Figure 1: SonicSMR concept diagram
IN MAY 2020, LASER ADDITIVE Solutions (LAS) secured grant funding of £826,633 from the UK’s Department for Business, Energy and Industrial Strategy (BEIS). It was part of the £505m Energy Innovation Programme – ‘Advanced Manufacturing and Materials Phase 2B’ competition, to lead a collaborative R&D project entitled SonicSMR. The 12-month project, which finished in June 2021, explored intelligent additive manufacturing methods for producing small modular reactor (SMR) components.
Alongside consortium partners — The University of Sheffield’s Nuclear AMRC, Brunel University’s Brunel Innovation Centre (BIC) and two specialist SMEs, IVY-TECH and Taraz Metrology — the project aimed to address the challenges of using metal additive manufacturing (AM) for SMR parts. It focused on issues such as interlayer adhesion and micro/macro-scale porosities, by developing and proving a method of in-line monitoring and inspection of metal AM components, utilising LAS’s expertise in laser AM equipment integration and component manufacture, with the specialist inputs of the other partners.
Using supplementary technologies such as power ultrasonics, optical process monitoring and artificial- intelligence-based automated defect recognition, project SonicSMR looked to develop a large laser-based AM system suitable for the manufacture of defect-free complex components.
The project is one of several that formed part of a £40m funding package to kick start the development of next-generation nuclear technologies, as part of the UK’s Nuclear Sector Deal launched in 2018.
Proof of concept
Nuclear AMRC’s role in the project was initially with a proof of concept. This would provide confidence progressing into the project that sonication did have a positive impact on the additive manufacturing process. They also did a design of experiment additively manufacturing small wall shaped parts. This was done with and without ultrasonication, the parts manufactured this way were all cut up then inspected with a Scanning Electron Microscope (SEM) and a Computer Tomography (CT) scanner to see how they performed metallurgically.
With the inspection giving promising results the next step was to produce larger block samples which were manufactured using the best parameter sets that were a combination of: laser power, traversal speed, wire federate, ultrasonication frequency and shielding gas settings. These block samples gave very good results, which were seen by CT scanning and utilising a SEM. There was a significant reduction in defect size and frequency throughout the part, and aesthetically the samples were also better as the sonication had improved the final part geometry.
Nuclear AMRC also hosted trials in November 2020. These combined all partner prototype equipment and tested them together, which allowed the project to progress into its final stages.
Figure 2: Trials at Nuclear AMRC (Photo credit: Nuclear AMRC)
Power ultrasonic system
Part of Brunel Innovation Centre’s role was to develop, optimise and implement the power sonication system for improving the additive manufacturing process. It also developed software and communication protocols to allow easy integration with the process.
The power sonication system consisted of a power amplifier and signal generator which was taken from concept to physical system during the project. As integration with LAS’s direct energy deposition system was key there was constant communication between LAS and BIC. Key factors that came out of the conversations were: ease of operator use, which was taken on board and resulted in an easy-to-use system; robustness, and the final product had a strong industrial feel when in operation; and multiple manufacturing scenarios, which BIC also considered and which gave the system scope for expansion beyond the current aims of the project. For example there can be multiple transducers driven from the unit, allowing larger part manufacture.
Development of AI-enabled in-line monitoring
The growing interest in AM technology and the benefits it offers has led to a real interest in improving the AM process. During this project Taraz and BIC developed an in-line process monitoring system that can scan, then analyse, surface defects between layers.
Taraz has developed a scanning head which is ideal for fitting onto current AM systems, giving them in-line process monitoring capability which is something most AM systems are lacking currently.
This technology makes it possible to improve AM process consistency and make development more cost effective.
BIC’s AI takes data obtained from the Taraz fringe projection system and highlights defects to the operator, who can assess the part and decide a route forward. This is especially useful for larger parts, where scrappage would incur huge costs, or for parts that have to adhere to stringent quality controls.
The most common manufacturing defects in an AM process are cracks, porosity and lack of fusion. Other material characteristics such as the correct microstructure, mechanical strength and corrosion resistance will need to be explored in future AM parts.
System design optimisation
Ivy-Tech’s role in the project was to design and develop a power ultrasonic transducer mounting (PUTM), providing consistent sonication to the AM part. It also provided optimisation recommendations for the full system to LAS based on a customer survey that was distributed at the beginning of the project.
Ivy-Tech used a live detailed requirements capture document followed by a number of design techniques. One example was brainstorming diagrams, separating key requirements such as ‘transducer contact force and technology’ to fulfil certain required functions. Four concepts were generated and assessed further by modelling in computer aided design, taking into account the requirements document. Ivy-Tech was able to produce a robust system meeting all requirements including a new thermal requirement, discovered in the proof of concept stage. The thermal requirement was to keep the transducer at a temperature below 200°C, otherwise it would depolarise, destroying the transducer.
Ivy-Tech’s transducer holder proved to be extremely effective when it was used at the Nuclear AMRC in a suite of trials. Minor modifications were made to ensure robust operation.
Integration and demonstration validation
LAS’s role was to project manage and to integrate all partner-developed equipment into a demonstrator cell.
The steps in achieving this were:
- Collect details about the partner equipment over the course of the project, such as weight, size, electrical and digital inputs and outputs.
- Provide additive manufacturing expertise to partners as they develop their equipment.
- Construct and decorate the SonicSMR cell.
- Work with integration company Olympus Technologies Ltd to install a high accuracy Kuka KR60 L30 HA robot with 3m KL1000 linear track.
- Integrate critical systems — Trumpf TruDisk 3kW laser; Dinse wire feeder; Oerlicon Metco powder feeder — to supply shielding gas.
- Integrate partner equipment with the SonicSMR cell.
- Manufacture nuclear-related demonstration components using the full operational cell and partner equipment.
Overall, the SonicSMR project has been a positive story from a year of COVID-19 restrictions. Using online meetings the consortium has achieved a huge amount and it is a testament to everyone involved.
The SonicSMR project will culminate in a fully operational metal AM demonstrator system, which will be assembled in Doncaster, and will be used to build example-scale SMR components.
Due to the highly adaptable nature of the system application in other industries — such as aerospace and hydrogen fuel generation — is also a possibility.
Author: Bradley Rogers is Project Engineer at Laser Additive Solutions Ltd