Performance Evaluation of Purely Mechanical Wireless In-Mould Sensor for Injection Moulding Florian M¨ uller, Christian Kukla, Thomas Lucyshyn, Clemens Holzer Abstract—In this paper, the influencing parameters of a novel purely mechanical wireless in-mould injection moulding sensor were investigated. The sensor is capable of detecting the melt front at predefined locations inside the mould. The sensor com- prises a movable pin which acts as the sensor element generating structure-borne sound triggered by the passing melt front. Due to the sensor design, melt pressure is the driving force. For pressure level measurement during pin movement a pressure transducer located at the same position as the movable pin. By deriving a mathematical model for the mechanical movement, dominant process parameters could be investigated towards their impact on the melt front detection characteristic. It was found that the sensor is not affected by the investigated parameters enabling it for reliable melt front detection. In addition, it could be proved that the novel sensor is in comparable range to conventional melt front detection sensors. Index Terms—Injection Moulding, In-Mould Sensor, Structure-Borne Sound, Wireless Sensor I. I NTRODUCTION I NJECTION moulding is a highly dynamic process de- signed to produce high-precision technical parts in mass production scale. To assure reproducibility of the injection- moulded parts, control strategies are necessary to compen- sate changing process conditions. Wang et al. [1] propose a three-level categorized system for injection moulding process control. The first level deals with controlling of machinery related parameters such as barrel temperature, injection rate or clamping force. These parameters can in general be con- trolled independently and accurately using closed loop control strategies [1]–[3]. Industry often selects this approach for part quality control ’hoping that the found process parameter set is good enough’ [4]. The second level comprises process parameters, e.g. melt temperature, melt pressure or melt front advancement. This set of parameters is closely related to the part quality. Finally, level three parameters involve the most complex parameters, the quality related parameters, such as part weight, shrinkage and warpage or optical defects of the part [1]. Of course, it would be great to directly control these parameters to ensure constant quality; however, the lack of appropriate in-mould sensors to detect quality parameters is undeniable. This lack is recognized by researches for some time now and they insist to lay a strong focus on developing in-mould sensor concepts for quality control measurements [2], [4]. Nowadays, two types of in-mould sensors are mainly used to The authors are with the Department of Polymer Engineering and Science at Montanuniversitaet Leoben, A-8700 Leoben, Austria. e-mail: florian.mueller@unileoben.ac.at Manuscript received July 30, 2013; monitor or control the injection moulding process: cavity wall temperature sensors and cavity pressure sensors [5]. Cavity wall temperature sensors are only in some cases used for temperature sensing while more often for melt front detection. Due to the low mass and small sensor head diameter, the sensors have short response times of around 1 10 ms and are suitable for melt front detection [6], [7]. Melt front detection is also possible using cavity pressure sensors which is reported in [8]–[10] for instance. In injection moulding there are several cases in which it is of special interest to know the transient position of the melt front. One of these points is at the end of the filling phase where the machine has to switch-over from a volumetric controlled to a pressure controlled filling condition [11]. This point is at around 98 % volumetric cavity filling [10]. In a machine centric approach this point is whether located using the current ram position or using a timer. However, when process parameters change, e.g. viscosity change due to batch- to-batch variation, the melt front propagation varies to the cycles before. Consequently, in a machine centric approach the switch-over point is not the same to the shots before resulting in varying part quality. In several publications it is emphasized to utilize in-mould sensors for precise switch-over point detection since it is crucial for having constant accurate part quality [6], [12], [13]. One of the major disadvantages of all common available in- mould sensors is the necessity of wiring for energizing as well as data transmission. A mould needs complex structural modification to enable implementation of wire ducts [14]. Moreover, sometimes it is not even possible to find sufficient space for implementation of in-mould sensors since cooling channels, sliders as well as ejector pins have a higher priority in the design process because they are crucial in defining part quality and mould functionality. In addition, wires introduce significant disadvantages to the overall lifetime of a mould due to their sensitive nature, e.g. rupturing. Furthermore, if a sensor fails during processing most parts of the mould have to be disassembled to enable the exchanging of the sensor. In recent years a movement towards wireless in-mould sensors has been initiated. Since the year 2002 a research group in the US develops and investigates a self energized wireless in- mould sensor [15]. The sensor uses the pressure inside the melt to generate electricity for sensing process parameters such as pressure and temperature. The measurement data is transmitted to the outside surface of the mould using ultrasonic structure-borne sound [16]–[18]. With this design 1 For a cavity wall temperature sensor with a head diameter of 1 mm. World Academy of Science, Engineering and Technology International Journal of Mechanical and Mechatronics Engineering Vol:7, No:9, 2013 1778 International Scholarly and Scientific Research & Innovation 7(9) 2013 scholar.waset.org/1307-6892/16594 International Science Index, Mechanical and Mechatronics Engineering Vol:7, No:9, 2013 waset.org/Publication/16594