TY - CONF A1 - Robens-Radermacher, Annika T1 - Efficient cooling time optimization in Wire Arc Additive Manufacturing using a multi-layer reduced order model N2 - Additive manufacturing (AM) has transformed the industry by enabling the production of complex geometries and parts with customized properties. Among various AM techniques, wire arc additive manufacturing (WAAM) stands out due to its high deposition rate and low equipment cost. However, WAAM’s complex thermal history poses challenges for real-time simulation, essential for online process control and optimization. Consequently, experimental optimization remains the state-of-the-art approach. A critical parameter to optimize is the cooling phase duration, which prevents structural overheating, controls the molten pool size, and influences the mechanical properties of the final product. For efficient cooling time optimization, a fast-to-evaluate model of the temperature field during multi-layer deposition is necessary. This study proposes a reduced order model (ROM) using the proper generalized decomposition (PGD) method as a powerful tool to minimize computational effort. Given the moving heat source in WAAM processes, a mapping approach is employed to achieve a fully separated representation of the temperature field. Building on the authors’ previous one-layer approach, this contribution extends the model to multiple layers through enhanced mapping and compression techniques. The compression reduces the total number of PGD modes as the number of layers increases. The extended mapping allows computations with a fixed mesh over the simulation time, in contrast to standard methods such as the element birth technique. For cooling time optimization, the cooling duration of each layer is incorporated as PGD variables, enabling time-efficient computation of the temperature field for varying cooling times. The developed ROM is applied to optimize the cooling time of a multiple layer example. Therefore a 5-10 layer wall structure is investigated using the austenitic stainless steel 1.4404 (AISI 316 L). The resulting cooling times and the efficiency of the approach are discussed. T2 - 12th European solid mechanics conference (ESMC) CY - Lyon, France DA - 07.07.2025 KW - Model order reduction KW - Proper generalized decomposition KW - Welding KW - Additive manufacturing KW - Optimzation PY - 2025 UR - https://esmc2025.sciencesconf.org/ AN - OPUS4-63855 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Santi, Alberto A1 - Schröder, Jakob A1 - Serrano-Munoz, Itziar A1 - Bayat, Mohamad A1 - Hattel, Jesper Henri T1 - Exploring the Flash Heating method in additive manufacturing for residual stress prediction: A comparative study with diffraction results from X-ray and neutron techniques N2 - Residual stress (RS) control is crucial for ensuring the performance and reliability of components produced through laser-based powder bed fusion (PBF-LB) additive manufacturing (AM). This study evaluates the Flash Heating (FH) method as an efficient approach for RS prediction, comparing its outcomes with multiple experimental techniques, including X-ray diffraction, neutron diffraction, and layer removal methods. These experimental assessments are conducted in different regions of the component, both before and after detachment from the baseplate. The study validates the FH method and analyzes key numerical parameters, such as meta-layer height, contact time, and time-stepping strategies. Results indicate that FH effectively predicts bulk RS distributions but shows discrepancies in surface stress estimations, likely due to unaccounted factors like surface roughness. Additionally, implementing experimentally derived material properties from as-built AM samples significantly enhances model accuracy compared to conventional material datasets. These findings underscore the potential of FH for efficient RS prediction in PBF-LB while identifying areas for further improvement. Refinements should focus on incorporating anisotropic, temperature-dependent material behavior derived from as-built AM samples and surface roughness effects. This work advances the understanding of key factors necessary for accurate and computationally efficient RS prediction, supporting the optimization of AM processes. KW - Finete element method KW - Inconel 718 KW - Metal additive manufacturing KW - Residual stress KW - Thermomechanics PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-639533 DO - https://doi.org/10.1080/01495739.2025.2541862 SN - 0149-5739 SP - 1 EP - 24 PB - Taylor & Francis AN - OPUS4-63953 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Schmidt, Jonathan A1 - Merz, Benjamin A1 - Poka, Konstantin A1 - Mohr, Gunther A1 - Hilgenberg, Kai T1 - Surface structure analysis using visual high-resolution in situ process monitoring in laser powder bed fusion N2 - Parameter studies are a common step in selecting process parameters for laser powder bed fusion of metals (PBF-LB/M). Density cubes are commonly used for this purpose. Density cubes manufactured with varied process parameters can exhibit distinguishable surface structures visible to the human eye. The layer-wise process enables such surface structures to be detected during manufacturing. However, industrial visual in situ monitoring systems for PBF-LB/M currently have limited resolution and are incapable of reliably capturing small differences in the surface structures. In this work, a 65 MPixel high-resolution monochrome camera was integrated into an industrial PBF-LB/M machine together with a high-intensity LED (light-emitting diode) bar. Post-exposure images were taken to analyse differences in light reflection of fused areas. It is revealed that the grey-level co-occurrence matrix can be used to quantify the visual surface structure of nickel-based superalloy Inconel®939 density cubes per layer. The properties of the grey-level co-occurrence matrix correlate to the energy input and the resulting porosity of density cubes. Low-energy samples containing lack of fusion flaws show an increased contrast in the grey-level co-occurrence matrix compared to specimens with optimal energy input. The potential of high-resolution images for quality assurance via in situ process monitoring in PBF-LB/M is further discussed. KW - Additive manufacturing KW - Powder bed fusion KW - In situ monitoring KW - Image processing KW - High resolution camera PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-626685 DO - https://doi.org/10.1007/s40194-025-01955-1 SN - 1878-6669 SP - 1 EP - 15 PB - Springer Science and Business Media LLC AN - OPUS4-62668 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Engelking, Lorenz A1 - Scharf-Wildenhain, R. A1 - Schröpfer, Dirk A1 - Hälsig, A. A1 - Kannengießer, Thomas A1 - Hensel, J. T1 - Influence of heat input on properties and residual stresses in hybrid addi-tive manufacturing of high strength steels using MSG processes N2 - The application of steels with a higher yield strength allows reductions in wall thickness, component weight and production costs. Hybrid additive manufacturing based on Gas Metal Arc Welding (GMAW) processes (DED-Arc) can be used to realise highly effi-cient component modifications and repairs on semi-finished products and additively manufactured structures. There are still a number of key issues preventing widespread implementation, particularly for SMEs. In addition to the manufacturing design, detailed information about assembly strategy and geometric adaptation of the component for modifications or repairs are missing. These include the welding-related stresses associ-ated with the microstructural influences caused by the additive manufacturing steps, particularly in the transition area of the substrate and filler material interface. The pre-sent research focuses the effect of welding heat control during DED-Arc process on the residual stresses, especially in the transition area. Defined specimens were welded fully automatically with a high-strength solid wire (yield strength > 790 MPa) especially adapted for DED-Arc on S690QL substrate. The working temperature and heat input were systematically varied for a statistical effect analysis on the residual stress state of the hybrid manufactured components. Regarding heat control, t8/5 cooling times within the recommended processing range (approx. 5 s to 20 s) were complied. The investiga-tion revealed a significant influence of the working temperature Ti on the compressive residual stresses in the transition area and the tensile residual stresses at the base of the substrate. High working temperatures result in lower compressive residual stresses, heat input E does not significantly affect the tensile stresses. T2 - 6. Symposium Materialtechnik CY - Clausthal-Zellerfeld, Germany DA - 20.02.2025 KW - DED-Arc KW - Residual stress KW - Heat control PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-632188 DO - https://doi.org/10.21268/20250506-3 SP - 110 EP - 122 AN - OPUS4-63218 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Breese, Philipp Peter A1 - Altenburg, Simon T1 - Absolute temperature determination in laser powder bed fusion (PBF-LB/M) via hyperspectral thermography N2 - Temperature is a key characteristic in laser powder bed fusion of metals (PBF-LB/M). As a quantitative physical property, the temperature can determine the actual process quality independently from the nominal process parameters. Thus, establishing a process evaluation on temperatures rather than the comparison of process conditions is expected to be more effective. However, quantitative in situ temperature measurements with classical thermographic methods are virtually impossible. The reason is that the required emissivity value changes drastically throughout the process. Additionally, large temperature ranges along with the highly dynamic nature of the PBF-LB/M process make temperature measurements difficult. Based on this challenge, this work presents a method for hyperspectral temperature determination. The spectral exitance (in W/m2⋅nm) was measured in situ at many adjacent wavelengths in the short-wave infrared (SWIR). This enabled a local temperature determination via Planck’s law in combination with a spectral emissivity function. The temperature field of the melt pool crossing the 1D measurement line was reconstructed from the information, gathered at nearly 20 kHz sampling rate. The reconstructed melt pool had a spatial resolution of 17 µm by 40 µm, and temperatures between 2700 and 1300 K were observable. Comparison of the 316L stainless