Added Value: Max design, min waste for Clean Sky's ASCENT AM
The weight-saving advantages and innovative design possibilities made available by additive manufacturing (AM), also known as 3D printing, are broadly recognised in the aerospace industry. However despite these advantages, verifying AM process reliability and product consistency consumes time and resources. Specifically, due to the high temperatures involved in building up multiple thin layers of metal, heat sinks (or gradients) of temperature occur.
These cause deformation of the part during the process, and the final component that emerges from the process often ends up being distorted in comparison to the geometry that was intended.
Clean Sky’s ASCENT AM (Adding Simulation to the Corporate Environment for Additive Manufacturing) project, which successfully concluded last year, developed a simulation tool that predicts and compensates for this deviation from the intended geometry by producing AM manufactured components that are inherently distorted — but once cooled and removed from the AM machine, revert to the correct shape.
By doing so, the resulting software developed in ASCENT AM enables designers – even without specialist additive manufacturing knowledge – to avoid material wastage by eliminating the 'trial and error' approach that previously existed in AM manufacturing, and replacing it with a 'first time right' process.
The ASCENT project, which is funded by the European Commission under Horizon 2020, achieves this by calculating temperature fields occurring in the component and the resulting deformations as well as residual stresses. The software even compensates for deformations in the post-processing phase of manufacture, which occurs after the component has been removed from the AM printing machine.
‘In our software we tried to simulate the whole process and get the right dimensional part after the process chain,’ says Daniel Wolf, coordinator of ASCENT AM at the Institute for Machine Tools and Industrial Management within the Technical University of Munich (TUM). The project was carried out in collaboration with project topic manager MTU Aero Engines AG in Munich.
‘A key factor was to ensure that simulation time would always be faster than the real printing process – that was one goal, and we managed to achieve that. Right now this simulation software is in daily use at MTU Aero Engines, and they're really happy with the results,’ he says.
The actual test components that were 3D printed during the project included turbine blades. These are particularly complex, where producing the correct curvature is critically important. Obstacles addressed in the project included simulating the components' supporting structures which are often of mesh construction. The intricacy of these finely detailed structures is challenging to replicate in digital simulation.
‘We managed to gain material parameters for these support structures. That allows us to discretise the support structure in the simulation as a whole block. To do that, we carried out many experiments to study how support structures behave, which is quite hard to do with Inconel (the nickel-chromium-based superalloys used for 3D printing in this project).’
Another challenge was to develop a creep simulation subroutine to compensate for the phenomenon where creep occurs. (Creep is the tendency of materials to deform due to stress or high temperatures).
‘For additive manufacturing, creep is not very well understood, and the normal creep subroutines don't work very well, so we have to develop our own creep subroutines to get good results,’ Wolf says.
Another task carried out by the ASCENT AM team was managing the difference between the build-up of the part in the simulation and the build-up in reality. In the real procedure the process involves melting about 20 micrometers of powder layer by layer, but to simulate every layer would take an unacceptable amount of time.
‘So we took 50 to 100 layers at a time, but if you do that, you have to think about what the thermal behaviour will do in reality compared to the simulation,’ explains Wolf.
Open source simulation tool enables “first time right” manufacturing
TUM carried out numerous investigations of how the thermal curves evolve in these layer compounds. They then compared the results with how the temperature gradients are in reality to come up with the necessary algorithms for the simulation software which was developed from scratch on a Linux open source platform.
The resulting software, says Wolf, ‘will be openly available’ and enables designers, even without specialist knowledge of additive manufacturing design prerequisites, to design ‘finished components’.
The software calculates and compensates for the deformities of the part, ensuring that the printed part turns out correct without wastage. This will be particularly beneficial to SMEs in Europe’s aerospace ecosystem says Wolf, ‘because right now it's really hard for small companies to get involved with AM because there is a steep learning curve.’
All of this adds up to a notable accomplishment within Clean Sky, because AM is an enabling technology that positions European industry advantageously in the fiercely contested aero-engine sector.
‘Our ultimate aim is to have rotating parts produced using additive manufacturing,’ says Roland Schmier, Clean Sky topic manager at MTU Aero Engines.
The strategy is to start off producing the less critical parts of the engine using AM technology – parts such as the ‘borescope eye,’ a component that allows a borescope to be inserted into the engine for maintenance inspections. As the science of AM matures, the ambition is to manufacture much more complex parts, such as turbine blades:
‘A turbine blade has a lot of internal channels for cooling the blade, and the manufacture of such a blade in a casting process costs an enormous amount of money, labour time and material,’ explains Schmier, adding that the software created in Clean Sky's ASCENT AM ‘increases the predictability of part fidelity.’
And, looking at the much longer term potential of additive manufacturing, industry could eventually get to a point in the future where, ‘if there's an aircraft on the ground and it needs a part, you could just print it when you need it.’
Of course, we're not there yet. But, says Schmier, ‘this is the longer term plan,’ and is the reason why industry – not only MTU but manufacturing industry as a whole – is investing into research associated with producing high quality parts through additive manufacturing.
Clean Sky's Rosario Trillo Rivas reports that ‘the simulation tool is already in use, so in terms of exploitation, it has been very successful – there's already use of a developed technology.’
‘The simulation tool's enablement of “first time right” manufacturing means savings in time, materials and manufacturing processes. These indirectly translate into CO2 reductions which will contribute to achieving the ACARE goals.’