Advances in Ultrasonic Inspection Methods for PIM Process Monitoring and Control

Abstract

Previous research efforts concentrating on proof of principle studies indicate that ultrasonic nondestructive evaluation (NDE) can be applied to process monitoring of a Powder Injection Molding (PIM) system [1, 2]. This research resulted in successful detection of several injection molding problems such as flashing, short shot, weld line defects, and jetting. Past techniques, however, require that the sensors be positioned so that the ultrasonic wave, bulk or guided, directly interacts with the defect. Recent work has shown that the data collected from an ultrasonic pulse-echo technique may be used as inputs into a neural network classification and simplex control algorithms for successful PIM process monitoring and control [3, 4]. This technique is sensitive to changes in the boundary conditions at the mold/specimen interface that result from the presence of various specimen defects. The current work investigates the use of ultrasonic data collected from a thru-transmission technique for PIM process monitoring and control. An overview of the data acquisition and analysis system is discussed along with data collection and preprocessing methods. In addition, results obtained from a simplex PIM process variable tuning algorithm experiment are presented. Adjustments in the PIM process variables are made based on information obtained from thru-transmission amplitude profiles.