steel solidification temperature and the observed solidification plateau in the gathered thermal data revealed a relative error of less than 6% in the absolute temperature measurement. These initial results of hyperspectral temperature determination in PBF-LB/M show the potential in the method. It allows for physically founded process evaluation, and the prospects for tuning and validation of numerical simulations are highly promising. KW - Additive Manufacturing KW - Infrared Thermography KW - In-situ Monitoring KW - Quantitative Temperature Measurement KW - Emissivity PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-631977 DO - https://doi.org/10.1007/s40964-025-01148-8 SN - 2363-9512 SP - 1 EP - 10 PB - Springer Science and Business Media LLC AN - OPUS4-63197 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Altenburg, Simon T1 - Measurement of real temperatures in metal powder bed fusion: Hyperspectral thermography N2 - Detailed knowledge about the physics of the PBF-LB/M process is still lacking, and the simulation of the fast and small-scale process is challenging. Especially the experimental validation of complex simulations lacks a suitable measurement technique for temperature distributions at high speeds and spatial resolution. The complicated process physics, specifically the rapidly changing emissivity in and around the meltpool, pose a severe challenge for usual thermographic approaches. Here, we present first results of a hyperspectral measurement approach to reconstruct temperature and emissivity maps during the PBF-LB/M process in a custom manufacturing machine. The camera setup measures the thermal radiation of the process along a line at a rate of 20 kHz, spectrally resolved between 1 µm and 1.6 µm. When the meltpool travels perpendicularly across this line, a typical meltpool can be reconstructed by pointwise fitting for temperature emissivity separation, based on typical spectral emissivities from reference measurements. T2 - Lasers in Manufacturing Conference - LiM CY - Munich, Germany DA - 23.06.2025 KW - PBF-LB/M KW - In situ monitoring KW - Thermography KW - Additive Manufacturing KW - Process monitoring KW - Hyperspectral PY - 2025 AN - OPUS4-63564 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Gupta, Kanhaiya A1 - Poka, Konstantin A1 - Ulbricht, Alexander A1 - Waske, Anja T1 - Identification and authentication of additively manufactured components using their microstructural fingerprint N2 - In the field of additive manufacturing, the ability to uniquely identify and authenticate parts is crucial for certification, logistics, and anti-counterfeiting efforts. This study introduces a novel methodology that leverages the intrinsic microstructural features of additively manufactured components for their identification, authentication, and traceability. Unlike traditional tagging methods, such as embedding QR codes on the surface or within the volume of parts, this approach requires no alteration to the printing process, as it utilizes naturally occurring microstructural characteristics. The proposed workflow involves the analysis of 3D micro-computed tomography data to identify specific voids that meet predefined identification criteria. This method is demonstrated on a batch of 24 parts manufactured with identical process parameters, proving capable of achieving unambiguous identification and authentication. By establishing a tamper-proof link between the physical part and its digital counterpart, this methodology effectively bridges the physical and digital realms. This not only enhances the traceability of additively manufactured parts but also provides a robust tool for integrating digital materials, parts databases, and product passports with their physical counterparts. KW - Authentication KW - Additive manufacturing KW - X-ray Computed Tomography KW - Digital fingerprint KW - Unique identification PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-630356 DO - https://doi.org/10.1016/j.matdes.2025.113986 SN - 1873-4197 VL - 254 SP - 1 EP - 12 PB - Elsevier Ltd. CY - Amsterdam AN - OPUS4-63035 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Serrano-Munoz, Itziar A1 - Agudo Jácome, Leonardo A1 - Thompsom, Sean A1 - Schneider, Judy T1 - On the transferability of post-processing heat treatments designed for PBF-LB IN718 alloys to directed energy deposition specimens N2 - Many processes are being developed for metal additive manufacturing (AM) which vary by their heat source and feedstock. The use of directed energy deposition (DED) is growing due to its ability to build larger structures outside of a contained powder bed chamber. However, the only standard exclusively for post-build heat treatment of AM IN718 is ASTM standard F3055-14a, developed for powder bed fusion (PBF). This study evaluates the applicability of this current heat treatment standard to AM IN718 specimens produced using two methods of DED: laser-blown powder (LP)-DED and arc-wire (AW)-DED. Electron microscopy and X-ray diffraction techniques were used to characterize the specimens in the as-built condition and after the full heat treatment (FHT) specified in F3055. No evidence of remaining Laves phase was observed in the two DED specimens after the FHT. Yield strengths for the DED specimens