AM System Using TIG Welding and Robot Relies on Flexible Enclosure for Gas Protection

Welding is a natural starting point for additive manufacturing. In welding, metal is added when melted for the sake of the strength it will have when it solidifies. A logical step therefore is to build 3D forms by welding in layers. Indeed, electron beam additive manufacturing originated with a company known for welding systems. Could standard arc welding technology be used for additive manufacturing as well?

Seemingly yes. The UK’s Cranfield University has now been studying this question for years. Researchers at the Welding Engineering Research Centre here developed the process of Wire and Arc Additive Manufacture (WAAM), using gas tungsten arc welding (also known as tungsten inert gas or TIG welding) to build forms in metals ranging from aluminum to nickel alloys. BAe Systems has used the process to build a titanium alloy wing spar 1.2 meters long.

With the aim of eventually making this process easy to apply in production, Cranfield researchers have taken the further step of integrating WAAM with a robot—meaning a process using standard welding could be applied with a standard automation device. However, the challenge in this is oxygen contamination. For some alloys, the welding torch’s inert gas shroud is sufficient protection against contamination, but titanium is particularly prone to reaction with residual oxygen leading to surface oxidation. Using WAAM with a robot therefore required some way to guard against this.

The Cranfield team solved this problem by working with welding technology company Huntingdon Fusion Techniques (HFT). Flexible enclosures developed by this company accommodate the robot and provide for inert gas protection throughout the AM build process. An initial small enclosure from the company contained just the welding head and was used for testing, while the enclosure seen above contains a volume of 27 cubic meters. Read much more about the enclosure technology enabling this robotic additive manufacturing in this paper on HFT’s site.