Predicting material properties for embedded structures created with polymer material extrusion additive manufacturing

• Published in: Additive manufacturing Vol. 34; p. 101247
• Main Authors: Sinha, Swapnil, Meisel, Nicholas A.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.08.2020
• Subjects: Cavity design; Design for additive manufacturing; Embedding with material extrusion; Cavity design; Design for additive manufacturing; Embedding with material extrusion
• ISSN: 2214-8604; 2214-7810
• DOI: 10.1016/j.addma.2020.101247

Embedding with additive manufacturing (AM) is a process of incorporating functional components, such as sensors and actuators, in the printed structure by inserting them into a specially designed cavity. The print process has to be interrupted after the cavity is printed to insert the component. This allows for multifunctional structures to be created directly from the build plate. However, previous research has shown that this process interruption causes failure at the paused layer due to the cooling between the layers. The presence of the designed cavity further impacts the strength of the part due to a reduction in the effective cross-section in contact between the paused and the resumed layers. This research presents a methodology to predict the weld strength between the layers of an embedded material extrusion structure by obtaining the thermal history at the layer interface as a result of process interruption. An infrared camera and an embedded thermocouple are used to obtain the thermal history of the depositing fresh layer and of the layer interface, respectively. The impact of toolpath design on the thermal history of the layer interface is considered by dividing the cross-section area into zones with similar thermal history. Polymer weld theory is utilized to predict the strength at these different zones, where material properties are obtained through rheology measurements. These strength values for the zones are then used to predict the load at failure for different specimens by treating them as composites. Findings confirm that this approach can be used to more accurately predict tensile loads at failure for embedded structures, with errors ranging from 1% to 20 % depending on the toolpath geometry.