ORNL 3D-Printed a Pressure Vessel: Getting One Past ASME and a CRN Reviewer Is Another Story

By Kopfkino Consulting Corp. · Published September 4, 2026 · 6 min read

Oak Ridge has wire-arc printed a closed steel pressure vessel on the MedUSA platform. The fabrication demo is the easy part; material data, essential variables, NDE, and CRN registration are the wall.

Oak Ridge National Laboratory and Idaho National Laboratory announced on August 27 that ORNL has produced a closed steel pressure vessel, roughly 3 by 5 feet, using wire-arc additive manufacturing on the MedUSA three-robot platform. The target problem is real and expensive: large forgings sit at the bottleneck of nuclear and chemical processing construction, with lead times that shape project schedules. A printed vessel that bypasses the forging queue is a genuine fabrication achievement. It is also, in code terms, the easy part.

What the Demonstration Actually Proved

The demonstration shows that a wire-arc process can deposit a closed pressure boundary at a useful scale, with coordinated multi-robot motion, in steel. That is a manufacturing result worth celebrating, and the laboratories involved are credible. The forging bottleneck it targets is not hypothetical; large forgings have dictated delivery schedules in nuclear and chemical work for years. What the demonstration does not show is that the vessel can carry a design rating, survive a code calculation, pass examination, or be registered anywhere. Between a printed object and a registered pressure vessel stands the entire qualification apparatus of the ASME Boiler and Pressure Vessel Code, and behind that, in Canada, the CRN process.

The Easy 20 Percent and the Hard 80 Percent

Printing the geometry is the visible 20 percent of the problem. The invisible 80 percent is qualification, and it decomposes into three hard problems that do not care how elegant the printing was. Until all three are solved for a given design, the vessel is a research artifact rather than equipment.

Material Property Data for a Layer-Built Process

Code allowable stresses rest on decades of production data for wrought and forged materials, where the mill process is standardized and the property distributions are known. A layer-built material has properties that depend on deposition path, interpass temperature, build orientation, and post-build heat treatment, and those properties can vary through the wall of a single component. Establishing design allowables for a printed material means generating a defensible property database for the specific process as run, not for the alloy in the abstract.

Essential Variables With No Code Pedigree

Procedure qualification under Section IX assumes a known process with defined essential variables. Wire-arc additive manufacturing sits awkwardly in that framework: the process resembles welding, but the object being qualified is the pressure boundary itself, not a joint in it. Which parameters are essential, how they are controlled and recorded, and what a requalification trigger looks like are questions the code is only beginning to formalize. The Q2 2026 ASME code meetings overview shows where committee attention actually sits: a Section VIII design-by-analysis working group is working through cyclic, creep, and buckling topics, and a substantial Section IX rewrite for tube-to-tubesheet welding is planned. Both matter to industry; neither is an additive manufacturing rule set. Additive progress in the committees is real but deliberate, and near-term code rules that make printed vessels routine should not be assumed in any project schedule.

NDE of Printed Geometry

Volumetric examination methods were developed around wrought and forged geometries and weld joints. A layer-built wall can present anisotropic ultrasonic behavior, surface conditions that complicate conventional techniques, and geometry that conventional scanning plans were never designed for. Demonstrating that an NDE method actually finds the flaws that matter in a printed wall is its own qualification program, and it has to be solved per geometry and per process, not declared in general.

The Canadian Question: Registering a Design Built by a Print Procedure

Suppose the code path matures and a fabricator arrives in Canada with a printed vessel design. The CRN reviewer now faces a design registration package whose material specification is, in effect, a manufacturing procedure. That is a genuinely new posture for design registration, which is built around referenced code editions, listed materials, and established construction methods. A reviewer can only register what can be evaluated against adopted rules, so an early printed-vessel submission should expect extended dialogue: requests for the property database, the process qualification records, the NDE validation, and likely conditions attached to any registration that issues. Our pages on material selection and impact testing for CRN and FEA in support of CRN applications describe the conventional evidence package; a printed design needs all of that plus the process qualification layer.

For a proponent preparing that first conversation with a provincial reviewer, the evidence package assembles in layers: the material property database for the process as run, the procedure qualification records with their essential variables and controls, the NDE validation demonstrating detection capability on the printed geometry, and the current status of any relevant code cases. Each layer exists to answer a question the reviewer is obligated to ask, and assembling them in advance is the difference between a dialogue and a rejection.

The practical guidance for anyone watching this space is to separate enthusiasm from scheduling. Additive pressure equipment will arrive, and the forging bottleneck gives it a strong commercial reason to arrive. But the qualification wall is measured in years of data generation and committee cycles, and Canadian registration will follow code acceptance rather than precede it.

Frequently Asked Questions

Can we order a 3D-printed pressure vessel today?

Not as registered pressure equipment in Canada. The ORNL result is a research demonstration at laboratory scale. A commercial, code-stamped, CRN-registered printed vessel requires material data, procedure qualification, and examination methods that are still being established.

Does ASME Section VIII currently cover additive manufacturing?

Section VIII's mainstream construction rules assume conventional materials and processes. Additive construction is being addressed through committee work and code-case-style pathways rather than through mature book rules. Anyone planning a project should verify the current state of that work directly with the committee record rather than relying on vendor claims.

Will a CRN reviewer accept a printed vessel design?

Expect a case-by-case posture. Reviewers register designs they can evaluate against adopted codes, and a design whose material is defined by a print procedure will face data requests that a conventional vessel does not. Early engagement with the destination province's reviewer is the right move for any proponent.

What should conventional fabricators do about this now?

Watch it, and otherwise continue. The demonstration does not change near-term capacity planning, quoting, or registration strategy. It does signal where forging-constrained sectors are investing, which is useful intelligence for long-range product planning.

Sources

• American Nuclear Society: ORNL and INL collaborate on 3D-printed pressure vessels
• CEI Think Tank: Overview of Q2 2026 ASME Boiler and Pressure Vessel Code meetings

Whether your next vessel is printed, forged, or rolled and welded, the registration package is what makes it installable in Canada. Contact us to talk through the evidence a reviewer will expect.

Topics: Additive Manufacturing, ASME BPVC, Qualification, NDE, CRN