The AMRC has joined forces with a UK tech start-up and a global engineering giant in a transatlantic bid to disrupt the $3 billion metal additive manufacture (AM) market by using Artificial Intelligence and robotics to carry out the labour intensive removal of the support structures essential in AM.

The research team behind Project SALSA (Separation of Additive-Layer Supports by Automation) is developing an agile and dexterous tool that will use robotics and machine vision to detect and remove supports from metal-additive parts, to ramp up productivity.

The Innovate UK-funded programme is part of a UK-Canada collaboration for research and development projects and follows a recent visit to Canada by the AMRC’s Head of Digital, Professor Rab Scott, as part of the High Value Manufacturing Catapult research mission to the country.

Led by Bristol based tech start-up Additive Automations, SALSA involves partners Renishaw Canada, the National Research Council Canada and the University of Sheffield’s Advanced Manufacturing Research Centre.

Ben Fisher, senior project engineer for AMRC’s Integrated Manufacturing Group (IMG), says that when a part is created using AM, it is often necessary to add support structures to build the piece successfully. About 95 per cent of parts produced in the metal additive manufacturing industry involve this time-consuming activity, often involving the use of chisels and grinders, making process control very difficult. 

SALSA will build on the transatlantic consortium’s wealth of AM and AI expertise to develop and test a small-scale prototype that automates support removal and transfer this to an operational prototype.

This project will develop a tool that can separate additive-layer supports using robotics and machine vision, freeing up skilled technicians for higher-value tasks.

“Support removal is a necessity for metal additive manufacturing,” explains Ben. “The aim of this project is to reduce the time it takes to do that. The current method is manual, very time consuming and can be a hazardous and toxic process for workers. This project will develop a tool that can separate additive-layer supports using robotics and machine vision, freeing up skilled technicians for higher-value tasks.”

Metal AM offers enormous potential as a method for manufacture: it can be cheaper, takes less time and offers unlimited design freedom. A leading aircraft engine supplier has already produced more than 30,000 fuel nozzles using metal AM for its new engine, but the big challenge is to enable small companies to access the same benefits.

To do this, the complexities, cost and risks associated with the technology need to be reduced to give SME manufacturers the confidence that metal AM is ready for full-scale production and capable of saving them time and money.

Post-processing can account for between 35 per cent and 65 per cent of the part costs and take a skilled technician several working days to remove the supports on a single part. Mistakes can also be costly - one instance reported was £17k on a single component – and in some cases removing the supports and finishing can be more expensive than the actual AM build process.

The project’s goal is to develop a tool that can remove all types of supports on all types of metal AM materials and ensure process time and cost is at least half of that offered by current methods. It also needs to make certain the process does not use or generate toxic substances and is a safe environment for people.

Project SALSA’s objective is to develop a post-processing machine that gives the UK industry the confidence, control and economics it needs to adopt this technology in full-scale production.

A report by Robert Bush, company founder of the consortium’s lead partner Additive Automations, says a lab-scale proof-of-concept for automating support removal has already proven that time and cost reductions up to 66% and 62% respectively are possible, with huge productivity gains estimated - 98% uptime, compared to 21% manually.

The consortium will explore two streams of work: the design and build of the end effector for removing the support material, being undertaken by Additive Automations, and the artificial intelligence and machine vision system to detect the removable support material, which will be developed by IMG.

Ben said: “We are deploying our expertise in collaborative robotics for this project; we do a lot of complex integration work and we have done a number of cobot projects. The work that IMG will be doing in this programme is the integration of the end effectors onto a cobot such that an operator can teach the routines to the part for the robot to remove the material. It’s about making these systems talk to each other.”

Ben added: “For additive to become competitive as manufacturing method, you need to remove or reduce the non-value added time, in this case the process for removing the support material. You need to have that support material to build the part, but there is a lot of post-processing time with AM parts, so it is shredding that time and automating that process of the removal.

“As AM is used in a huge range of applications – aerospace, automotive, energy, industrial, medical –  it means the opportunities and the potential impact of this project is massive. It has the power to lower the barriers to adoption.”

Additive Automations’ founder Robert Bush says SALSA has the potential to be a major disruptor.

“The collaborative robotics market size is $1bn and growing at 50 per cent. Artificial Intelligence in manufacturing market size is $1bn and growing at 25 per cent. Metal additive manufacturing is an industry worth $3bn and growing at 80 per cent, one of the fastest growing areas in engineering,” says Robert.

By adopting technologies that form part of the fourth industrial revolution, such as metal AM, machine learning and cobots, Britain will ensure a competitive, stable economy with sustainable innovation and jobs generation for decades to come.

“The benefits of additive manufacturing read like an engineer’s wish list; the ability to produce unitised parts with complete design freedom, novel geometries and graded materials at both reduced cost (no tooling) and reduced waste (near net shape solution).

“Moreover this can be achieved with a high-degree of customisation at the point-of-use (local manufacture) with leads times far less than current methods. The UK has world class additive manufacture machine manufacturing capability, a well-established national centre and is among the world leaders in research and innovation within high-performance applications such as aerospace and medical.

“However, the UK is seeing significantly slower adoption and use of additive manufacture than Germany, China, USA and South Korea. UK manufacturing companies view it currently as having too many barriers to entry for full-scale production.

“Project SALSA’s objective is to develop a post-processing machine that gives the UK industry the confidence, control and economics it needs to adopt this technology in full-scale production. SALSA will help make metal additive manufacturing cost effective, not just in the thousands of parts but into the tens of thousands.

“Since 1970 the UK’s manufacturing sector has had the most severe decline in output as a percentage of GDP. In 2015 manufacturing output was ten per cent of GDP, down from 28 per cent in 1970. To create wealth and build an economy something needs to be created, and manufacturing is the wealth generating industry.

“By adopting technologies that form part of the fourth industrial revolution, such as metal AM, machine learning and cobots, Britain will ensure a competitive, stable economy with sustainable innovation and jobs generation for decades to come.”