NuScale and Doosan Move From Nuclear Materials to First-Article Fabrication — A Qualifying Step, Not Yet an Order Book

On 29 October 2026, at Doosan Enerbility's headquarters in Changwon, NuScale Power signed a contract covering manufacture of the upper reactor coolant system (URCS) for the NuScale Power Module (NPM). World Nuclear News reported the signing the following day under a headline describing the item as an "upper reactor vessel"; NuScale's own release is more precise, describing the contract as moving the collaboration "from preparation into production of the First Article for the upper reactor coolant system."

That distinction is the whole story. This is not a series production order. It is the point at which a certified design becomes a manufacturable object — drawings, weld procedures, cladding schedules, heat-treatment cycles and non-destructive examination (NDE) acceptance criteria get fixed on real hardware. For the metals and heavy fabrication supply chain, that is a more consequential event than the headline tonnage suggests, and a less consequential one than a multi-module order would be.

What has actually changed in the relationship

The NuScale–Doosan partnership dates to 2019. Its trajectory is visible in three steps:

Neither the WNN report nor NuScale's release discloses contract value, duration, module count or delivery date. Any figure for these would be invented. What the sources do establish is direction of travel: Doosan's own framing is a shift "from material preparation to the fabrication of reactor components for commercial deployment."

Why the URCS is a qualification milestone, not a tonnage story

The URCS forms the upper section of the reactor pressure vessel and contains reactor coolant at high temperature and pressure. In pressurised water reactor practice, components of this class are heavy-section low-alloy ferritic forgings and plate — typically SA-508 or SA-533 Mn-Mo-Ni grades — with austenitic stainless or nickel-alloy cladding on wetted surfaces, plus SA-182/SA-336 forgings and SA-193/SA-194 bolting at nozzles and flanges. That material specification is my metallurgical inference from standard PWR practice for the component type, not an assertion in the sources.

The commercial logic follows from the metallurgy. First Article production is where forging, cladding, heat treatment, machining and NDE procedures are proven, and where the fabricator's ASME Section III N-stamp scope and nuclear QA programme are validated against a specific geometry. In heavy nuclear fabrication, that qualification consumes substantial forging and test-piece tonnage for comparatively little delivered product. The value is in the approved procedure and the qualified line, not in the kilograms shipped.

One line in NuScale's release deserves attention beyond the ceremony: the company says refinements made through the collaboration "improve manufacturability while preserving the performance and safety characteristics that earned NRC approval." Design-for-manufacture changes to a certified design normally require some form of regulatory review — a topical report, exemption or equivalent — and the mechanism is not disclosed in the material reviewed here. Readers should treat that as an open item rather than a settled one.

The capacity question behind Doosan's "significant manufacturing capacity" language

The NPM is a 77 MWe (250 MWt) integral PWR, the first SMR design to receive US NRC approval (September 2020), offered in VOYGR configurations up to the 12-module, 924 MWe plant. Because each module is a fraction of a large PWR, the bottleneck for a real order book shifts away from forging press capacity toward repeatable welding, cladding, machining and inspection throughput — and toward qualifying enough vendors to build hundreds of near-identical units.

A rough illustration of the scale (my arithmetic, purely illustrative; no source gives takt times or unit weights): 12 URCS units per VOYGR-12 plant means a 10-plant programme needs roughly 120 units. At a hypothetical 3–6 months per unit on a single qualified line, one line delivers about 2–4 units a year — implying 30–60 line-years for 120 units. That is the arithmetic behind Doosan's statement that "significant manufacturing capacity will be required" and its stated intent to strengthen its supply chain for follow-on orders.

Commercial and geographical context

Two things make this contract more than an engineering formality.

First, NuScale's order book is thin and its flagship US reference project is gone: in November 2023 the company and Utah Associated Municipal Power Systems mutually terminated the CFPP, a six-module, 462 MWe project pencilled in for 2029. The March 2023 long-lead upper RPV order was placed for CFPP; NuScale intends to transfer those modules to its next customer. The first-of-a-kind deployment is now slated for Doicești, Romania, on a former coal plant site. A First Article contract at Changwon is therefore also a signal to Romanian and other prospective customers that hardware is moving.

Second, Changwon is one of very few non-US heavy nuclear fabrication complexes able to serve Western SMR programmes. Each step NuScale takes there deepens its reliance on Korean capability — a sensible capacity play, but also a single-point dependency worth watching as competing SMR developers pursue the same constrained pool of qualified heavy fabricators. That assessment is my judgement, not a sourced statistic.

What to watch

• Whether the First Article contract is followed by a multi-module URCS and RPV production order tied to Doicești, and on what schedule.
• The regulatory route for any design-for-manufacture refinements, and whether it affects the certified NPM design.
• Whether Doosan's supply-chain strengthening extends to European forging or cladding vendors, which would broaden the SMR metals supply chain beyond Korea.
• Parallel qualification of other pressure-boundary components; NuScale's release index has referenced work with MillenniTEK on a "first-of-a-kind reactor safety component," which I did not read and cannot verify here.

For steel and heavy-forging suppliers, the message is that SMR demand will arrive first as qualification work — procedure development, test pieces, N-stamp scope — and only later as recurring tonnage. Fabricators positioned for the qualification phase, rather than the tonnage phase, are the ones likely to convert it into an order book.