Monday  |  Carnegie III  |  01:50 PM–02:10 PM

#13694, Detection Analysis for Sub-superficial Defects in Additive Manufactured Metal Samples by Means of Flash Thermography

One of the most used manufacturing processes for producing mechanical components with complex geometries in a short time is Additive Manufacturing (AM). Due to the complexity of the process and the amount of the involved parameters, it is necessary to assure the absence of defects or define their limit size in the final component, to accept or discard it. Furthermore, the surface quality and the accessibility of the surfaces to be checked in additive manufacturing parts is one of the major concerns of the industries. Non-destructive evaluation has been identified as one of the most effective methods for resolving these issues.
This work is focused on the application of active thermography as non-destructive technique to assess the quality of AISI 316L metal samples produced by means of Laser Powder Bed Fusion (L-PBF) process. It is known that one of the most widespread and common defects in AM is porosity, diffuse and localized, due to sudden and unexpected changes in process conditions. From a thermal point of view, this type of defects is very far from the one normally simulated with non-destructive techniques, that are flat bottom holes or delaminations. More in general, sub-superficial defects, like porosities, affect the quality of the signal and the detection limit. For this reason, a complete experimental plan has been carried out to print sub-superficial spheres of different size (depth and diameter) in different specimens, for a total of about 150 defects, including replications at fixed nominal aspect ratio and shape factor. In particular, these spheres have non-melted powder of the same material inside.
A reflection set-up with two flash lamps and a MWIR cooled sensor has been used to perform different tests and to define the limits and the advantages of the proposed technique. To improve the quality of the thermal signal and the signal to noise ratio, post processing algorithms have been used to process the raw thermal data. The study is a preliminary research activity aimed to calibrate the technique for the quantitative estimation of porosity in metal additive manufacturing components.

Ester D'Accardi   Politecnico di Bari
Davide Palumbo   Politecnico di Bari
Vito Errico   Politecnico di Bari
Andrea Fusco   Politecnico di Bari
Andrea Angelastro   Politecnico di Bari
Giuseppe Danilo Addante   Politecnico di Bari
Umberto Galietti   Politecnico di Bari