STRESS-STRAIN RESPONSE OF POLYMERS FOR PREDICTING THE BEHAVIOR OF INTEGRAL FASTENERS

Experimental Approach
The standard engineering stress-strain tension test is performed on uniform section shaped test samples which are placed in an Instron 4204 load frame [20] with an extensometer attached to measure the elongation on the sample. Load readings are taken from an Instron 50 kN
load cell. Tests are performed at three crosshead rates: 1, 10, and 100 mm/min. Next, true local stress and true local strain tension tests are performed in a similar manner, except that the uniform section specimen is replaced with an hourglass sample as shown in figure 1. Here the accepted definition of true strain is observed as defined in equation 1, εtrue=ln( λ) (1) where εtrue is the true strain measure and λ is the stretch. The hourglass sample allows an extensometer to be attached transversely to the waist of the sample to measure the change in width of the sample during the test. In this manner the extensometer can be used to measure the local true strain at the waist of the hourglass sample, and continue to do so after yielding and the formation of the neck which is typical of polymer tension tests. It is assumed that the material is isotropic and the deformation is volume preserving. This allows the calculation of the extensional true strain. The true stress is determined from the load cell readings, the local true strain measurements, and the initial cross-section of the sample. This type of test is performed at the same three crosshead rates as for the standard uniform section samples. Depth 3.18 mm 31.75 mm Radius y x Width 6.73 mm z
The tensile stress-strain response of polymers for large deformations can be obtained in many ways. Two methods are presented and compared to a true constant strain rate controlled tension test. The differences in the results suggest that improved modeling of polymers with respect to strain rate can be achieved using the true constant strain rate test method. These improved models can have a large impact on the predictive capabilities of finite element models. Such improvements are necessary for accurately predicting insertion, retention, and drop-test performance of integral fasteners. A simple finite element material model is implemented incorporating experimental results. focused mainly on addressing the issues of convergence in the models. Several authors [14-18] have proposed rate dependent material models for polymers, however they are aimed at modeling material response at moderate strain rates (<0.01 sec-1 ), or extremely high rates such as those experienced when an anvil crushes a small, short cylindrical test sample [19]. Strain rates encountered in integral fastening tend to be between these two groups. There are a variety of standard tension testing techniques used to determine the stress-strain behavior of a material, and eventually aide in developing a material model. The problem is that the test specimen is often subjected to a variety of strain rates during the test making determination of material properties with respect to strain rate rather difficult. Some of the different types of tension tests conducted are as follows: •Standard displacement controlled engineering stress-strain tension tests using standard uniform section test samples. •Displacement controlled tension tests that measure local true strain and local true stress using hourglass shaped test samples. •True constant strain rate controlled tests that measure local true strain and local true stress using hourglass shaped test samples. In this paper the results are compared for each of the three test methods. Then, the last method of controlling the tension test in a true constant strain rate manner, which is considered to be more useful in determining rate dependent properties for polymers, is implemented in a finite element analysis (FEA) to verify that the experimental data can indeed be used in a basic “rate dependent” material model. Differences in the resulting material response are also compared and show significant difference in the results obtained based on the test mete of integral fasteners has many benefits for both design for assembly, product safety, and design for disassembly [1-6]. However, not much is known about the performance characteristics of plastic integral fasteners. The proper design and application of integral fasteners is often a time consuming, risky task. In particular, one of the most common areas of failure for integral fastening occurs when they are subjected to high rates such as in a product drop test. Typically, drop tests occur late in the development process using production parts. A product failure at this late stage can be extremely costly. Ideally, it would be beneficial for a development team to be able to run a simulation of the drop test early in the development process before money has been committed to tooling. However, adequate material models do not currently exist that adequately capture high strain rate properties of plastics to allow for a finite element simulation of a drop test. This paper details an approach used to more accurately determine the rate dependent behavior and material properties of a polymer with the goal of eventually using this type of test to improve existing rate dependent models. Work has been performed in the are of modeling integral fasteners and their performance [7-13]. However all of these works have used basic rate independent material models while performing a quasi-static analysis. Little effort has been focused on improving the material model to include rate effects. The efforts have been