Volume 75, Issue 4

14,000 data sets from an industrialized bainitization process, consisting of process gas furnace, salt bath and circulating air furnace, were used to predict the resulting Vickers hardness of cylinder heads made of 100Cr6 based on process data such as temperature and pressure. For prediction, machine learning methods such as ANNs, CNNs, ensemble methods and support vector regressors were compared. Meta features such as the furnace number as well as features extracted from the recorded time series were used. Data preparation and feature extraction were performed according to the machine learning methods used. The random forest achieved the best predictions with an R 2 score of 0.406 and also allows the evaluation of the most important features.

In case-hardened steel components, fine-grain stability is one of the most important properties of the microstructure, since a single coarse grain in a fine-grain matrix is often sufficient to cause premature failure of the component under appropriate loading. In order to better understand the influence of heat treatments on cold formed components, the case hardened microstructure of 16 different heat treatment combinations was investigated on the case hardening steel 20MnCr5 (1.7168). The process chain influences grain growth inhibiting precipitations, which in turn influence the fine grain stability. For this reason, aluminium nitrides in size ranges from approx. 15 to 250 nm were analysed in two samples using high-resolution scanning electron microscopy in both fine and coarse-grained areas. The respective areas examined had an average area of 1,250−2,000 μm 2 . The statistical analysis showed that within the samples no significant difference in morphology and density between aluminium nitrides in fine and coarse-grained areas could be found. On the other hand, more clusters appear in larger grains. Furthermore, a significant influence of the heat treatments on the aluminium nitrides could be detected.

Eddy current testing is particularly suitable for non-destructive 100 % testing in the near-surface area of electrically conductive materials due to its high sensitivity, high testing speed and the possibility of easy automation. Eddy current testing is therefore frequently used, especially in safety-relevant applications such as aviation. After a short description of the basics of eddy current testing, this paper reports on experiences with the testing of rolling element bearing components for engine bearings by means of eddy current. The advantages of the multi-frequency method in material mix-up testing and estimation of the retained austenite content are clearly shown. It is also shown, however, that in particular cases material-typical properties can lead to sham indications.

Directed energy deposition is a metal additive manufacturing process that builds parts by joining material in a layer-by-layer fashion on a substrate. Those parts are exposed to rapid thermo-cycles which cause steep stress gradients and the layer-upon-layer manufacturing fosters an anisotropic microstructure, therefore stress relief heat treatment is necessary. The numerical simulation can be used to find suitable parameters for the heat treatment and to reduce the necessary efforts to perform an effective stress relieving. Suitable validation experiments are necessary to verify the results of the numerical simulation. In this paper, a 3D coupled thermo-mechanical model is used to simulate the heat treatment of an additive manufactured component to investigate the application of a C-ring geometry for the distortion-based validation of the numerical simulation. Therefore, the C-ring samples were 3D scanned using a structured light 3D scanner to quantify the distortion after each process step.