South Korea's Daegun Tech says a 316L stainless-steel impeller made by laser powder bed fusion has completed 18 months of defect-free operation in Unit 1 of the Saeul Nuclear Power Plant. The part operates in the generator circuit-breaker cooling system. Korea Hydro & Nuclear Power and other South Korean partners participated in the development; trade press reported the operating milestone on 8 October 2026, based on Daegun's 22 September announcement.

What changed in the additive design?

A conventional impeller may be assembled from multiple blades and a carrier plate, making the joints separate potential failure locations. Daegun integrated the blades and plate into one component. That does not by itself prove longer life, but it removes some interfaces and allows geometry to be driven by flow and manufacturing requirements rather than assembly access.

The impeller was produced from 316L on a PBF-LB machine developed by Daegun. Laser powder bed fusion consolidates thin powder layers, so quality depends on much more than the alloy designation. Powder history, atmosphere, laser power and scan strategy, orientation, supports, heat treatment, machining and the inspection plan together determine whether the finished part suits its operating environment.

What do 18 months prove—and what remains unknown?

Daegun's public statement and Metal AM identify the same part, alloy, manufacturing process, plant and 18-month operating period. The company reports no defect during that time. Public information does not disclose rotational speed, temperature, fluid chemistry, vibration loading, dimensional results, nondestructive examinations or acceptance limits.

The result is therefore operating experience for one component in one application. It does not qualify every Daegun parameter set, every 316L component or PBF-LB generally for nuclear use. Installation in a working power plant matters, but the evidence is limited by a single publicly described part and the absence of a detailed qualification dossier.

Why this matters to the supply chain

Daegun presents the project as localization of a previously imported component. For rarely ordered, obsolete or long-lead-time parts, additive manufacturing can reduce tooling dependence and create a digital production route. In nuclear service, however, a digital file is not a spare part. A controlled geometry revision, approved feedstock, fixed process window, qualified machine and operators, traceable post-processing and documented inspection are still required.

The scaling question is whether the project can establish a repeatable qualification package. If the next part is built on another machine, from another powder lot or in a revised orientation, the organization must define which change triggers requalification. Eighteen months of operation is useful evidence in that chain, not a substitute for it.

Editorial conclusion

The news is not merely that a printed metal part could be installed, but that it accumulated a year and a half in a real nuclear auxiliary system. That is closer to industrial acceptance than a laboratory coupon. The next meaningful milestone would be publication of inspection evidence, results from multiple parts and longer service, together with the change-control strategy. Until then, this is a promising application reference rather than a blanket approval of the technology.