Colloquia

Summary

At the University of Twente, the Friction Screw Extrusion Additive Manufacturing (FSEAM) approach has been developed as a solid-state additive manufacturing alternative to overcome solidification and other problems of aluminium and magnesium alloys in fusion-based additive manufacturing. The approach is based on a rotating threaded tool placed in a stationary housing. The screw softens, transports and bonds the feedstock material to the build surface.

Recent research has successfully produced wall-like structures, but the current system is prone to unsteady, varying material flow rates at certain process conditions affecting the product’s geometrical stability. Furthermore, a research gap exists on the role of the tool gap, the distance between the tool and the housing, and the tool rotation rate on the system’s temperatures, tool torque, feed force and normal force applied to the build surface. This knowledge is crucial for further optimization of FSEAM to a full-grown production technique that is promising to be applied in aerospace products and beyond.

In this master thesis, an extensive systematic experimental study was performed with a modified, Friction Screw Extrusion (FSE) setup that is based on the same principles as FSEAM but which facilitates a more straightforward analysis of the role of the process parameters. The FSE experiments were performed with various tool rotation rates, tool gaps and extrusion speeds, while recording process temperatures and forces. The extrusion speed was determined afterwards from video analysis of the motion of the extrudate.  

The FSE experiments were conducted successfully. Detailed analysis of the video recordings, based on the Farneback method, successfully retrieved material extrusion speed data. In general, the extrusion speed was constant except for a small region in the process window that showed speed variations compliant with FSEAM observations.

Correlation analysis, based on the Pearson correlation coefficient, revealed clear linear correlation between variations in extrusion speed and normal force in that region of the process window. In fact, the developed correlation method was found to be a clear indicator in predicting the occurrence of extrudate speed variation. Severity coefficients were proposed as a first step to indicate the severeness of this variation which could possibly be used at a later stage for improved process control.