TL;DR:
- ORNL and ZEISS signed a licensing agreement for advanced 3D-printed component inspections using CT scanning and Simurgh algorithm.
- Collaboration aims to expedite inspections, cut costs, and enhance quality.
- Five-year research initiative supported by DOE’s Advanced Materials and Manufacturing Technologies Office.
- Focus on using CT scanners and analytics to detect defects in 3D-printed parts during manufacturing.
- Challenges in 3D printing inspection due to unique characteristics of printed parts.
- ZEISS’s role as a leader in multidimensional metrology solutions tailored for additive manufacturing.
- Simurgh framework accelerates CT scanning and analysis through deep learning.
- Successful use cases include certifying nuclear fuel assembly brackets and testing 3D-printed turbine blades.
- Exploration of expanding CT scanning applications to microelectronics and batteries.
- Aspiration to make CT scanning a standard procedure in production lines for comprehensive component scrutiny.
Main AI News:
In a groundbreaking stride towards enhancing the quality control of 3D-printed parts, a strategic licensing agreement has been solidified between the prestigious Oak Ridge National Laboratory (ORNL) and renowned research collaborator ZEISS. This pioneering partnership aims to harness the power of industrial X-ray computed tomography (CT) and ORNL’s innovative machine learning algorithm, Simurgh, to facilitate swift and thorough evaluations of intricate 3D-printed components. This collaboration is projected to not only slash inspection durations and costs by over tenfold but also to elevate the overall quality standards.
The dynamic licensing agreement is an integral facet of an overarching five-year research consortium between ORNL and ZEISS, bolstered by the backing of the Department of Energy’s Advanced Materials and Manufacturing Technologies Office, alongside a notable Technology Commercialization Fund award. The crux of this multifaceted research initiative revolves around the adept integration of CT scanning and advanced measurement mechanisms, proficiently delving into the internal dimensions of 3D-printed parts. This in-depth exploration is geared toward detecting potential imperfections, such as fractures, during the intricate manufacturing process.
A crucial hurdle in the wider assimilation of 3D printing is ensuring the soundness of parts, safeguarding them from concealed defects that could undermine their operational efficiency. Unlike conventional manufacturing methodologies, where decades of experience inform expectations, the innovative realm of 3D printing necessitates a novel approach. This entails employing advanced characterization techniques, specifically CT scanning, to unravel the distinct attributes hidden within these items.
CT scanning, a prevailing non-destructive methodology, finds widespread application across diverse industries to ascertain component quality. However, this technique has conventionally entailed a substantial expenditure of time and resources. The pivotal challenge lies in leveraging the realm of physics and technology to expedite CT scanning, effectively broadening its industrial utility. According to Amir Ziabari, a prominent researcher at ORNL, the quest is to merge scientific acumen with technological innovation to render CT scanning more accessible for industry-wide adoption.
This transformative research is currently unfolding at the Manufacturing Demonstration Facility of the Department of Energy’s Oak Ridge National Laboratory. Home to the MDF Consortium, this collaborative endeavor strives to propel the vanguard of U.S. manufacturing technology forward under the adept guidance of the Advanced Manufacturing and Materials Technologies Office.
A key player in this monumental collaboration, ZEISS Industrial Quality Solutions, stands at the forefront of multidimensional metrology solutions. Their expansive repertoire encompasses coordinate measuring machines, optical and multisensor systems, as well as cutting-edge 3D X-ray metrology and microscopy systems for industrial quality assurance. By tailoring their solutions to additive manufacturing, covering aspects from process validation to automated defect analysis, ZEISS fervently champions consistency and replicability in quality standards.
Paul Brackman, Additive Manufacturing Manager at ZEISS, acknowledges the enduring partnership between ZEISS and ORNL, which has catalyzed the development of ingenious solutions for automated analysis and qualification. The focus now pivots towards optimizing process development and qualification in additive manufacturing, ushering in widespread adoption and a paradigm shift from prototyping to full-scale production.
Within the characterization laboratory at MDF, ZEISS boasts a remarkable array of equipment, including industrial computed tomography systems and scanning electron microscopes. These tools serve to minutely scrutinize the subtlest flaws concealed within 3D-printed components. However, the scan itself constitutes just the inaugural step. Extracting insights mandates intricate analytics to pinpoint flaw locations. This endeavor entails considerable computational prowess, which translates to both time and financial resources. Enter the Simurgh framework, a revolutionary system underpinned by deep learning. This innovation dramatically accelerates scanning and analysis processes, thereby engendering heightened precision.
Such meticulous characterization holds paramount significance for high-value components destined for extreme operating environments, where failure is inconceivable. ORNL leveraged CT scanning to certify the efficacy of nuclear fuel assembly brackets inserted into the Browns Ferry Nuclear Plant in Alabama—a historic first for 3D-printed parts within a nuclear reactor. This technique was equally pivotal in the creation of 3D-printed turbine blades, rigorously tested within a land-based engine operating at speeds surpassing 12,000 revolutions per minute and enduring temperatures soaring up to 800 degrees Celsius. These blades exhibited remarkable resilience, impeccably fulfilling expectations amidst the rigorous operational milieu.
The scope of CT scanning isn’t confined solely to the present domains; ORNL is diligently exploring its expansion into industries such as microelectronics and batteries, where its applications remain largely untapped. This exploratory trajectory holds the potential to usher transformative breakthroughs in fields pivotal for the impending clean energy transition.
Notably, current CT scanning technology imposes restrictions on scanned object dimensions, shapes, and materials. While it is justifiable for scrutinizing a limited number of high-value components or validating a subset from a larger production batch, the collaborative efforts of ORNL and ZEISS strive to transcend these constraints. The objective envisions rendering CT scanning as commonplace as visual inspections on assembly lines of yesteryears.
Amir Ziabari outlines an ambitious aspiration: to accelerate CT scanning to an extent where it becomes seamlessly integrated into production lines, enabling the swift and reliable scrutiny of every single component. Achieving this ambitious milestone could undoubtedly revolutionize the 3D printing landscape, unearthing its unparalleled potential.
Conclusion:
This collaborative endeavor between ZEISS and ORNL marks a significant leap in the inspection of 3D-printed components, utilizing advanced CT scanning and machine learning to enhance quality while drastically reducing inspection time and costs. As industries explore wider applications and strive for cleaner energy solutions, the accelerated and accurate scrutiny of critical parts can potentially reshape manufacturing practices and standards.
