AMS Speaker Spotlight: The Future of WAAM 3D Printing

Filomeno Martina CEO and co-founder, WAAM3D will be participating in Additive Manufacturing Strategies 2022, Panel 1: The future of DED and WAAM. Additive Manufacturing has been around for quite a few years now, with powder-bed technologies becoming relatively widespread. Here Martina discusses the role wire-arc additive manufacturing (WAAM) is now playing in industry and its potential to change the future of medium to larger component design. 

The WAAM process saw its first introduction a long time ago. In fact, a patent was granted almost a Century ago – in 1925 – for the manual deposition of superimposed layers of metal by arc welding.  Due to the type of feedstock used (wire) and the size of the melt pool generated by its heat source (an electric arc), WAAM can print very large components in a cost effective and timely manner; it also eliminates the storage, contamination and safety problems associated with powder-based systems. So why has WAAM yet to see its full potential realised?

Closeup of WAAM3D’s End Effector, part of their RoboWAAM system.

Being able to print parts by WAAM, with the required level of integrity, requires mastering the interaction of the many variables often involved in different steps of the engineer’s workflow. This is no different from any other process within AM, which truly is the epitome of multidisciplinarity – with complex interactions of material science, physics, mechanical engineering, software and hardware frameworks and, indeed, human factors. The scale at which these complex phenomena occur in WAAM makes them even more evident. Therefore, the technology needs to mature in the direction of 1) the integration of the tools needed to go from a part’s concept to a successful print, 2) the automation and control of the underlaying processes, and 3) the improvement in print speeds to being able to tackle even more components in a timely manner.

With this in mind, we have been developing WAAM-specific software that covers the whole digital chain: life cycle assessment and business case analysis, tool-path planning, process parameters calculation, trajectory simulation and collision detection, residual stress prediction, machine control, data analysis, and health and safety management. Together with this, we have developed our hardware backbone that comprises the motion system, power source, a suite of innovative sensors, the fume management system, the automatic wire reload, and the whole machine enclosure – everything managed centrally from the operator’s station. This really is a massive step forward in terms of robustness and usability, compared to the obsolete approach of relying on a standard welding cell used for WAAM purposes. The intimate level of integration means that, at any point in time, anyone in the organisation can know what the machine’s performances are, with any undesired deviation from the ideal behaviour being flagged immediately.

Moreover, just as the industry needs to move away from re-branded welding cells, it also needs to develop new WAAM processes and dedicated ancillary solutions. It is of no surprise that we have already hit the productivity limits of welding tools – these were never meant to be used in an AM context. For high-value material such as titanium, we must be able to increase build rates and control its geometry – specifically of layer height, wall width and distortion – at the same time. For low-value materials such as iron-based alloys, increasing the build rates will be key to increased market adoption. Hence our efforts on developing radically new WAAM variants that have already demonstrated such capability and that we are painstakingly industrialising. Finally, maximising the flexibility of robots, there will be more and more attention on the integration of helpful processes such as cold-work (to achieve forged-like mechanical properties), metrology and inspection (to sign-off parts immediately), and machining (to increase the geometrical freedom, and potentially repair defects in situ).

A case in point:

Let’s look at putting WAAM into practice. Once we have received the CAD from the customer, within a few hours we will have been able to select wire and substrate material, chosen where to place the substrate feedstock, built the WAAM pre-form, calculated the printing cost and assessed the part’s environmental performances. The tool-path plan can then be produced in less than 10 minutes, with process parameters calculated automatically. We can then run a simulation of the printing process in our virtual environment and fix any alignment issues. As soon as the machine is available, we can clamp the starting feedstock in our RoboWAAM machine using existing tooling and execute the print. Data coming from our 15 sensors logs progress throughout and compares it to predicted and expected values, including geometrical ones (thanks to the in-process 3D reconstruction happening whilst the material is being deposited). If any non-conformity arises this is then flagged immediately. The part is then signed off and sent to machining.

First WAAM full scale prototype of a titanium pressure vessel to be used in the future of manned missions – by a team comprising of Thales Alenia Space, WAAM3D, Cranfield University and Glenalmond Technologies.

Future advancements:

Whilst serial production will initially rely on frozen configurations (locked parameters, trajectories, and flows), for spare parts the user needs to have immediate absolute certainty of their integrity. For time pressured projects or those with a batch size of one, extensive mechanical testing campaigns make little financial and operational sense.  However, as more work is done in serial production and repair/remanufacture of parts, increased confidence in the operational capabilities of WAAM and data to trust will help address these issues. This will be no doubt assisted by in-process quality control, founded on on-line inspection, and on qualification frameworks based on key variables such as material-based thermal fields, as opposed to part-specific parameters configuration. This incredibly flexible, powerful approach will also overspill back to serial production. The automation and the incredible number of sensors of these machines will be instrumental in building up and renewing confidence in the whole technology.

At the same time, the focus will shift to the bigger picture – on how to make all the WAAM cells part of a global factory and manufacturing environment, to enable a truly decentralised, on-demand manufacturing solution. This is beyond Industry 4.0 – not just the one smart factory, but a globalised smart network made up of many local smart factories.

In summary, on paper the case for applying WAAM to metal parts is strong, but still very much limited by the usability of the systems and overall process performance. Addressing these two issues is key to enabling firstly users to just use (as opposed to develop), and secondly organisations to look at a much larger fraction of their metal parts portfolio.