Design and manufacturing of thermoplastic composite anisogrid structures

The replacement of fibre reinforced thermoset components with thermoplasticbased composites has been growing more and more in recent years. This is due to the fact that thermoplastics have an indefinite shelf life, low moisture absorption, excellent thermal stability, high toughness and damage tolerance, short and simple processing cycles and potential for significant reductions in manufacturing costs. In addition, they have the ability to be remelted and reprocessed and also damaged aircraft structures can be repaired by the application of heat and pressure. Taking into account these market trends and the above-mentioned technical issues, the PhD thesis aims at designing and manufacturing two- and three-dimensional cylindrical anisogrid lattice structures by using thermoplastic prepreg. Anisogrid stands for “anisotropic grid” and refers to the production of structural grids in composite material. Two-dimensional anisogrid lattice structures have been prototyped with 3 different geometries and 5 different numbers of layers from 4 to 8 by using a metallic pattern and E-glass/polypropylene prepregs. The “filament gun deposition” manufacturing process consists of a hot-melt gun loaded with narrow thermoplastic prepreg tapes. A finite element model has been used to predict mechanical stiffness of these structures by using material properties coming from the sample characterization. A good agreement has been found between experimental and numerical data. Three-dimensional lattice structures, cylindrical and conical geometries, have been manufactured by using an enhanced process called “heat shrink compaction”. It consists of a combined effect of pressure and temperature by using a tubular thermal shrinker and an exposition into a muffle. Also here, three-dimensional anisogrid lattice structures have been prototyped with different geometries and different number of layers. Before producing anisogrid structures, circular rings have been produced by the same process in order to verify the right compaction level and the output density. A finite element model reproducing the ring have been used for the calibration and validation process. Another numerical model reproducing the anisogrid structure has been used to predict mechanical stiffness of these structures by using material properties coming from calibration process. A good agreement has been found between experimental and numerical data with an average difference about 9% for the complex structure of anisogrid lattice.