Midplane mesh

This is the direct approach in which a solid model is read into a program and an automatic midplane mesh is generated (Fig. 7.61). Details of such a system were given in Ref 42 and will not be discussed here. In practice, the fully automatic generation of a midplane model is difficult. In many cases there is some need to clean up the resulting model before analysis. Nevertheless, midplane generation can save an enormous amount of time for many types of part. 

Fig. 8.1 An example of midplane mesh (Courtesy of Dr. Peter Kennedy)...

The midplane FE approach is by far the most efficient numerical method for the simulation of injection molding of thin-walled parts. However, the preparation of a midplane mesh can take a considerable amount of time. 

The limitations of the automatic midplane generation stimulated the innovation of the so-called dual domain approach (Yu and Thomas 2000 Yu et al. 2004). This approach uses the external mesh on a 3-D geometry. It still solves the Hele-Shaw equation but eliminates the need for a midplane mesh. 

Figure 5.6 Sample three-dimensional simulations of the IP process. (Top) Finite element mesh. Total length 30 cm (0 < X < 30), total height 1 cm (0 < Z < 1), total width 3 cm (0 < Y < 3). Fluid is injected from both sides through the thickness (i.e., in the Z-direction through a 1 cm x 1 cm square). (Bottom) Flow front progression at the midplane (i.e., Z — 0.5), Ka = Kzz = 2Kyy K, - = 0.0...

Figure 8.7 illustrates the basic idea. Figure 8.7a shows a rectangular plate injected at its center. Figure 8.7b shows the flow in a cross-cross section. Figure 8.7c is the midplane representation of the flow, and Fig. 8.7d is the dual domain representation of the flow, where two analyses are performed simultaneously on the top and bottom surface meshes. 

---