Lack Of Fusion Research Articles

Weld Defect Detection of a CMT Arc-Welded Aluminum Alloy Sheet Based on Arc Sound Signal Processing

The cold metal transfer (CMT) process is widely used in thin plate welding because of its characteristics of low heat input and stable arc. In actual production, a larger weld gap, misalignment, or other problems due to assembly error lead to serious welding defects, such as burn-through and a lack of fusion. The arc sound contains a wealth of information related to the quality of the weld. This work analyzes the mechanism of CMT arc sound generation, as well as the correlation between the time–frequency spectrum of the arc sound signal and welding quality. This paper studies the extraction of the multi-channel time–frequency spectrum of an arc sound and inputs it to a custom convolutional neural network for the CMT welding defect identification of thin aluminum alloy plates. The experimental result shows that the average accuracy of the proposed model is 91.49% in the defect identification of a CMT arc-welded aluminum alloy sheet, which is higher than that of the single-channel time–frequency convolutional neural network and other traditional classification models.

Systematic approach to process parameter optimization for laser powder bed fusion of low-alloy steel based on melting modes

In the metal additive manufacturing (AM) process of laser powder bed fusion (LPBF), there are a limited number of materials suitable for producing parts with high density and desired mechanical properties. To establish novel materials, it is essential to determine optimized process parameters in order to overcome process-related challenges and mitigate defects such as lack of fusion, keyholing, and balling. Scaling laws based on thermophysical properties and process parameters can be used to transfer knowledge from other materials or LPBF systems. In this work, a scaling law is used to adjust process parameters for single-track experiments over a wide range, which are laser power PL (100–1000 W), scan speed vs (300–2500 mm/s), and laser spot size ds (0.08–0.25 mm). Compared to existing studies, the parameter range is thus extended towards large laser spot sizes and high laser powers. The scaling law used is based on the calculation of the normalized enthalpy frac{Delta H}{{h}_{s}}. The ratio of the deposited energy density DeltaH and the melting enthalpy hs correlates with the dimensions of the melt pool. According to the aspect ratio {delta }_{mathrm{c}} of the melt pool of each single track, the respective melting mode—conduction, transition, and keyhole mode—was identified. The process parameters of the single tracks in transition mode were used to optimize the density of the LPBF specimens with varying hatch distance hd (0.06–0.12 mm), resulting in specimens with a relative density of > 99.8%. The proposed methodology can accelerate the process parameter finding for new alloys and avoid process-related defects.Graphical

The effect of heat treatment on fatigue response of laser powder bed fused Al-Cu-Mg-Ag-TiB2 (A20X) alloy

The paper aims at assessing the effect of heat treatment on the fatigue behavior of a novel laser-powder bed fused (L-PBF) Al-Cu-Mg-Ag-TiB2 composite, otherwise known as A20X alloy. Heat treatments, (i) stress relieving, and (ii) T7 over-aging and stabilizing, were performed on L-PBF A20X materials, followed by advanced microstructural characterizations, mechanical property measurements (utilizing tensile and depth-sensing indentation testing), and force-controlled fatigue testing. Microstructural assessments revealed the ultra-fine and fine equiaxed grain structure of the L-PBF stress-relieved and aged A20X materials. Mechanical properties (yield stress, ultimate strength, and nano-hardness quantities) of the aged material were recorded as higher than the stress-relieved counter materials. By the same token, the aged material showed better low-cycle and high-cycle fatigue performance and higher fatigue strength relative to the stress-relieved conditions. Based on detailed fractography assessments, the life-limiting fatigue failures of the studied materials (stress relived and T7 aged) were primarily governed by sub-surface L-PBF-induced volumetric defects such as pores and/or lack of fusion. The findings of this study present useful information on the role/mechanism of inevitable process-induced interior defects on fatigue behavior of the Al-Cu-Mg-Ag-TiB2 composites and the effect of post-processing (i.e., aging in this research) on fatigue behavior, which is of great significance for the high-demanding and critical applications of lightweight high-strength L-PBF Al-Cu-Mg-Ag-TiB2 composites in the space and aviation sectors.

