The authors discuss drawbacks to treating plastics like metals for the purpose of FEA. Thermoplastics exhibit complex behavior when subjected to constant, increasing, or cyclical mechanical loads. The typical approximation of a linear relationship between stress and strain is often invalid because of the extremely non-linear behavior of plastics. Failure to account for this phenomenon can lead to over prediction of the stiffness of a plastic part which might then fail in actual use. The current solution to this problem is significant over design which results in wastage of raw material, and sometimes leads to other unanticipated problems.

The problem of non-linearity is further compounded due to the large displacements that tend to occur in plastic components. The elastic modulii of plastics are routinely as much as two orders of magnitude less than those of metals. Plastics can undergo an order of magnitude more strain than metals before incurring damage Consequently, these materials will tend to undergo much larger rotations and displacements so that the deformations carry further into the non-linear regions of the stress-strain curve. Standard data sheets and most computerized databases commonly provide design engineers with three relevant categories of ‘design properties’: flexural, heat resistance, and impact. While acceptable for comparative purposes, these properties are not useful for predicting the structural performance of plastics components because the data are not
independent of the test method, specimen geometry, and conditions of the test. The authors present methodologies for the generation and use of engineering data.

The book seeks to provide a proper understanding of thermoplastic material behavior and its relationship to measured properties, so accurate predictions of component behavior can be made. Because of the widely differing behavioral characteristics of these materials, no general procedure for the design of plastic parts can be proposed; instead, the book suggests practical approaches to handle the design of thermoplastic components. Treatments of stiffness, failure, impact, time dependent behavior, and fatigue are presented. Numerous examples in the book highlight areas of concern for design analysis of thermoplastics and illustrate the expected level of accuracy from such analyses.

By G. Trantina and R. Nimmer. McGrawHill, Inc., 1993. 384 pp

- Dr. Gerald Trantina is manager and Dr. Ronald Nimmer is a mechanical engineer at the Mechanics of Materials Program in GE’s Engineering Physics Research Center. The book was reviewed by Hubert Lobo, president of DatapointLabs.