Prof Dermot Brabazon explains additive manufacturing, and how data processing and M2M communication could revolutionise the way complex high-performance products are made.
There are many commercialised additive manufacturing (AM) technologies (commonly called 3D printing, or 3DP), most of which work under the same fundamental principles.
To start, the product design digital data (CAD file) is processed and oriented in an optimal build position. This data is then sent to the AM machine, where it is numerically sliced into thin layers. The AM machine fabricates each two-dimensional cross-section and concurrently bonds it to the previous layer. A complete part is thereby built by stacking layer upon layer until the component is completed.
What is additive manufacturing?
Chuck Hull, co-founder of 3D Systems, is considered the father of additive manufacturing.
In 1984, Hull developed a process known as stereolithography. In this process, a platform is lowered into resin, such that the surface of the platform is a layer-thickness (eg 0.1mm) below the surface of the resin. A highly focused UV laser then traces the boundaries, and fills in a two-dimensional cross section, of a model, solidifying the resin wherever it touches. Once a layer is complete, the platform descends another layer thickness, resin flows over the first layer, and the next layer is built.
This process continues until the model is complete, at which point the platform rises out of the vat and the excess resin is drained. The model is then removed from the platform, washed and placed in a UV oven for a final curing.
Other polymer-based AM technologies commonly used nowadays included selective laser sintering (SLS), fused deposition modelling (FDM) and polymer ink-jetting.
3D printing metals
Metal additive manufacturing technologies represent the largest growth area for AM and include selective laser melting (SLM), laser metal deposition (LMD), and electron beam melting (EMB).
In SLM, an electronically controlled galvanometer is used to move a fibre laser beam according to the 2D cross-sectional area of the part, melting and solidifying the metal particles together. The platform is then lowered, a fresh powder layer is spread across the build area and the process is repeated to produce the full 3D part.
The EMB process is very similar, but uses an electron beam to melt and solidify the metal powder rather than a laser. This process is operated in vacuum and the powder bed chamber is typically at a higher temperature than in SLM, which results in lower final thermal stress and higher part production rates.
In the LMD process, a laser beam is used to melt and solidify metal powder which is deposited by gas flow, or a feed metal wire, to the focal point of the laser beam. While this process is slower than SLM for mass production, it does allow for specific materials to be deposited at different positions, including the layering of different materials to enable a functionally graded coating.
In the 1940s, numerical controlled (NC) machine tools were developed for control of production processes, while, from the 1950s, computer numerical controlled (CNC) machines were developed. This phase of computer-controlled production is known as the third industrial revolution. Currently, the fourth industrial revolution (industry 4.0), also called smart manufacturing, is underway.
The product development landscape has evolved tremendously over the last 20 years. Product designs have become much more complex in both their shape and functionality. This is driven by the requirements of increased personalisation and more optimised product functionality.
In tandem with these developments, the need of reduced time-to-market has increased as new designs and product offerings are required to remain competitive and take advantage of the latest technology developments. In order to automate and integrate AM in an optimal manner, there is currently a strong demand and increased requirement for part and process data capture, analysis, adjustment and control.
M2M for AM
For industry 4.0 implementation, cyber-physical system implementation is required, in which an individual production machine is digitally connected with its environment and produced products have a ‘digital shadow’ bound to the product, which enables optimised production process parameters and product life cycle functionality.
The digital shadow data allows for horizontal connectivity between upstream and downstream processing. Considering additive manufacturing, upstream information includes powder type, morphology and fluidity. Downstream data includes measurement of porosity and surface roughness.
Data should be securely digitally connected with the part being produced, and flow with the part such that processing step parameters – such as laser power, process speed, polishing and cleaning – can be informed. The method of implementation of the digital shadow needs to allow for complete and secure traceability of component characteristics throughout the product lifetime.
‘Data should be securely digitally connected with the part being produced, and flow with it’
Some difficulties with the implementation of industry 4.0 are the handling of the large amounts of data generated and collected; the selection of which data is most appropriate for specific tasks; and the implementation of the use of this data for machine closed loop control, predictive maintenance, process step parameter alteration and life-cycle quality, as well as resource optimisation. Additive manufacturing, however, clearly helps to compress time-to-market of products and is, therefore, is a competitiveness-enhancing technology.
Fast prototyping and testing can be implemented and – importantly, to an increasing extent in recent times – rapid serial production of small and high value-add components can be produced. Application areas for additive manufacturing are diverse, and include aerospace and automotive components, injection moulding tooling, medical implants and tools, jewellery, food and pharmaceuticals.
AM remains a rapidly developing area in which significant commercial gains are to be made from the relevant industry sectors. Development of machine-to-machine (M2M) communication for AM is of critical importance in order to be able to most effectively harness the capabilities of AM by the implementation of more effective closed loop control, process data analytics and secure file transfer, as well as robust upstream and downstream process integration.
Prof Dermot Brabazon holds a degree in mechanical engineering and a PhD in materials science. He worked with Materials Ireland, a state-funded materials science research centre, later joining the School of Mechanical and Manufacturing Engineering at Dublin City University (DCU) where he now holds a professorship. He was conferred with the President’s Award for Research in 2009 and is currently director for the Advanced Processing Technology Research Centre at DCU.