Characterization of galvanized steel-low alloy steel arc stud welded joint

This paper investigates the possibility of successfully welding a Low Alloy Steel (LAS) stud to Galvanized Steel (GS) plate.Arc Stud Welding (ASW) was performed on joining LAS studs to GS plates. Welding parameters were selected based on weld trails. The first tests of the welded joints were based on visual inspection for welding defects such as lack of fusion and undercut welding defects. The good quality should be free of these defects and have full weld reinforcement. Other weld qualifications included torque strength test, microhardness test, and microstructure examination.The LAS studs have been successfully welded to a galvanized steel plate using the arc stud welding process. Higher welding current with adjusted welding time (800 A, 0.3 s) gave full weld reinforcement, the best joint appearance, and strength. Martensite phase was detected in the weld area and heat affected zone (HAZ), affecting the joint mechanical properties. Hardness property varied across the welded joint, and maximum hardness was recorded at the HAZ at the stud side. Hardness increased with the increasing welding current. At 800 A, welding current hardness was 10% higher than at 400 and 600 A. Torque strength was affected by weld reinforcement, and 800 A gave the best weld reinforcement that produced the highest torque strength.The main research limitation is the difficulty of welding LAS studs and GS plates. In conventional welding methods, such as gas metal arc welding, it is hard to get full weld penetration due to the geometry restrictions of the joint, which results in partial weld penetration between the studs and the plates. Furthermore, the issue of zinc evaporation during welding can be reduced by the advantage of the very high welding speed (in milliseconds) of ASW that overcomes the problem of continuous welding that usually results in the formation of harmful porosities and poor weldability.In this research, galvanized steel plates were successfully welded to LAS studs using the ASW process. The welding parameters for this dissimilar welding joint were carefully selected. Microstructure changing due to the welding process was investigated. The joint mechanical properties were evaluated.

Experimental investigation on deballing and surface finishing of selective laser melted 18Ni300 steel using polymer rheological abrasive medium

Additively manufactured metallic parts compete with casted and forged parts in physical and mechanical properties. However, the presence of various surface defects, such as balling and staircase effects, lack of fusion, surface undulations, etc., hinders the direct functional application of the as-built part and demands further post-processing. Abrasive flow finishing (AFF) is an extensively employed post-processing technique to finish additively manufactured parts. This process uses polymer rheological abrasive medium (abrasive medium) to finish internal and external surfaces. Even though significant research has been carried out on the surface finishing of selective laser melted metallic parts, the mechanism of material removal and surface finish improvement is not explored much. In the present work, the mechanism of deballing and surface finish improvement on selective laser melted 18Ni300 steel using the AFF process is studied. A newly developed polysaccharide-based abrasive media with monomodal and bimodal abrasive particle distribution are employed for finishing. At an extrusion pressure of 7.5 MPa, the influence of abrasive particle distribution and the number of finishing cycles on the amount of material removed and percentage change in surface roughness (% ΔRa) is observed. It was found that the abrasive medium with bimodal abrasive particle distribution resulted in more material removal and % ΔRa than monomodal. Surface roughness reduction of up to 93% was obtained using the abrasive medium with bimodal abrasive particle distribution. It was noticed that the loosely adhered powder particles were removed within the initial 15 cycles, and the balling effect was completely eliminated within 75 cycles. After 375 cycles, the as-built side surface was free from all other surface defects, such as micro-cracks, shallow pits, material ridges etc. Based on the obtained results, a suitable material removal mechanism in selective laser melting parts using the AFF process was proposed.

Tribological performance under abrasive wear of Fe-Cr-C+Nb coating deposited by FCAW process

This paper aims to investigate the abrasion resistance performance of Iron-Crhomium-Carbon + Niobium (Fe-Cr-C + Nb) coatings deposited by Flux-Cored Arc Welding (FCAW) process. Two coatings conditions were prepared in a 1-layer setup, with a selfshielded tubular wire over SAE 1020 carbon steel substrate, varying the FCAW parameterization levels, searching to reproduce different conditions of welding energy and their respective effects on the coatings' properties. Optical Microscopy (OM) and Scanning Electron Microscopy (SEM) were employed in the microstructure and wear micromechanism analysis. X-ray Energy Dispersive Spectroscopy (EDS) was used for chemical composition assessment and Vickers microdurometer for hardness evaluation. In the tribological attempt, it was used the dry/sand rubber wheel apparatus according to ASTM G65. From the results, in neither sample of the two coating conditions, deleterious defects were observed on the surfaces, as well as in the cross-sections, such as lack of fusion and process interruptions. Dilutions were sufficient to ensure metallurgical bond without compromising layer geometry. Cold cracking characteristics of the process and material were observed, which did not significantly influence the performance of the coatings. The microstructure of both coatings showed CrC and NbC carbides dispersed in the Fe-matrix, whose volume fraction and mean free path between them varied as a function of dilution. The greater Nb-content and less intense heat input in the lower dilution coating provided greater refining of the CrC, NbC, and Fe-matrix strength, which was reflected in higher hardness, lower wear rate, and less severe abrasive wear micromechanism.

A coupled ray-tracing based CFD and cellular automaton model for predicting molten pool formation and microstructure evolution in narrow gap laser welding

Laser beam reflection plays a significant role in the temperature development history during various laser material processes. Different from other processes where laser beam reflection main happens inside the keyhole, in narrow gap laser welding (NGLW), the laser beam would interact with the sidewalls multiple times before existing the narrow groove, which would have a great impact on the defect formation as well as the grain structure development. However, the underlying mechanism of this process remains unknown in multi-pass NGLW. In this study, an integrated ray-tracing based CFD and cellular automaton (CA) modelling framework is established to investigate the impact of laser beam reflection on molten pool formation and microstructure evolution in multi-pass NGLW. From the CFD simulation, it is found that the overall laser beam absorption of NGLW is greater than laser metal deposition (LMD) while smaller than laser powder bed fusion (LPBF) process. Besides, the overall absorptivity in NGLW is increased 28% when considering the ray reflection during the first welding pass. The absorptivity would increase and then stabilise through the multi-pass welding process, which is attributed to the change in incident angle and number of reflections. Additionally, it is found that the undercut defect is more likely to occur with ray reflection due to the excessive temperature rise on the sidewalls, while lack of fusion is observed in the case without ray reflection. By incorporating the thermal history of each welding pass with CA, the impact of laser beam reflection on grain structure development is evaluated. Columnar grains would govern the microstructure evolution in the case with ray reflection, which grow epitaxially from the fusion line to the molten pool centre. Grain coarsening would occur around the lack of fusion region in the case without ray reflection. Moreover, the analysis on grain size distribution is conducted and a greater maximum grain size is obtained in the case without ray reflection as a result of lower cooling rate caused by lack of fusion. However, a higher grain size increase is obtained in the case with ray reflection during multiple welding passes, and larger grains are expected to form if more welding passes are to be conducted. The simulated molten pool morphology and grain structure are compared with the experimental observation, and a good agreement is achieved.

3D numerical modeling for thermo-mechanical behavior of additively manufactured titanium alloy parts with process-induced defects

The defects due to the rapid heating and cooling in powder bed fusion affect significantly the mechanical properties of additively manufactured as-built parts. However, the effect of process-induced defects on the thermo-mechanical behaviors of as-built parts is still unclear. In this paper, considering the process-induced interfacial lack of fusion (LoF) and pores, we developed a coupled thermo-mechanical model of titanium alloy parts in the powder bed fusion process. The equivalent temperature-dependent thermal properties of the parts with defects are calculated based on mixture theory. The three-dimensional thermal, mechanical, and displacement fields are predicted based on the differential quadrature method (DQM). The accuracy of the proposed model is verified by the comparison with the predictions and experimental data in the references. The effect of crucial parameters on the thermo-mechanical responses of as-built parts is analyzed, including energy densities, scan patterns, preheating temperature, LoF dimension, porosity, and powder layer thickness. The predicted results show that the process and geometry parameters play a vital role in the temperature distribution, residual stress buildup, and deformation of the printed parts. In addition, the LoF defect triggers the stress jump at the interface of the defect/powder bed and powder bed/substrate. The predicted results can be beneficial to provide a guideline for the design and additive manufacturing of titanium alloys.

Deep Multi-Modal U-Net Fusion Methodology of Thermal and Ultrasonic Images for Porosity Detection in Additive Manufacturing

AbstractWe developed a deep fusion methodology of nondestructive in-situ thermal and ex-situ ultrasonic images for porosity detection in laser-based additive manufacturing (LBAM). A core challenge with the LBAM is the lack of fusion between successive layers of printed metal. Ultrasonic imaging can capture structural abnormalities by passing waves through successive layers. Alternatively, in-situ thermal images track the thermal history during fabrication. The proposed sensor fusion U-Net methodology fills the gap in fusing in-situ images with ex-situ images by employing a two-branch convolutional neural network (CNN) for feature extraction and segmentation to produce a 2D image of porosity. We modify the U-Net framework with the inception and long short term memory (LSTM) blocks. We validate the models by comparing our single modality models and fusion models with ground truth X-ray computed tomography (XCT) images. The inception U-Net fusion model achieved the highest mean intersection over union score of 0.93.

Evaluation of the role of hatch-spacing variation in a lack-of-fusion defect prediction criterion for laser-based powder bed fusion processes

Lack of fusion (LOF) defects impact adversely on the mechanical properties of additively manufactured components produced via laser-based powder bed fusion. Following a stress-relieving heat treatment, the tensile properties and hardness of Ti6Al4V components were found to be negatively impacted by the presence of LOF defects. This work considers a geometrical-based inequality for the prediction of LOF defects. We critically evaluate an LOF criterion using both the experimentally and analytically obtained melt pool geometries. Experimentally, we determined melt pool dimensions by analysing a single-layer, multi-track deposition with oversized hatch spacing in order to establish depth and width from non-overlapping melt pools. Analytically, Rosenthal-based predictions of melt pool size (width and depth) are applied. To investigate LOF defects, we used hatch spacing as the main parameter variation to investigate defects while keeping all other controllable parameters unchanged. An original LOF criterion from the literature was found to be an adequate predictor of LOF defects when experimentally obtained melt pool geometry was used. Critically, however, the analytical expressions for melt pool geometry were found to be in error and this caused the LOF criterion to fail in predicting LOF defects in all cases where defects were observed experimentally. However, an adaptation to the LOF prediction criterion is proposed whereby it is recommended that a correction factor {R}_{c}^{2}=0.7 (or {R}_{c}=0.83) is used with the analytically derived melt pool geometry. Furthermore, this correction is extended into the laser power versus scanning speed operating space to give minimum (corrected) line energy for LOF avoidance in Ti6Al4V components.

Real-time process monitoring and closed-loop control on laser power via a customized laser powder bed fusion platform

Additive manufacturing (AM) is one of the most effective ways to fabricate parts with complex geometries using various materials. However, AM also suffers from printing quality issues resulting from the defects such as over-melt, lack of fusion, swelling, etc. One of the root causes of those issues is that the process parameters remain constant during the entire printing process, regardless of the dynamic heat accumulation and various printing feature sizes. For instance, raster is the most common scanning strategy in the laser powder bed fusion (L-PBF) process. The length of the raster line varies depending on the printing feature size. When scanning small features, the raster line is short, resulting in heat accumulations and over-melt. These variabilities may cause severe quality issues and thus suggest adaptive process parameters be applied. Aiming to address this challenge, this study develops a closed-loop control system to regulate the laser power based on melt pool thermal emission to avoid over-melt, balling, and high surface roughness. The control target is determined by correlating the printing quality (dimensional printing error in this study) with the thermal emission through thin-line printing trials using variable power. A high-speed thermal sensor and controller are designed, tuned, and implemented on a newly developed L-PBF testbed. The system successfully maintains a low dimensional error by regulating the laser power at 2 kHz. A significant improvement in printing quality was achieved, as validated by both microscopic imaging and 3D scanning.

Processing of Haynes® 282® Alloy by Direct Energy Deposition with Arc and Wire

Direct energy deposition with arc and wire (DED-AW) is a versatile, low-cost, and energy-efficient technology for additive manufacturing of medium- and large-sized metallic components. In this study, the effects of arc energy and shielding gas in cold metal transfer (CMT) welding of walls and blocks on cooling time, mechanical properties, and macro- and microstructure have been studied using precipitation-hardenable Ni-based superalloy Haynes® 282®. The arc energy and consequently the cooling rate were varied by changing the wire feed rate and the travel speed. As expected, increasing the arc energy leads to higher cooling times for the walls. Due to the 2D thermal conduction, the thin walls cool down much slower than multi-layer welded blocks, but this reduces the strength values only very slightly. While the walls have no sensitivity to the occurrence of unacceptable seam irregularities, the multi-layer blocks show isolated seam defects, such as hot cracks or lack of fusion. Despite shielding gas variation, the as-welded blocks show acceptable strength properties at room temperatures (RT) and impact values at RT and -196 °C. However, the use of an N-containing shielding gas results in lower elongation and notched bar impact energy. Precipitation-hardened specimens tested at 871 °C exhibit a similar strength level to transverse tensile specimens of gas metal arc welding (GMAW) welded joints on 12.7 mm thick plates with fracture in the weld metal.

XFEM Analysis of Strain Rate Dependent Mechanical Properties of Additively Manufactured 17-4 Precipitation Hardening Stainless Steel

AbstractAdditively manufactured (AM) specimens of 17-4 precipitation hardening (PH) stainless steel (SS) corresponding to the three-point bend test, compact tension test, and single edge cracks were analyzed using the extended finite element method (XFEM) approach. A two-dimensional and three-dimensional elastic-plastic simulation were conducted using “abaqus 6.14” software based on the experimental results and validated with the simulation results. In XFEM, the partition of unity was used to model a crack in the standard finite element mesh. Based on simulation results, the present study compares the mechanical properties of AM 17-4 PH stainless steel samples with those of wrought 17-4 PH samples. Stress intensity factor and J integral were used to measure fracture toughness of the specimens. The change in fracture toughness with strain rate was evaluated by simulating two-dimensional compact tension specimens. The presence of defects such as pores resulting from entrapped gas, un-melted regions, and powder particles resulting from lack of fusion were the main reasons for lower elongation to failure of laser powder bed fusion (L-PBF) produced 17-4 PH SS reported in the literature.

A Method to Optimize Parameters Development in L-PBF Based on Single and Multitracks Analysis: A Case Study on Inconel 718 Alloy

In the context of the use of AM, particularly in the L-PBF technique, the printability characterization of material occurs through the identification of its printability map as a function of printing process parameters. The printability map identifies the region where the powder melting is optimal and ensures a dense and defect-free material. Identifying the zones affected by physical phenomena that occur during the printing process which lead to material defects such as keyhole, lack of fusion and balling mode is also possible. Classical methods for the characterization of material and the identification of its printability map require the printing of a large number of specimens. The analysis of the specimens is currently time-consuming and costly. This paper proposed a methodology to identify optimal process parameters in L-PBF using an integrated single and multi-tracks analyses embedded in an overall algorithm with detailed metrics and specific factors. The main scope is to speed up the identification of printability window and, consequently, material characterization, reducing the number of micrographic analyses. The method is validated through an experimental campaign assessing the material microstructure in terms of porosity and melt pool evaluation. The case study on IN718 superalloy shows how the application of the proposed method allows an important reduction of micrographic analysis. The results obtained in the case study are a reduction of 25% for the complete definition of the printability map and more than 90% for identifying the zone with a high productivity rate.

Effect of the Position in the Build Chamber on the Fatigue Strength of Additively Manufactured Maraging Steel MS1

The quality of additively produced parts and the achievable mechanical response may be affected by several factors, such as build orientation, heat treatment, or machining. A further rarely investigated factor is the position of the built part in the chamber with respect to inert gas flow. Previous studies have highlighted that the interaction between gas flow and laser track may induce an intense vaporization with consequent lack of fusion, particle entrainment, drop in density and denudation of the produced part, which is likely to detrimentally affect mechanical properties. This study addresses the effect of part position on the fatigue strength of heat-treated maraging steel MS1 produced by an EOSINT M280 machine in a nitrogen environment. Novelty arises from the lack of studies in this field, especially under fatigue. A factorial plan with subsequent statistical analysis highlighted that positioning the part upstream with respect to the gas flow leads to a slightly lower fatigue strength; however, no significant differences are observed. The failure mode, involving initiation from subsurface porosities of the same size, is also unaffected. Finally, a fatigue limit of 26% of the ultimate tensile strength is found, which is consistent with previous outcomes.

The Effect of Postprocessing on the Fatigue Properties of Ti-5Al-5Mo-5V-1Cr-1Fe Produced Using Electron Beam Melting.

Despite the significant potential advantages of processing Ti-5Al-5Mo-5V-1Cr-1Fe alloy (Ti-55511) using Electron Beam Melting (PBF-EB/M), when compared to conventional manufacturing technologies, the resulting internal defects are an important characteristic of such additive technologies and can highly decrease mechanical properties. One of the most dangerous defects formed during metal additive manufacturing processes are material discontinuities such as a lack of fusion. Defects of this type, due to their "flat" nature, are difficult to characterize. For cycle-loaded specimens, where the loading force acts perpendicular to the lack-of-fusion plane, defects of this type can significantly reduce fatigue properties. This paper presents the results of research aimed at improving the fatigue properties of Ti55511 alloy by reducing the influence of the lack-of-fusion defect on fatigue damage. The static and fatigue properties of specimens in the as-built state, as well as after hot isostatic pressing (HIP) treatment, were analyzed. The effect of HIP on both the reduction of pores and the degree of sphericity when using the X-ray computed tomography (XCT) system was presented. The change in the microstructure after HIP was analyzed in terms of the change in the size of individual phases, as well as the change in the phase ratio. This paper also contains a fractographic analysis of the samples after tensile and fatigue tests.

A multi-scale experimental investigation for fatigue limit and fatigue crack initiation behavior of powder bed fusion-laser beam 316L stainless steel

