Fiber optic flow monitoring system and methodology for sensor placement to identify flow disturbances in liquid composite molding

In the liquid composite molding (LCM) process, preforms are placed in a closed mold and resin is injected to fill all the empty spaces between the fibers to make a composite. The filling behavior inside the closed mold may not be repeatable due to disturbances introduced during cutting or placing of the preform, which could lead to flow patterns that result in large regions of unsaturated fibers. Hence, there is a need to monitor the filling process so that the quality of the final product may be controlled. Since embedded sensors can cause other complications, having as few sensing elements as possible in the mold would be desirable.; The goal of this work is to develop a sensor placement methodology for robust and reliable flow monitoring in liquid composite molding processes by implementing only one or two sensing elements within the mold. By establishing an appropriate methodology, the best locations to position these sensors can be systematically determined, allowing timely identification of different mold filling patterns. Initially, the feasibility of two types of fiber optic sensors, the evanescent wave fluorescence sensor (EWFS) system and the long period grating (LPG) sensor, were evaluated for this application. Their responses to the presence of resin were characterized and then tested to see how well they would function under realistic manufacturing environments.; Subsequently, two versions of the sensor placement methodology were developed to accommodate both the linear continuous response of the EWF system and the discrete on/off response of the LPGs. Both these forms of the methodology would be applicable to any other sensor technology with either of these responses to resin arrival. Development of the methodology was achieved by simulating the sensor responses based on the resin arrival times at specific locations in the mold for a variety of ideal and nonideal mold filling scenarios due to induced disturbances, which are generated through filling simulations. Race tracking is the featured anomaly employed as it is one of the most prevalent issues associated with the filling stage of LCM. The responses were assessed based on how well they could distinguish the different flow patterns from each other and how quickly that was accomplished. As a result of quantifying the effectiveness of the responses in this manner, the performances of different sensor configurations can be easily compared. More emphasis was placed on the procedure developed for the LPG sensors since it was concluded in the preliminary investigations that they stood a better chance of being successfully implemented as a flow monitoring device for LCM.; Through the course of the investigation, it was demonstrated that sufficient information regarding the progress of the filling process could be gathered from a minimal number of sensing elements within the mold provided they were strategically placed. The methodology was further tested to ensure that it was flexible enough to accommodate a variety of potential scenarios while retaining as much physical relevance to the process itself as possible. In addition to comparing results from this evaluation scheme with those obtained from other existing sensor placement schemes, experiments were also carried out to validate the results from these studies.