October AMP_Digital

A D V A N C E D M A T E R I A L S & P R O C E S S E S | O C T O B E R 2 0 1 9 2 3 lead to a reduction in scrap production. Because each subsequent volume el- ement is initially accessible as a sur- face during the production of additively manufactured components, it is also expected that defects can be detected that cannot be found using convention- al NDT methods due to, for example, complex geometries of the finished component. To achieve these improve- ments and efficiencies, the extraction, processing, and analysis of very large amounts of data generated by these in situ process-monitoring techniques (big data) poses a challenge. To gener- ate added value from the data volume, reduction, visualization, and correla- tion of the generated data sets from in- dividual processes must be performed (data fusion). In this domain, BAM co- operates with partners such as the Zuse Institute Berlin (ZIB). Using modern data analysis meth- ods, BAM is developing a technique for process validation. BAM is also evalu- ating the influence of different process variables on manufactured-component quality. a focus on safety (both for applications and the environment) and AM technolo- gy innovation for a wider range of mate- rials and manufacturing locations. IN SITU PROCESS MONITORING Additively manufactured compo- nent quality is significantly influenced by process parameters used during pro- duction. Process sensors and measur- ing systems for controlling the energy source, the melt pool, and component geometry are already commercially available in metal-based AM. However, it is not yet possible to detect defects and inhomogeneities directly or indi- rectly during the building process. The BAM project Process Monitoring in Ad- ditive Manufacturing (ProMoAM) with- in the materials subject area aims to develop spectroscopic and NDT meth- ods for in situ evaluation of the quality of additively manufactured metal com- ponents in laser and arc-based AM pro- cesses. In situ techniques in addition to passive and active methods of ther- mography, include optical tomography (Fig. 1), optical emission spectroscopy, eddy current testing, particle emis- sion spectroscopy, laminography (ra- diography), x-ray backscattering, and photoacoustic methods. Previous in- vestigations show that, for example, thermography can be used to record temperature distribution (Fig. 2). How- ever, predictions of the occurrence of defects are only possible by evaluating heating and cooling rates and spatial temperature gradients. These processes are being devel- oped on BAM AM systems, i.e., laser powder bed fusion, laser metal deposi- tion, and wire and arc AM. Key charac- teristics of measured data are extracted to correlate defects and inhomogene- ities with production parameters. Re- sults of these evaluations are compared with results from benchmark reference procedures such as x-ray computer to- mography and ultrasonic immersion technology. This process monitoring should result in a significant reduction in expensive, time-consuming nonde- structive tests performed after produc- ing a component and, at the same time, NONDESTRUCTIVE MATERIALS CHARACTERIZATION Nondestructive materials charac- terization is a central component for understanding additively manufac- tured materials. These methods serve as a reference to underpin develop- ment of in situ process monitoring, and they are used to characterize and vali- date additively manufacturedmaterials from powder to component failure. In this context, BAM develops and evaluates nondestructive testing meth- ods to characterize internal structures and defects in additively manufactured materials and structures in the microm- eter range and below. The experience of BAM in the development of micro-com- puted tomography (µ-CT) for quantita- tive 3D materials characterization has been extended by means of synchro- tron µ-CT at the BAMLine of the Ber- lin Electron Synchrotron BESSY. The technique is complemented by syn- chrotron x-ray refraction (SynRef-CT) developed at BAM, which is suitable to investigate smaller defects due to the Fig. 2 — Thermogram recorded using a mid-wave infrared camera during AMmanufacturing of a 316L austenitic stainless steel cube in a laser powder bed fusion machine. The area of liquid material is shown in a separate grayscale.