were 1049 MPa for FHT AW-DED and 1096 MPa for LP-DED, higher than the minimum stated for PBF-LB IN718 of 920 MPa. The size, morphology, inter-spacing, and diffraction patterns of the γ´ and γ´´ strengthening precipitates are found to be similar for both DED processes. Differences were observed in the microstructure evolution where the F3055 heat treatments resulted in partial recrystallization of the grain structure, with a higher content of annealing twins observed in the AW-DED. These microstructural differences correlate with differences in the resulting elongation to failure. Thus, it is proposed that variations in heat treatments are needed for optimizing IN718 produced by different AM processes. KW - Additive manufacturing variants KW - Directed energy deposition (DED) KW - Post-process heat treatments KW - SEM-EBSD and TEM microscopy KW - XRD phase analysis PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-628165 DO - https://doi.org/10.1007/s00170-025-15386-1 SN - 1433-3015 VL - 137 IS - 7-8 SP - 3949 EP - 3965 PB - Springer Science and Business Media LLC AN - OPUS4-62816 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Winkler, Michael T1 - Automated Repair of Gas Turbine Blades Using DED-Arc N2 - Gas turbine blades are critical components in aerospace and power generation, often subject to wear, erosion, and fatigue-induced damage. Traditional repair methods are labor-intensive, costly, prone to inconsistencies, and not rapidly adaptable. This work presents an automated approach for repairing gas turbine blade tips using Wire and Arc Directed Energy Deposition (DED-Arc) in combination with a high-precision point to point registration technique of laser line triangulation (LLT) 3D scans. The proposed workflow begins with affixation of the milled down turbine blade to a work piece manipulator using a 3D printed clamping mechanism and a rough alignment of the turbine tip. Subsequently, the turbine blade’s geometry is acquired using a fully integrated 3D laser triangulation sensor, transforming, and aggregating the captured 2D line data into a 3D scan in the working user coordinate system using live feedback data from a finely calibrated industry robot. This point cloud representation of the real-world turbine blade is then used as the target during an advanced point-to-point shape registration technique transforming the digital representation of the repair process containing all relevant tool path and geometry data into the coordinate system of the real-world turbine blade. Afterwards, the turbine tip is then iteratively repaired whereby the turbine tip geometry is divided into differentiated sections, each with its own optimized process parameter set. A key innovation in this approach is the adaptability of the repair process through a closed-loop monitoring system. After each DED-Arc deposition, a 3D scan is performed to document the deposited geometry, to detect the interaction of the different process parameter sets, to activate an intervention if necessary, and calculate subsequent tool paths based on current geometry data. The results indicate that the combination of precise 3D scan registration with DED-Arc is a viable solution for the industrial-scale repair of gas turbine blades leading to significant reduction in labor, tooling, process, and time related cost. T2 - IIW Assembly CY - Genoa, Italy DA - 22.06.2025 KW - DED-Arc KW - Additive Manufacturing KW - Repair KW - Turbine Blade KW - Automation PY - 2025 AN - OPUS4-63624 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Huo, Wenjie A1 - Schmies, Lennart A1 - Gumenyuk, Andrey A1 - Rethmeier, Michael A1 - Wolter, Katinka T1 - Prediction of mean strain from laser beam welding images and detection of defects via strain curves based on machine learning N2 - With the advancement of machine learning, many predictions and measurements in visual tasks can be achieved by convolutional neural networks (CNNs). Solidification hot cracking is a significant defect in laser beam welding, commonly encountered in practical applications. Existing theories indicate that the formation of cracks is closely related to strain accumulation near the solidification front. In this paper, we first leverage supervised Regression networks to design CNNs that achieve real-time average strain estimation for each frame in the collected welding videos. Two different architectures are proposed and compared: the first model stacks two frames at a set interval and feeds them into the network, while the second model extracts image features individually and predicts the results by calculating the correlation between them. Each network has its own advantages in Terms of computational efficiency and accuracy. Finally, we further train a multilayer perceptron (MLP) classification model that can detect the occurrence of cracks based on the predicted strain behaviors. KW - Laser beam welding KW - Mean strain prediction KW - Solidification cracking detection Convolutional neural networks KW - Convolutional neural networks PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-644495 DO - https://doi.org/10.1016/j.optlastec.2025.113975 SN - 0030-3992 VL - 192, Part F SP - 1 EP - 8 PB - Elsevier Ltd. AN - OPUS4-64449 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -