Powder bed fusion (PBF) is the most commonly adopted additive manufacturing process for fabricating complex metal parts via the layer-wise melting of a powder bed using a laser beam. However, the qualification of PBF-manufactured parts remains challenging and expensive, thereby limiting the broader industrialization of the technology. Powder characteristics significantly influence part properties, and understanding the influencing factors contributes to effective quality standards for PBF. In this study, the influence of the particle size distribution (PSD) median and width on powder flowability and part properties is investigated. Seven gas-atomized SS316L powders with monomodal PSDs, a median particle size ranging from 10 μm to 60 μm, and a distribution width of 15 μm and 30 μm were analyzed and subsequently processed. The PBF-manufactured parts were analyzed in terms of density and melt pool dimensions. Although powder flowability was inversely related to the median particle size, it was unrelated to the distribution width. An inverse relationship between the median particle size and the part density was observed; however, no link was found to the distribution width. Likely, the melt pool depth and width fluctuation significantly influence the part density. The melt pool depth decreases and the width fluctuation increases with an increasing median particle size.
Powder bed fusion (PBF) is the most commonly used additive manufacturing process for fabricating complex metal parts via the layer-wise melting of powder. Despite the tremendous recent technological development of PBF, manufactured parts still lack consistent quality in terms of part properties such as dimensional accuracy, surface roughness, or relative density. In addition to process-inherent variability, this is mainly owing to a knowledge gap in the understanding of process influences and the inability to adequately control them during part production. Eddy current testing (ECT) is a well-established nondestructive testing technique primarily used to detect near-surface defects and measure material properties such as electrical conductivity in metal parts. Hence, it is an appropriate technology for the layer-wise measuring of the material properties of the fused material in PBF. This study evaluates ECT’s potential as a novel in situ monitoring technology for relative part density in PBF. Parts made from SS316L and AlSi10Mg with different densities are manufactured on a PBF machine. These parts are subsequently measured using ECT, as well as the resulting signals correlated with the relative part density. The results indicate a statistically significant and strong correlation (316L: r(8) = 0.998, p < 0.001, AlSi10Mg: r(8) = 0.992, p < 0.001) between relative part density and the ECT signal component, which is mainly affected by the electrical conductivity of the part. The results indicate that ECT has the potential to evolve into an effective technology for the layer-wise measuring of relative part density during the PBF process.
10 April 2024 in Progress in Additive Manufacturing
Powder bed fusion of metals (PBF-LB/M) is currently the most widely adopted additive manufacturing technology for the fabrication of metal parts. However, the inconsistent quality of PBF-LB/M-manufactured parts and high costs for part certification are impeding wider industrial adoption. In-situ monitoring technologies are expected to enable process control in order to ensure consistent quality, and to replace some of the post-process inspection steps, therefore, reducing part certification costs. Eddy current testing (ECT) is a standardized nondestructive testing technique, which can be used as an in-situ monitoring technology to measure the part quality during the PBF-LB/M build cycle. However, the process-induced complex temperature fields in PBF-LB/M parts during the build cycle are among the most relevant disturbances due to the temperature dependence of the electrical conductivity. This study investigates the process-induced temperature influence on in-situ monitoring of relative density using ECT. Parts made from AlSi10Mg were manufactured on a PBF-LB/M machine and the build cycle was monitored using ECT and an infrared camera, which was used to extract the part surface temperature right before the ECT measurement. The results demonstrate that the temperature increase of the parts during the build cycle decreases the electrical conductivity independently of the relative part density, which was measured via micro-computed tomography. Therefore, a temperature compensation method was proposed and applied demonstrating that a layer-to-layer difference of 0.15 % relative density can be detected via ECT. Consequently, it has been demonstrated that ECT is an effective in-situ monitoring technology for PBF-LB/M, even in the presence of temperature disparities within parts.
Background/Objectives: To evaluate the diagnostic accuracy of two visual large language models (vLLMs), GPT-4o (OpenAI) and Claude Sonnet 3.5 (Anthropic), for detecting brain metastases in routine MRI using combined imaging and textual input. Methods: This retrospective study included 31 patients with and 46 without brain metastases with underlying melanoma (n = 24), lung cancer (n = 23), breast cancer (n = 17), or renal cell carcinoma (n = 13). In total, 100 MRI examinations (50 with, 50 without metastases) were provided to both vLLMs using a single representative slice per sequence, together with clinical history and the referring question. The generated free-text reports were evaluated for detection accuracy, overdiagnosis, correct sequence recognition, anatomical localization, lesion laterality, and lesion size estimation. Results: Both vLLMs showed perfect sensitivity (100% for both) but very low specificity (GPT-4o: 8%, Sonnet 3.5: 4%; p = 0.625), resulting in low diagnostic accuracy (GPT-4o: 54%, Sonnet 3.5: 52%; p = 0.625). Sequence identification was highly accurate in both models, with GPT-4o performing significantly better (100% vs. 93%; p < 0.05). Identification of the anatomical brain region (70% vs. 72%; p = 1.00) and lesion laterality (62% vs. 76%; p = 0.189) was comparable. Both models hallucinated additional lesions in 12% of cases. Lesion size measurements showed no significant differences between the models or in comparison with the radiologist. Conclusions: GPT-4o and Claude Sonnet 3.5 can generate radiological reports and detect brain metastases with excellent sensitivity, but their very low specificity, frequent hallucinations, and limited spatial reliability currently preclude clinical application. Future work should address how the balance between visual and textual input influences diagnostic behavior in vLLMs.