GEOmetrical Control of part produced by metal Additive Manufacturing – GEOCAM

The AM, more than any other process, causes thermal shocks and successive cooling phases in the part during the production. The characteristics of the materials, the orientation of the part, its location in relation to the walls, the characteristics of the powders and the manufacturing trajectories are all technological parameters that drastically modify the physical and thermomechanical phenomena involved and final result in geometrical defects and leads to significant variability on the part. We are convinced that a joint approach between thermal metrology, physical modelling and simulation is a key to overcome these scientific bottlenecks. GeoCAM project targets three axes:
(a) Statistical analysis of part geometry, with a phenomenological approach based on statistical tools to investigate the influence of physical and process parameters on the geometrical defects of manufactured parts, applied to simple (laboratory) geometry parts and extended to more complex parts (close to industrial parts),
(b) Thermal metrology, with the integration of remote local measurement devices located in the room and in the powder bed (thermocouple, pyrometer and thermal cameras) to characterize the manufacturing conditions and thus, to quantify the various physical phenomena (diffusion, conduction...) according to the conditions of realization,
(c) Simulation of manufacturing at the part scale, with the development of an hybrid approach taking into account thermomechanical phenomena and the basic principle of part generation in successive layers while obtaining results with a reasonable calculation time.

The project has enabled us to gain a better understanding of part deformation processes during additive manufacturing of metal parts on a powder bed, and to develop suitable simulation methods. The first method is based on approximate temperature histories, using layer-by-layer repetition of temperature histories in key areas of the part. The second is based on an explicit scheme, which offers considerable time savings compared with traditional implicit schemes. In order to provide the simulations with data on material properties, particularly of metal powders at high temperatures (>1000°C), samples have been designed specifically for measurement by modulated photothermal radiometry. Finally, an original procedure for measuring part temperatures using thermocouples within the industrial machine was developed to validate temperature maps. A project in the Nouvelle Aquitaine region on the additive manufacturing of lumbar implants has been based on the results obtained, and two funding applications have been submitted to the ANR (the French national research agency) concerning, on the one hand, the choice of process parameters to optimize part manufacturing and, on the other, the simulation method based on explicit schematics.

The methods of statistical analysis of three-dimensional measurements will be applied, not only to other additive manufacturing processes but also, in general, to any manufacturing process. The range of applications is therefore very wide.
As regards the hybrid simulation methods developed, they could be extended to other additive manufacturing processes such as CLAD (Construction Laser Additive Directe) or WAAM (Wire and Arc Additive Manufacturing) processes.
At the end of the project, depending on the relevance of the simulation algorithms developed and the state of the market for AM support software, it is envisaged to work, in partnership with a software development company, in the development and the integration of thermomechanical simulation into a specialized AM software suite.

Ledoux, Y.; Ghaoui, S.; et al. Fast simulation for powder bed fusion process based on thermal field pattern repetitions: application on electron beam melting process. IJAMT. 2023, 131, 585-594.

Ledoux, Y.; Ghaoui, S.; Vo, T.H.; et al. Geometrical defect analysis of overhang geometry produced by electron beam melting: experimental and statistical investigations. IJAMT. 2022, 122, 2059–2075.

Essongue, S.; Ledoux, Y.; Ballu, A. Speeding up mesoscale thermal simulations of powder bed additive manufacturing thanks to the forward Euler time-integration scheme: a critical assessment. Finite Elements in Analysis & Design. 2022, 211.

Ghaoui, S.; Ledoux, Y.; Vignat, F.; et al. Analysis of geometrical defects in overhang fabrications in electron beam melting based on thermomechanical simulations and experimental validations. Additive Manufacturing. 2020, 36, 101557.

The new technical possibilities offered by the Additive Manufacturing process (AM) of metallic materials set the stage for designing and manufacturing new parts with complex and innovative shapes. At this stage of the process development, it is hard to guaranty the quality of part (ie. material health, roughness, geometric defects).
The project focuses on the control of the overall geometry quality of the workpiece manufactured by SLM (Selective Laser Melting) and Electron Beam Melting (EBM) processes. This investigation field corresponds to the major originality of this project since the classical research activities deals with micrometric or macrometric scales of defects. The geometrical distortions are mainly due to the coupling between the thermal gradients imposed during manufacture and the shape evolution during the cooling phase and the successive layer deposits generating important residual stress states. The characteristics of the materials, the relative location of the part into the FA machine (close to the walls), the characteristics of the powders and the laser paths are all technological parameters that drastically modify the physical and thermomechanical phenomena involved.

The first objective of the project is to identify the geometrical defects of the manufactured parts and their variability. Multivariate statistical analysis methods on batches of manufactured parts with different process parameters allow the identification of relations between the physical phenomena of the processes, the parameters of settings and the part geometry.
The second objective is to measure the thermal fields during manufacture process, on an experimental device or on the commercial machine, by means of various remote local measurements in the part or in the powder (thermocouples, pyrometers, IR cameras). This complex instrumentation on an industrial machine will be developed on EBM machine and on an experimental SLM dedicated to research and developed on another project.
The third objective is to simulate the deformation of the parts. The simplest approach is based on the integration of thermal gradients and expansion phenomena within the part; another approach is founded on the simulation of the successive deposition of layers with thermal gradients. In a second step, a complementary approach based on hybrid simulations will be developed in order to drastically reduce the computation time with the development of typical macro-element approach.

Thanks to the generated knowledge, decision support tools will be available for the design, the industrialization of parts and the control of the process. The influence of the parameters on the geometry leads to efficiently define and adjust the process parameters for part quality. More to this, the generated knowledge will be relevant for the part design and will be expressed through expert rules or constraints corresponding to useful data for topological optimization. This work will reduce the development cycle of parts with a first-compliant part (first trial leads to a conform part geometry).

The consortium is composed of two research laboratories: the I2M (Institute of Mechanics and Engineering, Bordeaux) works on the AM process since 2 years on SLM process, has an important knowledge in the field of 3D metrology of parts and surface analysis, in numerical simulation and has revelant expertise in the measurement and simulation of thermal phenomena, G-SCOP (Sciences for Conception, Optimization and Production, Grenoble) has an important know-how in AM, mainly on the EBM process, with many industrial and scientific feedbacks.