MX3D’s DNV AMC3 Approval and the Shift Toward Certified Metal AM

On September 30, 2026, Dutch wire-arc additive manufacturing company MX3D announced that DNV had certified its facility at AMC3, which MX3D describes as DNV’s most stringent additive manufacturing level. According to the announcement, the certification covers the company’s facility, its MetalXL software, and its WAAM processes for critical, high-load structural applications in maritime, defense, and energy markets. A DNV representative quoted in the release describes MX3D as the first company to reach AMC3 for DED-Arc/WAAM technology.

On its own, this is one company’s milestone. Read alongside other announcements from the same week, it is part of a broader shift: metal additive manufacturing is moving from proving that parts can be made to proving, with documents an outside party will accept, that they are made the same way every time.

A certification route outside aerospace

Most public discussion of AM qualification has followed aerospace and medical programs, where large OEMs and regulators define the rules. MX3D’s approval points to a second route. Classification societies such as DNV set requirements for ships, offshore structures, and energy equipment, and their standards are written to be applied across many manufacturers and owners.

DNV’s public page for DNV-ST-B203 lists a November 2025 edition covering additive manufacturing of metallic and polymer parts. It describes requirements for procurement, technology-specific qualification of processes and parts, quality management, and production, testing, and inspection. It also includes an informative appendix on choosing additive manufacturing categories for parts according to their function, and appendices on inspection and test plans for powder bed fusion and directed energy deposition.

That structure matters for the industry. A category system linked to the consequence of failure gives asset owners a common language for deciding where printed parts can be used. A manufacturer approved at the highest category has shown that its organisation, software, and process controls can support that level. Each part still has to be assigned its own category and produced to the matching plan, so the approval opens a door rather than settling every application in advance.

It is also worth noting what the announcement does not say. MX3D does not list specific alloys or part types, and the detailed category definitions sit in the full standard rather than the public summary. The practical scope will become clearer as certified parts enter service.

Large-format arc deposition is a different market

WAAM builds near-net-shape metal parts by depositing welding wire with an electric arc. It is slower to reach fine detail than powder bed fusion, but it can produce large structures and uses wire feedstock that is widely available.

That positions it differently from most metal AM. The competition is not a machined block or a precision casting, but large fabrications, forgings, and castings with long lead times, especially in sectors where replacement parts for older assets are hard to source. Our earlier analysis of digital spare parts inventories and the 3D printed tooling business case described the same pressure from the spare-parts and tooling side.

For those markets, a classification-society approval at the top category is arguably more important than any single performance figure. It gives shipyards, offshore operators, and energy companies a recognised basis for accepting printed structural parts, which has been one of the main barriers to adoption.

The rest of the evidence chain is moving too

Certification focuses on the manufacturing process. In the same week, the material feeding that process and the inspection that follows it both showed signs of becoming more formal.

Feedstock is becoming contractual. On September 30, TCT Magazine reported a 30-month supply agreement between 6K Additive and ADDMAN for Nickel 718 powder, valued at up to $10.8 million with an $8.1 million guaranteed minimum. The report says the agreement includes a structured revert buy-back program: 6K Additive purchases used powder from ADDMAN and reprocesses it with its microwave plasma system. Closed-loop powder handling has existed informally for years. Writing it into a multi-year contract between a powder producer and a service bureau makes the material’s history something that is planned, priced, and recorded. We discussed why that history matters in our article on recycled titanium and reused AM powder.

Inspection data is converging. In an October 5 interview with TCT Magazine, ZEISS product manager Paul Brackman described how powder characterisation, in-process scanning, and X-ray CT are being combined, and pointed to a higher-voltage CT system aimed at dense metal AM parts. A September 29 overview from 3Dnatives reached a similar conclusion: in-situ monitoring and post-process inspection are complementary, with monitoring flagging anomalies during the build and post-build methods confirming the finished part. The article also notes the limits: monitoring produces large volumes of data that are hard to act on, and CT sensitivity falls as part size grows.

Standards still have gaps. The public scope of ASTM E3166, a guide to nondestructive examination of metal AM aerospace parts after build, covers powder bed fusion and directed energy deposition with laser or electron energy sources. It explicitly excludes in-process monitoring, input material control, and accept–reject criteria. Arc-based deposition and in-process data, two of the areas moving fastest, are therefore not covered by this guide. Requirements for them still come from company specifications, classification-society rules, or customer agreements.

What we expect next

The following is our assessment, not a forecast from any of the companies or bodies cited.

1. Approval will become a market-entry requirement for structural AM. As one WAAM producer reaches the top category, owners and shipyards will increasingly ask competitors for the same level of documentation. Certification is likely to separate suppliers that can sell into regulated structural markets from those limited to prototypes and tooling.
2. Evidence will be linked from powder or wire to finished part. Buy-back contracts, lot-level material records, build monitoring, and CT results are each useful separately. The competitive advantage will come from connecting them, so that a finished part can be traced to the feedstock and build conditions that produced it.
3. Standards will lag behind practice in arc DED and process data. Until published standards cover these areas, much of the acceptance logic will be written into individual certificates and customer specifications. That favours manufacturers able to document their own methods clearly.

Signals worth watching over the next year include certificates that name specific alloys and part classes, the first publicly reported AMC3 parts in service, further revisions to DNV-ST-B203, and whether other classification societies and asset owners publish comparable routes for large printed structures.

For companies working with metal additive manufacturing, technical ceramics, or precision machining, Como Precision is available to discuss specific projects.