In this study, the fatigue limit of powder bed fusion-laser beam (PBF-LB) 316L prepared by various printing power intensities was studied. After finding the PBF-LB printing parameters giving the best fatigue limit, the fatigue crack initiation (FCI) behavior of dense PBF-LB 316L manufactured by the parameters corresponding to the best fatigue limit was investigated. To this end, the fatigue limit was evaluated by self-heating experiments, and a novel method based on the acoustic emission (AE) technique was suggested to assess the FCI stage during the whole fatigue process. In addition, the AE source mechanism was studied to validate the proposed method, and the fatigue behavior and the fatigue cracking mechanism were deeply clarified through detailed characterizations of microstructures and metallurgical defects. The results indicate that the fatigue limit of dense PBF-LB 316L is affected by a combined effect of the subgrain size and gas pores. However, for the defective PBF-LB 316L, the LOF (lack of fusion) defects induced by low printing power play a crucial role in deteriorating the fatigue limit mainly by promoting FCI. Besides, the printing parameter corresponding to the best fatigue limit is found to be 60 J/mm3 in terms of the volume energy density (VED). Furthermore, as for the FCI behavior, it is shown that the percentage of the FCI stage of dense PBF-LB 316L ranges from 12.4% to 45.3% during the whole fatigue process, and the fatigue crack scale after FCI is on a micro level (from 3.8 to 46.6 μm). Moreover, areas around the fine grains (FGs) and coarse grains (CGs) regions and the junctions between ferrite and austenite are the sites where fatigue cracks are prone to initiate.

Multi-track laser metal deposition of Stellite6 on martensitic stainless steel: Geometry optimization and defects suppression

Direct Laser Deposition has been considered as a method to repair and rebuild the industrial parts. Considering the possibility of the formation of structural defects due to the process parameters, assessing the selection of the optimum laser cladding parameters to minimize these defects is a crucial subject. Therefore, a model to predict the single-track geometric characteristics based on the three main process parameters namely, laser power, scanning speed and powder feed rate was presented to prevent the formation of defects in the deposition of Stellite 6 on martensitic stainless steel. Application of the linear regression method coupled with genetic optimization algorithm resulted in the optimum combined parameters (PαVβFγ), which were the basis for the generation of the process map. Eventually, based on track geometric attributes, an optimum region was introduced on the process map and to validate the capability of the prediction model, three single tracks were selected for multi-track laser deposition with a 30%, overlapping ratio concluded from derived equations. Investigation of these three multi-tracks indicated that the one with the least dilution (2.14 %) and the most wetting angle (52°), contained various defects, namely lack of fusion, delamination, and unmelted powder defect. Furthermore, the multi-track with high dilution resulted in the penetration of alloying elements from the substrate to the deposited clad and vice versa. Finally, the multi-track which was selected from the optimum region, presented the best deposited layer with no visible defects, which validated the proposed model.

Use of ResNet modelling for TIG weld feature digitization and correlation : a technique for AI based welding system

TIG Welding is being practiced in the manufacturing industry and it demands highly skilled labour. Artificial Intelligence (AI) is developing rapidly as researchers are constantly finding new ways in which intelligent machines can add value to their industry. An AI-based welding system stands to add value by increasing production rates, improving safety, and decreasing the human input required. Weld monitoring is a key activity in the TIG welding process and successful use of AI system will enable failure prediction and the proactive corrective actions. The aim of this project is to explore, test, and compare ResNet modelling based machine learning algorithms and examine their ability to monitor welds. In this project the weld monitoring process includes collecting images of weld joint for weld feature digitization. Also, the study enables predicting whether the weld shows good quality, contamination, burn through, misalignment, lack of fusion, or lack of penetration through a ResNet modelling based image analysis.

Material properties of AISI H10 (32CrMoV12-28) hot work tool steel processed by Laser Powder Bed Fusion with 200°C substrate preheating temperature

In this study the processability of hot work tool steel AISI H10 (32CrMoV12-28) using the Laser Powder Bed Fusion (L-PBF) process was investigated. The tool steel was processed in two build jobs with different parameter settings. The produced samples were examined for defects such as cracks, pores and lack of fusion. Due to the high cooling rates in the L-PBF process and the resulting residual stresses, cracks were found despite substrate preheating. In a second build job the process parameters were adapted without changing the preheating temperature of 200°C. Changing the process parameters led to crack-free AISI H10 hot work tool steel samples. Process parameters with the highest density and lowest number of defects were used to produce tensile and notch impact samples which were used to determine the mechanical properties.