface based clamping conditions (derived of the real clamping device) and effects of previous forming processes are considered. The simplified model was examined due to an extensive design of experiments. Not only simple, but even complex simulated specimens match with the experimental results very well.