Part Specific Process Optimization in SLM

Informations projet

PASSPORT

N° de convention de subvention: 785562

DOI

10.3030/785562

Projet clôturé

Date de signature de la CE 25 Janvier 2018

Date de début 1 Janvier 2018

Date de fin 30 Juin 2019

Financé au titre de

• SOCIETAL CHALLENGES - Smart, Green And Integrated Transport

Coût total

€ 599 860,00

Contribution de l’UE

€ 599 860,00

Coordonné par

TWI LIMITED
United Kingdom

Periodic Reporting for period 1 - PASSPORT (Part Specific Process Optimization in SLM)

Période du rapport: 2018-01-01 au 2019-06-30

Résumé du contexte et des objectifs généraux du projet

Metal Additive Manufacturing (AM) technology has developed rapidly in the last decade and has demonstrated significant potential to reduce the costs and improve the quality and efficiency of aerospace components. This can be realised through improved design freedom and light-weighting via topology optimisation, improved buy-to-fly ratios, and a reduction of tooling cost – all of which have a demonstrable impact on the carbon footprint and waste in manufacture.

Aluminium alloy AlSi10Mg is highly demanded for many applications in aerospace, automotive, micro-electronics and heat exchanger industries. AlSi10Mg has also seen increasing adoption and application as a powder metallurgy alloy for AM. However, there are many factors that affect the final quality of SLM parts. The complex geometry of topology optimised parts – designed to take advantage of the flexibility of SLM processing – means that almost every location in the build sees a unique thermal history. As thermal history controls part quality, there is uncertainty in the performance of and lack of homogeneity within AlSi10Mg AM components.

As the European AM industry and design optimisation applications are growing, lead time and financial costs associated with optimising process parameters to high-quality AlSi10Mg parts with complex geometry is a significant barrier to widespread adoption.

PASSPORT sought to remove this barrier and advance the state-of-the-art. Specifically, the PASSPORT project aimed to:

•Undertake a detailed characterisation of AlSi10Mg SLM parts in- and post-process with a unique laboratory setup.
•Employ advanced, nonlinear process simulations to understand the relationship between different scan strategies, transient/spatially-varying process parameters, and part attributes.
•Develop state-of-the-art, optimised process parameters that vary with local part topology and geometry characteristics to ensure homogeneous mechanical properties.
•Produce a bespoke, stand-alone process parameter selection software solution for AlSi10Mg SLM parts that can communicate with multiple vendors’ SLM machines.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

"An extensive work programme involving advanced numerical modelling, comprehensive experimental studies, and software development was undertaken.

The first activities focused on the setup of the experimental facilities. The activities at the start of the project involved the development of a bespoke thermocouple substrate and the procurement and characterisation of the feedstock for the project. A novel thermocouple substrate was designed. Initial experimental activities focused on the production of reference cube samples (see Image 01). A finite element manufacturing process simulation of laser powder bed fusion was developed. The model was calibrated and validated against measurements collected using the thermocouple substrate during manufacturing of reference samples. A best practice guide on finite element modelling of laser powder bed fusion was also produced (see Image 03).

The next activity was to undertake a review of topology optimised parts. This identified four ""building blocks"" characteristic of such parts: an X-shape, a Pi-shape, Cylinders and Solid Members. A comprehensive full-factorial, design-of-experiments study was performed. Over 500 samples were built and thoroughly characterised using optical microscopy, hardness measurements, X-ray CT scanning and surface roughness measurements. An example of some of the samples are shown in Image 02.

To store the results, Granta Design developed the PASSPORT database and software. A quality metric was defined to enable this link between physical test data, optimal processing parameters, and features of the as-built parts. Once the PASSPORT Database had been developed and populated, the final activity was to create a way to ""break down"" a topology optimised part into the fundamental building blocks. This was done by creating a ""classification algorithm"" wrapped into a Matlab Graphical User Interface for ease-of-use. The Matlab PASSPORT GUI was also connected to the Granta Design PASSPORT Database.

The PASSPORT Software allows an end-user to import an STL file and then the software breaks the STL file down into individual building blocks, associating each block with optimised process parameters. This then theoretically enables a user to use part-specific process parameters and achieve improved homogeneity in the quality and propertis of an AM part. This is illustrated in Image 04. This was then demonstrated on a realistic component provided by the Topic Manager.

The exploitation and dissemination included:

A publication: Zavala-Arredondo et al, 2019: ‘Use of power factor and specific point energy as design parameters in laser powder-bed-fusion (L-PBF) of AlSi10Mg alloy’, Materials and Design 182.
Seminars including the 2018 TWI Annual AM Symposium; 2018 UK SIMULIA Regional User Meeting where a presentation on best practice modelling of AM processes developed in PASSPORT were presented; 2019 AILU Conference on Laser Additive Manufacturing where the PASSPORT was described; STEM Outreach Event with local students about digital technologies as a career using PASSPORT as a case study; a keynote at the UK National NAFEMS Conference on the validation of AM modelling and best practice guidelines


"

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

Significant technical progress beyond the state-of-the-art was achieved. This includes:

•The development of a thermocouple substrate for monitoring of temperatures during processing. The design can be used to enhance the accuracy of models by providing validation data. Improvements in thermal models will enable greater insight into the fundamental physics of LPBF.
•An extensive DoE study undertaken on LPBF AlSi10Mg involving the production and characterisation of over 500 samples.
•The finite element model developed during PASSPORT extended the current state-of-the-art at the beginning of the project. The validation data provided by the project enabled the model to be optimised and a best practice guide to be drafted.
•The feasibility of building a single AM part with multiple, region-specific process parameters optimised to achieve homogeneous properties were developed and demonstrated. This was beyond the current state-of-the-art
The expected and potential impacts of PASSPORT are focused on the aerospace industry, but there are spillover benefits and impacts for other industries. Fundamentally, the PASSPORT project will enable companies to reduce product costs by minimising trial-and-error experimentation to optimise processing conditions for parts. This will save material wastage but also enable improved design freedom. The latter will allow for reduction of component weights (leveraging topology optimisation), leading to reduce environmental impact at the point of manufacture and during service life. Whilst PASSPORT was focused on AlSi10Mg only, the blueprint is wholly applicable to other material systems. By improving quality and integrity via region-specific optimised process parameters, the certification of more advanced, structural AM parts can become more achievable. This technology will thus enhance the competitiveness of the European AM community and lead to follow-on targeted research (in a number of numerical, software and experimental fields) to improve the adoption of AM processing

Example of the reference sample characterisation.

Example sample plan and as-built parts for the DoE study.

PASSPORT Solution Diagram

Example of finite element process model for SLM.