Additive Manufacturing: Equipment, Instrumentation and Measurement: Session I
Program Organizers: Ulf Ackelid, Freemelt AB; Ola Harrysson, North Carolina State University
Monday 8:00 AM
November 2, 2020
Room: Virtual Meeting Room 5
Location: MS&T Virtual
Session Chair: Ulf Ackelid, Freemelt AB
8:00 AM
Adaptive Multi-Beam Laser Additive Manufacturing (AMB-LAM) Technology: Instrumentation and Processes Development and Demonstration: Mikhail Vorontsov1; Nathan Farwell2; Michael Massey2; Geunsik Lim2; 1University of Dayton; 2II-VI Corporation
We describe an innovative powder bed selective laser melting (SLM) system based on a fiber array laser energy source composed of seven laser beams whose parameters (power, focal spot centroid coordinates, scanning speed, amplitude and direction) can be simultaneously and individually controlled to adaptively shape spatio-temporal distribution of the laser power on the powder material providing capabilities for thermal gradients management during the SLM processes. The developed adaptive 7-beam powder bed SLM system was utilized for printing basic coupons using different beam shapes from the following materials: Inconel 718, 7075Al, Ti-6Al-4V, 316L stainless steel and Ti-48Al-2Cr-2Nb. The coupon characterization demonstrated potentials for improvements in surface finish, density, tensile and yield strength, ductility and micro-cracks mitigation. Microstructures of the test coupons were evaluated with EBSD imaging techniques. The obtained EBSD images clearly demonstrated ability for local modification (transitioning from columnar to equiaxed) of material micro-structures during the SLM process.
8:20 AM Invited
Dynamics of Laser-powder-metal Interactions in L-PBF Captured by High Speed Imaging: Manyalibo Matthews1; Nicholas Calta1; Philip Depond1; Gabe Guss1; Saad Khairallah1; Jonathan Lee1; Aiden Martin1; Rongpei Shi1; Maria Strantza1; Alexander Rubenchik1; 1Lawrence Livermore National Laboratory
In laser powder bed fusion (LPBF) additive manufacturing, complex hydrodynamics driven by vapor recoil and Marangoni convection lead to liquid metal interfaces that are steeply curved thereby affecting laser-material coupling efficiency. Changes in dynamic absorptivity due to melt pool motion can lead to fluctuations in local microstructure and residual stress. To clarify the complex physics involved, a LPBF test bed equipped with high speed optical imaging and microcalorimetry is used to study melt pool dynamics as a function of laser parameters for several commercial-grade alloy systems. To capture keyhole dynamics, a second testbed is used at synchrotron beamlines for high speed x-ray radiography. Hydrodynamic finite element models simulate the melt pool morphology and dynamics, providing insight into energy coupling, keyholing and spatter generation mechanisms. The combination of unique diagnostics and high-fidelity modeling creates an exceptional capability for L-PBF physics validation and process optimization. Prepared by LLNL under Contract DE-AC52-07NA27344.
9:00 AM
Investigations on Optical Emissions and Their Relation to Processing Parameters and Processing Regimes in The Laser Powder Bed Fusion Process: Christopher Stutzman1; Abdalla Nassar1; 1Penn State University
Additive manufacturing (AM) is a rapidly growing field where complex components are produced without complex tooling. While the desire to move into new defect critical industries exists, it is important to understand that AM processes can, and often do, create flaws within the component. Therefore, it is important to monitor the process to determine when flaws are formed. In this work, we investigate the sensing of plume fluctuations during the deposition of nickel alloy 625 via a custom-built, off-axis multi-spectral sensor during a laser powder bed fusion process. We show that spectral emission lines from the vapor plume, relate to processing parameters used during deposition and enable sensor-based process map development. We conclude by proposing a novel calibration technique which will permit the correction of intensity variations thus enabling analysis on the fly. Expected applications of the work include real-time sensing and control of PBFAM build conditions.
