Warpage Of Injection Molded Parts Manufactured With Polypropylene And Its Composites

With Growing demands on dimension stability and quality of auto products in modern manufacturing, warpage has become one of the most important criteria in assessing the processing quality of injection molded products. The researchers prefer to optimize the injection molding process parameters to reduce the warpage by experiment design and simulations, treating the warpage as a black box. However, it is difficult to know the exact warpage of part without molding when given a group of process parameters.In this study, the simulation system was developed to provide accurate warpage prediction for Polypropylene(PP) and its composites(thermoplastic olefin and long glass fiber reinforced PP), using the existing simulation software. For the PP and thermoplastic olefin(TPO), two critical material properties were investigated on to build the mathematical models. One was the PVT behavior; the other was the constitutive relationship. The warpage analysis was divided into two steps to simplify the FEA model for warpage. In the first step, the flowing analysis was performed in the commercial software Modex3 D to get the temperature distribution of part at different time. In the second step, the FEA model was built in the software Abaqus for the warpage analysis, embedded with the user subroutines UEXPAN(PVT behaviors) and UMAT(constitutive relationship). The temperature distributions obtained in the first step were imported as the boundary condition in each time step. For the long glass fiber reinforced PP(LGFRP), the fiber orientations of long fibers were measured and simulated, which were used to predict the mechanical property of composite. The local mechanical property and warpage of instrumental panel were simulated and validated also. The conclusions were drawn as follows:1) The PVT and crystallizing properties were studied for PP and TPO in order to accurately describe the PVT behaviors in the conditions close to the actual manufacturing. The standard PVT tests showed that the pressure didn’t change the trend of curves at different pressures. The curve in one pressure could be obtained by moving the curve in another pressure in the T-V plane. DSC tests indicated that cooling rate was critical to the crystallizing curves. The mathematical PVT model was developed by modifying the standard PVT curves according to the crystallizing curves, considering the effects of pressure, cooling rate and crystallinity. The Abaqus user subroutine UEXPAN was coded and validated by the experiments, exhibiting a good agreement.2) The tension-compression tests were conducted to work on the quasi-static mechanical properties of PP and TPO. These results showed that the mechanical properties were sensitive to the strain rate and temperature. The slope of stress-strain curve in the unloading path depended on the loaded strain limit. The Chaboche elastic-viscoplastic model was adopted and modified to describe the mechanical behavior of PP and TPO. The parameter values of modified Chaboche model were determined by fitting the measured stress-strain curves. The comparison of fitted and measured stressstrain curves agreed well with each other.3) The simulation system was built for the warpage analysis in Abaqus, combined with the user subroutines UEXPAN and UMAT for material properties, introducing the effects of cooling rate, pressure and crystallinity. It could predict the deflection of PP plaque well in both the distribution trend and magnitude. For complex plastic part and glove box inner with TPO, it was able to catch the trend of the deflection though the parts. Furthermore, the predicted deflections were in the same order with test results.4) A system was built to simulate the mechanical properties of long fiber reinforced part made by injection molding. The fiber orientations were predicted by the commercial software and validated by the experimental results obtained from the X-Ray CT scan and image analysis. Then, the predicted fiber orientation was used to predict the mechanical properties in material design software Digimat. Finally, the simulated mechanical properties of long fiber reinforced polymer were applied to predict the local Young’s modulus in auto instrumental panel. The results showed that the fibers had typical skin-core distribution in the thickness direction, highly oriented in skin and core. The fibers rotated from the skin to core in the flow plane. They preferred clustering in the core also. The simulated orientation distribution matched the experimental one well. The average fiber orientation tensor was close to the measurement. Furthermore, it was able to predict the rotations of fibers in flow plane from skin to core. The difference of Young’s modulus in plaque, which was induced by the orientation of fibers, was larger than 20%. Though, the difference between the simulation and measurement was less than 8%. The predicted Young’s modulus in the instrumental panel was close to the measurement at most locations, indicating that this system could be used to predict the mechanical properties of long fiber reinforced polymer. The predicted warpage of instrumental panel matched the actual part and showed very slight warpage.