| Injection molding, one of the primary polymer processes, is widely used in many industries because it is suitable for mass production. It is, however, not easy to determine the optimal injection conditions due to the complex multivariate and dynamic nature of the injection molding process. Moreover, as the use of polymer increases, additional demands, such as intricate geometry, high tolerance, quality finish, etc., have been placed on the injection molding process. To improve the molding quality and lower the cost, manufacturers and researchers have been working on improving understanding and control of the injection molding process. Transducers are a basic technology that assists in interpreting the processing, verifying simulated results, and in the precise control of injection molding process.; This dissertation investigates some theoretical and experimental aspects of ultrasonic monitoring of the injection molding process with practical applications for process monitoring, model validation, and production control. This work involves development of an ultrasonic apparatus and a mold instrumented with ultrasonic and cavity pressure sensors. A series of experiments are performed under ultrasonic monitoring. Both longitudinal and transverse sensors are used to detect the ultrasound velocity and attenuation in the polymer during the injection molding of polystyrene. The measured signals are examined as the polymer state in the cavity varies.; To properly interpret the behavior of the experimental signals a model is then developed. The model consists of three modules. First, a viscoelastic constitutive equation is developed to characterize the temperature and pressure dependent properties of polymer materials. Second, the wave velocity and attenuation as a function of viscoelastic properties of polymer are derived by using the developed constitutive relationship. Third, a process model for injection molding is developed to calculate polymer temperature and flow profiles as a function of injection time. The calculations are carried out using a one dimensional finite difference method. Based upon this model the amplitude of ultrasonic echo from the far-end polymer/mold interface is predicted as a function of injection time. The predicted results illustrate how temperature and pressure affect the detected amplitude, and are then used to examine which factors contribute to the observed amplitude profiles, making it possible to see inside the mold.; The far-end echo amplitude of the compression wave and shear wave signals (velocity in polymer and amplitude of the far-end echo) are analyzed for information regarding the state of the polymer in the cavity. The potential application of this information for process monitoring is then discussed. With these sensors it will be possible to develop control strategies that allow the injection molding to operate with less supervision and to yield high quality, consistent, and economical products. |
