In the context of sustainability, the repair of damaged components is of paramount importance. Rolling bearings facilitate the operation of rotating components with high loads and torques, which are utilised in various applications and can attain diameters of several metres. The larger the bearing, the more economically and ecologically viable the repair becomes. In the present study, the repair was achieved through the development of a process chain based on welding processes. To demonstrate the effectiveness of the proposed strategy, an example is provided of a bearing disc of the type 81212 made of 100Cr6 steel. The process started with the removal of the defect area at the bearing by turning. The groove was then filled by laser-based directed energy deposition (L-DED). A wear-resistant steel containing chromium, nickel, and vanadium was employed as the cladding material as part of a tailored forming method. This type of repair represents a novel approach, where the goal is to realize new ways to produce load-adjusted hybrid solid components. One challenge encountered in processing this high-alloy steel was the presence of welding defects through the repair process. To optimize the quality, a preheating process based on induction was introduced, which led to a significant reduction of defects by 88.3% based on a scanning ultrasonic microscopy analysis. After the geometric contour was restored through machining, a heat treatment was applied, leading to a modification of the carbides and a more refined microstructure. This was found to significantly increase the hardness of the base material by 109% resulting in 780 HV0.1 and the cladding by 8.3% resulting in 753 HV0.1, bringing it into a range similar to that observed prior to the repair (803 HV0.1).

Laser-based directed energy deposition (L-DED) enables the additive manufacturing of copper alloys for electrical applications, where electrical conductivity is a key functional property. This study investigated the effects of laser power, welding speed, and powder feed rate on the microstructure and properties of L-DED-fabricated CuSi3Mn1 walls. Electrical conductivity was evaluated in horizontal and vertical directions alongside grain size, porosity, hardness, and oxygen and hydrogen contents to investigate the relationships between processing conditions and material properties. Increasing laser power reduced the mean porosity by approximately 60% while promoting grain coarsening. Higher laser power and lower welding speed resulted in reduced hardness and lower oxygen content. Horizontal electrical conductivity decreased with increasing laser power and powder feed rate, despite the concurrent reduction in porosity and increase in grain size associated with higher laser power. In contrast, vertical electrical conductivity decreased with increasing welding speed, potentially due to differences in the number of interfaces between deposited layers. These results reveal direction-dependent effects of L-DED process parameters on electrical conductivity and demonstrate that changes in porosity and grain size alone cannot account for the observed conductivity trends.

Contact arc metal grinding (CAMG) is a thermally driven underwater cutting method developed for decommissioning in nuclear facilities. In contrast to conventional mechanical cutting methods, CAMG employs a hydraulically actuated rotating disk electrode that is supplied with high electric current. As the electrode approaches the workpiece, an electric arc is ignited, generating localized melting that facilitates the controlled removal of material. Inherent radiation shielding, effectively prevents the spread of airborne contamination, and enhances overall safety of the operating personnel. Thus, CAMG offers a particularly effective solution for the dismantling of highly activated or hard to access components where traditional mechanical tools cannot be used. Monitoring electrode wear plays a vital role in CAMG. The condition of the electrode can be evaluated through the analysis of current and voltage signals recorded during operation. The present study introduces a novel deep learning based approach for monitoring electrode wear in CAMG using signals recorded from experimental CAMG trials. Convolutional neural networks (CNN) are used to extract representative features from the signals, followed by Principal Component Analysis and t-distributed Stochastic Neighbour Embedding to evaluate clustering behaviour. The results show that wear progression is reflected in distinctive temporal patterns. Worn electrodes exhibit fewer characteristic peaks and lower electrical power. CNN-derived latent representations form clearly separable clusters corresponding to early, intermediate, and advanced wear levels. These findings demonstrate that combining CNN-based feature extraction with unsupervised clustering provides a reliable basis for real-time electrode condition assessment in CAMG and supports progress toward automation in underwater dismantling operations.

Contact Arc Metal Grinding (CAMG) is a fast, robust underwater cutting process. This makes it a potentially valuable solution for nuclear decommissioning. Until now, the measurement of tool wear has been based on parameters such as diameter, volume, or weight of the electrode. All of these variables require direct manipulation and contact with the electrode in order to be determined. In contaminated environments, contactless condition monitoring is crucial. The study proposes and evaluates a signal-based, contactless indicator for tool electrode wear, derived from electrical process data. Transient current signals were transformed into histograms, and the integral over the high-current region was used as a wear descriptor. CAMG trials with various additively manufactured tool electrodes with functional areas (different monolithic metallic alloys and CuSi3Mn1-based composites with fused WC) were conducted under water to evaluate this feature. A dedicated model test setup with different electrodes and high-speed imaging for the isolated observation of individual arc events was designed to facilitate the interpretation of process signals. For monolithic metallic functional areas and a reference steel electrode, the high-current integral decreased monotonically with wear and showed strong, statistically significant negative linear correlations. Despite initial disparities in levels, most notably for the thinner reference disc, all electrodes converged to similar values at the end of service life, once the functional area had been consumed. In contrast, CuSi3Mn1-based composites with fused WC particles exhibited weak or non-significant correlations, consistent with wear-driven changes in surface composition that alter local contact and arc attachment conditions. High-speed imaging was employed to confirm the occurrence of discrete arc events and to demonstrate that discontinuous voltage transients are consistent with rapid current interruptions and inductive voltage spikes. This enabled a correlation previously assumed in the literature to be confirmed experimentally for the first time.