About Composites Link

Below is a general description on how Composites Link works.

This page discusses:

Composites Development Process

The composites development process is defined schematically below.

General Development



The basic geometry of a part is usually predicated by its function or bounding constraints. The definition of basic geometry is usually completed by a designer using a CAD system.

Next, the zone modeling stage begins where the structure is divided into simplified regions and a basic composites layup is specified. In a zone model, material layers are not considered to extend beyond the region of the zone. Basic fiber directions are specified with respect to a coordinate system, but no detail is given as to the manufacturing process or sometimes even the order of application of material to the surface. For example, in a region the total layup thickness is specified as 5mm, with 50% of fibers in the nominal 0 direction, 10% in the +/- 45 directions, and the remainder in the 90 direction. This stage is usually controlled by an analyst using a FEA system to predict structural response and to size the zones to achieve the required performance. At the end of this stage, the basic composites structure is defined but no manufacturing constraints have been considered.

Next, the plies required to achieve the required total thickness are specified and producibility simulations are performed. This is the most time-consuming part of the process as it may be found that some plies cannot even be manufactured, so they have to be split into smaller sections, or even redesigned completely. But this stage considers manufacturing constraints implicitly, and allows the generation of manufacturing information such as flat pattern shapes. As a result, the ply model has a good chance of being manufacturable on the shop floor without difficulty.

Finally, manufacture is completed based on the ply model. In an aerospace environment, use is typically made of laser projection systems to project the ply boundaries onto the mould to ensure that manufacture matches the design and analysis models.

Composites Models for Analysis and Design

During the composites development process, designers and analysts model the composites structure in different ways, as summarized below.

The analyst defines zones and plies on shell finite elements, or surfaces associated with finite elements. The finite element model is a sufficiently detailed abstraction of the real component for the purposes of defining the response. The model has a very close link to the analysis code. This means that models can be created and analyzed very quickly.

By contrast, designers define zones and plies using geometrical surfaces and boundary curves. This approach is more exact but time-consuming. In general, the design model can be sent directly to manufacture.

Comparing the two different approaches, the analysis approach is far quicker at the expense of detail. For example, defining ply coverage on shell elements takes seconds per ply, whereas defining an exact boundary curve might take orders of magnitude longer. So, where structural performance is critical and time pressure great, such as in top-level motor sport, the ply model is generally first defined by the analyst. By contrast, for many aerospace components developed with less time pressure, the ply model is generally defined by the designer.