9:20 AM
Machine Learning Enabled Acoustic Monitoring for Flaw Type Detection in Laser Powder Bed Additive Manufacturing: Brandon Abranovic1; Wentai Zhang1; Haiguang Liao1; Jack Beuth1; Levent Kara1; Qingyi Dong1; 1Carnegie Mellon University
This work focuses on the analysis of acoustic data as a means to monitor laser powder bed additive manufacturing processes for key outcomes. This is of interest as it enables robust quality assurance, control and optimization of component properties, and improvement of process stability while reducing operator burden. Process mapping for Ti-6Al-4V was employed in determining parameter sets that would reliably induce keyholing, lack-of-fusion, bead-up, as well as a fully dense component. Using acoustic data collected during builds using these parameter sets, bag of words (BOW), support vector machines (SVM) and convolutional neural networks were evaluated for their performance in effectively classifying porosity flaws. Preliminary results have shown that these methods are able to reliably distinguish between the classes of interest. In future work, the application of recurrent neural networks (RNN) such as long-short term memory (LSTM) networks will be assessed for their viability against CNNs for baseline testing.
9:40 AM
Mechanical In-situ µCT Testing of Lattice Structures Manufactured by Selective Laser Melting: Pascal Pinter1; Stefan Dietrich2; Lukas Englert2; 1Volume Graphics GmbH; 2KIT / IAM-WK
In the present contribution, a lattice structure was manufactured additively in the shape of a dog-bone sample and tested using a micro computed tomography (CT) in-situ stage. Tensile tests were conducted in order to evaluate the local and global stiffness. Therefore, digital volume correlation (DVC) was utilized to match CT images of different load steps by calculating a transformation from one load step to the other. This geometrical transformation was then used to evaluate the predominant strain tensor on different scales of the structure. Our study shows that the strain tensor can be derived from DVC tests and used to calculate the local stiffness of the structure. Recommendations for parameter selection will be given to receive valuable results and reduce noise. Moreover, it is shown how DVC can be used to detect damage, such as cracks and other singularities that appear during testing.
10:00 AM
Using In-situ Process Monitoring Data to Identify Defective Layers in TI-6AL-4V Additively Manufactured Porous Biomaterials: Darragh Egan1; Denis Dowling1; 1UCD
The detection of process anomalies is critically important during the additive manufacturing of metallic components. In this study Ti-6Al-4V based porous structures were produced on a Renishaw 500M. A co-axial process monitoring system was utilised to generate laser and meltpool related data. When an optimised Generalized Extreme Studentized Deviate (GESD) test was applied to data generated during the processing of structures containing a known number of anomalous layers, the test precisely identified each of the anomalous layers. When the GESD test was run on the meltpool data generated during the processing of an unintentionally defective sample, 30 layers were identified as defective. Upon visual examination, the identified layers corresponded to the physical location of the defect within the sample. This study thus demonstrated the ability of the GESD test to detect anomalous layers created during the L-PBF process.
10:20 AM
Polyspectral Analysis for In-situ Prediction of Deviations in Laser Powder Bed Fusion Additive Manufacturing: Arthur French1; John Sions2; Yuri Plotnikov2; Kyle Snyder2; Kaushik Joshi2; Afroditi Filippas1; 1Virginia Commonwealth University; 2Commonwealth Center for Advanced Manufacturing
Due to the high cost and long build times of additive metal manufacturing in laser powder bed fusion (LPBF), it is essential to advance our ability to identify micro defects in real time through advanced data analytics on a variety of sensor modalities. Multi-modal data gathered through acoustic emissions (AE), IR, build plate position, HR camera and photodiode is parsed to distinguish signals generated during specific LPBF cycles. The data is then analyzed separately for each stage in the process, with weight being placed on the melting cycle, which is when porosity would be more likely to form. Test parts were designed and built in Inconel 718 with varied laser energy and power settings to explore a normal run vs a run conducive to porosity formation. The parts were then characterized in terms of density and nature of porosity. Results from our data analysis and build characterization will be presented.