Erkennung von Schweißfehlern mit Hilfe von zerstörungsfreien Prüfverfahren und Sensorik in Wire Arc Additive Manufacturing (WAM)

Wire Arc Additive Manufacturing (WAM) represents a welding based additive manufacturing process. It belongs to the grouping of Direct Energy Deposition (DED) and can be divided into three processes, Gas Metal Arc Welding (GMAW) based WAM, Gas Tungsten Arc Welding (GTAW) and Plasma Arc Welding (PAW) based WAM. Whereby each of these processes has different strengths and weaknesses in terms of susceptibility to manufacturing defects. These manufacturing defects are caused by the formation of welding flaws, which affect the mechanical properties of the WAM components and may lead to premature failure of the mechanical structure. Here, the four most common weld defects are divided into, pore formation, formation of hot and cold cracks, bond defects and geometry changes. For the further development of the WAM process, it is therefore of utmost importance to detect these defects at an early stage, during the manufacturing process. Thus, guiding steps can be taken to correct these defects or to create abort criteria for the WAM process. In order to realize this, an appropriate sensor system must be determined which can detect welding defects in sufficient size and quantity. The first step in the project relates to the calibration of existing sensors. Since current and voltage sensors are already implemented in the existing WAM system, they have to be adapted to reality in terms of magnitude and time synchronism. The same applies to the wire feed and a newly implemented sensor. This sensor is a microphone that is added to extend the measurement spectrum. In order to determine if weld defects can be provoked and determined during the WAM process, test welds will be performed. During these welds, the weld paths are artificially influenced to create specific weld defects. During the WAM process, the implemented sensors measure current, voltage, wire feed and the microphone signal. The different signals can be used to detect different defects in the test welds. These sensor signals will then be used to determine whether defects form, and which signals are used to detect them.