JEC COMPOSITES MAGAZINE - Issue #59 - August/September 2010 - 29

Design
- Lower panel face strains are measured by strain gauges (typically three at 0, 45, 90°). - Ball shape and impact duration are measured using a high speed Photron Fast-CamX digital camera (1,000 frames/ sec.), and a second digital camera records rebound height. Impacters with weights up to 45 kg are now dropped from heights to 7 meters, providing impact energies over 3 kJoules and pressures up to 100 tonnes/m². All these data are recorded and stored for subsequent analysis. ball. The sandwich skins were represented by quadratic multilayered shell elements and an orthotropic linear elastic material and the core was modelled by quadratic 3-D elements and an orthotropic (honeycomb) or isotropic (foam) material (Figure 5a). Provided the medicine ball is correctly modelled, a reasonable correlation between test results and simulations can be obtained (Figure 5b). More complex material models (elasto-plastic or viscoelastic) can be used to simulate the response of foam cores.

Test programmes
Several test programmes were implemented, the details of which can be found in [2,3]. These revealed a range of damage mechanisms (Figure 4) that depend on core type, facing thickness and panel geometry.

c
Fig. 5: a) FE model, b) example of results of results for central displacement during impact

d
Fig. 4: Different damage mechanisms, medicine-ball impact on sandwich panels. a) Core crushing, Nomex honeycomb b) Facing compression failure, ductile foam core c) Core shear, aluminium honeycomb d) Core shear, brittle foam core

Future work
Modelling work is continuing in order to integrate material response in the structural models used at HDS to dimension racing yachts. Other aspects of wave impact are also being studied, including the influence of stiffeners on structural response and the effects of repeated impact on damage development. Contact: peter.davies@ifremer.fr, hdv@hds-design.com
References [1]. Choqueuse D., Baizeau R., Davies P Experimental studies of .: impact on marine composites; Proceedings ICCM12, (1999), Paris. [2]. Baral N., Cartié D.D.R., Partridge I.K., Baley C., Davies P .: Improved impact performance of marine sandwich panels using through-thickness reinforcement: Experimental results; Composites: Part B 41 (2010) 117–123. [3]. Davies P Choqueuse D., Bigourdan B., Baral N., Cartié D.D.R., ., Partridge I.K., Baley C.: Pinned foam core sandwich for improved damage tolerance of racing multi-hull yachts; Proc. International Conference on Composite Materials (ICCM-17), Edinburgh, July 2010.

These mechanisms are more or less favoured by the panel geometry. For example, a thicker core is more susceptible to shear failure. Given these differences, it is essential to develop modelling tools that will allow the test results to be analysed correctly and used for design.

Numerical modelling
Numerical modelling was therefore performed. Figure 5 shows an example of a finite element model of an impact test. One quarter of the medicine ball and panel are represented. This is a dynamic model using ADINA. Quadratic shell elements and an isotropic linear elastic material were used for the medicine ball envelope. Quadratic 3-D elements and a non-linear Mohr-Coulomb material were used for the sand in the

No59 August - September 2010 / jec composites magazine

JEC COMPOSITES MAGAZINE - Issue #59 - August/September 2010

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #59 - August/September 2010