JEC COMPOSITES MAGAZINE - Issue #66 - July/August 2011 - 33

Today, these models cover all of the abovedescribed material properties and can be applied to both static and dynamic testing of automotive parts.

Fig. 3: Two different time points in this dynamic event have been investigated for homogeneous isotropic and inhomogeneous anisotropic materials (courtesy of JSOL Corporation) Fig. 4: Injection-moulded part with varying fibre orientation for a sunroof system (courtesy of Ticona and Inteva Products)

Fig. 2: Intake manifolds are prone to dynamic pressure loads caused by engine piston backfire (courtesy of JSOL Corporation)

Fig. 5: Comparison between isotropic and anisotropic approaches for two load cases. The DIGIMAT approach correlates excellently with the experiment (courtesy of Ticona and Inteva Products)

Applications
The use of micromechanical material models proves highly advantageous. Compared to the traditional isotropic approach, vast differences can be seen in terms of part performance prediction. A typical injection-moulded plastic part such as an air intake manifold should be able to withstand a high pressure peak resulting from piston backfire (see Figure 2). Two different time points in this dynamic event were investigated during the simulation (see Figure 3). A dramatic, qualitative difference in performance was observed between the traditional isotropic and the state-of-the-art micromechanical approach. The traditional method completely fails to predict the stress distribution not only over the part but also in the time domain, whereas DIGIMAT software modelling provides a realistic picture. To overcome this well-known limitation of isotropic models, a scaling factor is usually introduced to describe an average of the material properties as varying over the part. However, this procedure is not predictive as it is not possible to derive a unique scaling factor from a rule of thumb. As a consequence, even though experimental curves might be fitted via this approach, the simulation itself is not predictive. This is clearly seen on a plastic part for a sunroof system where two load cases – global and local loading – were

investigated (see Figure 4). To fit the experimental curves for both cases, vastly different scaling factors had to be used (see Figure 5). Conversely, re-running the analyses with one single DIGIMAT material directly gave excellent correlation with the experiment.

Future challenges
In a similar manner, many types of composite properties have been investigated so far using the DIGIMAT approach. As a result, this technique is developing a standard on the market. Engineers can correctly predict static or dynamic scenarios including the failure behaviour of the part. Temperature dependency was recently included to be able to tackle all kinds of under-the-hood applications.

No66 July - August 2011 / jec composites magazine e-Xstream engineering, the software developer, is already looking into the new challenges of the future. Following injection moulding and short fibres, draping as a processing technology for continuous-fibre composites will enter the stage soon. New material models are also in preparation to describe the dependency of local fatigue properties on the material microstructure. The final goal is to go for a full lifetime prediction of composite parts. n
More information: www.e-Xstream.com Jan.Seyfarth@e-Xstream.com Roger.Assaker@e-Xstream.com

Feature automotive

http://www.e-Xstream.com

JEC COMPOSITES MAGAZINE - Issue #66 - July/August 2011

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #66 - July/August 2011