Metal 3D printing currently relies on a single laser beam to melt powder into solid objects, but this project proposes swapping that single laser for an array of dozens or hundreds of lasers working simultaneously. This matters because existing metal 3D printing is too slow and unreliable for mass production. A single laser creates melt instabilities, requires painstaking trial-and-error to tune parameters, and often produces parts with internal flaws or distortion that need costly post-processing. The laser-array approach replaces the problematic "keyholing" melt regime with a stable conduction-limited one, boosting part density to 99.99% and slashing residual stress and distortion. If successful, the system could achieve build rates of 153 kilograms in 24 hours—roughly 20 times faster than current machines—and cut part costs by up to 80%. In-process metrology would detect and correct errors layer by layer, enabling a right-first-time production process. This could transform metal additive manufacturing from a niche tool for aerospace and biomedical prototypes into a viable route for high-volume manufacturing of precision components across multiple industries, quietly reshaping supply chains and reducing material waste.
View original technical description
The early prospects of Additive Manufacturing (AM) technologies promised to provide greater design freedoms, raise productivity levels, minimise material usage, compress supply chains, and enable the producer to attain greater levels of competitiveness by delivering enhanced product capabilities. Metal based LPBF AM systems have developed steadily over the past 20 years and now represent a multibillion-pound global market in machines, materials, and software. They find niche low volume applications in many industrial sectors and somewhat wider applications in aerospace and biomedical sectors. However LPBF AM processes are still slow compared to traditional manufacturing routes and are quite complex. They require precise focusing and manipulation of high energy laser beams over large powder beds in order to consolidate metal powder into a 3-dimensional solid through laser melting. Melting strategies play a significant role in part quality. Single laser beam melting strategies employed in all commercial systems suffer from melt instabilities, low melting efficiencies, and complex scanning strategies to reach high densities. They require a high level of labour-intensive part-specific build parameter refinement and time-consuming post processing operations. Despite the clear attractiveness of this production route, there remain several challenges in terms of build rates, process stability, part accuracy, repeatability, and part cost. In this project we propose to investigate several technology solutions that address these fundamental problems. To improve build rate we will establish a new class of LPBF AM capability by re-configuring the laser powder interaction process away from the current single laser interaction to large scale laser arrays. This approach offers increased melting efficiencies and true power scalability in the multi-kW domain. Since laser arrays are readily scalable, a 20kW system could deliver build rates of 153 kg in 24 hours. This is some 20 times faster than current systems. Our approach could offer world leading performance figures for LPBF AM systems. The use of laser arrays enables the problematic keyholing regime to be replaced with conduction limited regime leading to dramatic increases in process stability and part densities routinely reaching 99.99%. More stable melting regimes with reduced thermal gradients and reduce residual stress, reduce part distortion, and ultimately increase part accuracy. In process metrology will be applied to detect errors in the build layers and enable corrective steps thereby increasing process repeatability and deliver a right-first-time production process. With the combined innovations cited above we estimate that part costs savings up to 80% could be achieved compared to conventional LPBF AM systems.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know