Experimental and computational appraisal of the shape accuracy of a thin-walled virole aero-engine casing manufactured by means of laser metal deposition

Finite element analysis of in-situ distortion and bulging for an arbitrarily curved additive manufacturing directed energy deposition geometry

“…The method includes all elements into the simulation from the start but reduces the material properties of the elements that are added due to material deposition with a penalty factor of 10 −5 to the temperature dependent thermal and mechanical properties except for heat capacity. In order to prevent excessive deformation of the deactivated elements prior to activation as observed in [2], the Young's Modulus was set to a higher value of 10 GPa instead of applying the full penalty factor. After first trials, this formulation was identified as essential to model bulging: due to the increased stiffness of inactive elements the deformation in higher layers before activation can be reduced and a comparison is shown between 1 GPa and 10 GPa Young's modulus for the inactive zone.…”

“…The transformation allows the consideration of the distortion evolution regardless of the local curvature and is identified to be a necessary step in the evaluation of a curved part. Similar coordinate system notations were already utilized by Papadakis [2] and Ghasri-Khouzani [8] for cylinder and disc builds, respectively. The results are shown in Figure 4 for tangent, normal and height displacements with experimental mean and standard deviations from three independent trials and the simulation results superimposed at all six investigated points.…”

“…During welding, the substrate and the already-deposited material are re-heated periodically with each new layer creating complex interlocking effects such as inhomogeneous thermal strains, local melting and shrinking, annealing and phase transformations. When building DED parts such as turbine housings [2] or excavator arms [3], the whole process is set up experimentally and has to be iterated a number of times to identify suitable parameters regarding -for instance -power and material feed, path planning strategies and cooling times between layers. These setups consume valuable personnel and machine times, before an adequate result quality is achieved and the part can go into production.…”

“…the sample cannot be prepared before welding. To work around this difficulty, several industrial-scale AM-parts were simulated and compared to measurements taken on the finished part: Papadakis et al [2] conducted a structural numerical welding simulation for a large aero-engine housing build and compared the results to experimental 3D-scans on the finished component. Marimuthu et al [6] calculated the displacement of the substrate in an aero-engine build and compared it to ex-situ coordinate machine measurements (CMM).…”

“…Consistent with recent studies employing 3D-optical scanning, the sample is removed from the substrate and the distortions due to stress relaxation as well as the final simulated and actual part shapes are discussed. As described in literature [2,23], structural simulation models tend to estimate bending of the whole geometry instead of bulging in DED. To address the bulging modelling an addition to the structural simulation approach is introduced: The Young's Modulus in the not-yet activated upper layers is artificially increased to inhibit bending and keep the model upright for bulging prediction.…”