CNC Tech in Shipbuilding: How Marine Parts Get Made

• 1
  Why CNC tech in shipbuilding replaced hand layout
• 2
  What five-axis motion actually buys you in a hull
• 3
  Tolerances, finishes and the size ceiling
• 4
  Which marine alloys machine well and which fight back
• 5
  When CNC is the wrong process for a marine part
• 6
  Inspection and documentation in marine supply
• 7
  Lead time, quantity and how marine jobs get quoted
• 8
  Which process fits which marine part

Shipyards once built every component from a full-size wooden template, a chalk line and a hand-fed drill. That method still works on one-off hull plate, but it fails the moment a part has to fit a pump, a gearbox or a bearing housing. The tolerance chain closes up. CNC tech in shipbuilding exists because a machined mating face is repeatable: the tenth bracket comes off the machine the same as the first.

Cutting is only half the story. A CNC program also produces the setup sheet, the tool list and the in-process inspection record. When a classification surveyor asks how a rudder carrier was made, that paperwork is the answer. Hand layout leaves you with a part and a story. CNC leaves you with a part and a data trail.

The marine environment sets the real constraint. Salt spray, vibration and temperature cycling punish any joint that is not tight. A flange face with 0.1 mm of mismatch will leak, no matter how good the gasket is. Machining brings that face into the range where the seal works, usually Ra 0.8–1.6 μm and flat within a few hundredths of a millimeter.

None of this makes CNC a universal answer. It is a slow, expensive way to make a flat plate with four holes. The rest of this page is about where the line sits.

A three-axis mill moves the tool in X, Y and Z while the part stays bolted to the table. To machine five faces, you unclamp, rotate and re-zero the part. Every re-clamp adds a setup error, and on a 1.5 m housing that error stacks fast.

Five-axis machining adds two rotary axes, so the tool reaches the part from almost any direction in one setup. On a propeller shaft bracket or a strut, that means the bearing bore, the mounting pads and the fairing can all be cut without the part leaving the fixture. One datum, one setup, one tolerance stack.

The gain is not only accuracy. Complex geometry that a designer once split into welded sub-assemblies can be cut as one solid piece. Fewer welds means fewer heat-affected zones, fewer stress risers and less post-weld straightening. On a sea water pump housing, that is the difference between a part that lasts ten years and one that cracks at a weld toe.

Five-axis also handles undercuts and deep pockets that a three-axis tool simply cannot reach. Impeller passages, valve bodies and manifold cores fall into that group. If the geometry is prismatic and open from one direction, three-axis is faster and cheaper.

We run 16 simultaneous five-axis machining centers, 12 four-axis mills, 27 three-axis machines and 16 mill-turn centers. That mix matters more than any single machine. The right answer for a deck cleat is rarely the right answer for a rudder stock.

Our standard machining tolerance is ±0.005 mm (±0.0002 in) on critical features. That is a capability number, not a promise on every dimension. A 3 m long weldment will move after machining as residual stress releases, so we hold tight tolerances on short, rigid features and open them up where the part is long and thin.

Surface finish follows the same logic. A bearing bore or a hydraulic sealing face may need Ra 0.2–0.8 μm. A structural bracket is fine at Ra 3.2 μm as-machined. Specifying a mirror finish on a part that gets painted wastes money and adds lead time.

Size is the harder ceiling. Maximum processing size is 4,000 mm, with travels of 4,000 × 400 × 150 mm on the large machines. Medium work sits at 750 × 1,150 × 550 mm or 600 × 600 × 600 mm. Compact parts run on 500 × 500 × 450 mm and 500 × 310 × 200 mm machines, with a Ø400 mm rotary table for round work.

Anything beyond 4,000 mm has to be split, welded or machined in sections. That is a design decision, not a machining one. If you can keep a critical bore inside the envelope, you avoid a welded joint in the most highly loaded area of the part.

Stainless 316 and 316L are the default for sea water contact. They machine cleanly, weld well and resist pitting. The trade-off is galling: 316 threads tend to seize, so we use thread inserts or specify a different alloy for fasteners. 17-4PH (SUS630) gives roughly three times the yield strength of 316 after heat treatment, and it still machines predictably. Use it for shafting, pins and high-load brackets.

Aluminium 5083 is the workhorse for superstructures and non-pressurized hull sections. It keeps strength after welding better than 6061 and resists sea water reasonably well. 6061-T6 machines faster and holds a better finish, but it loses strength in the heat-affected zone, so it belongs on brackets and interior frames rather than welded hull plate.

Titanium TC4 (Ti-6Al-4V) and Inconel come up on exhaust components, fasteners and high-temperature valve parts. Both are tough on tooling. Cutting speeds drop to a fraction of what aluminium allows, tool life is short, and the cycle time can be five times longer. They are worth it only where the temperature or corrosion duty demands them.

Bronze and copper alloys still have a place on propeller hardware, bearings and bushes. C36000 brass machines fast and holds threads well but dezincifies in sea water, so it stays on the dry side of the system. Beryllium copper is a niche choice where you need conductivity plus strength.

Thin sheet metal with a handful of holes is a laser or waterjet job. Cutting 6 mm plate on a mill is slow, and the flatness you gain usually does not matter for a cover panel. Waterjet also avoids the heat-affected zone that laser cutting leaves on thicker stainless.

Large, simple weldments are a fabrication job. If a part is three plates welded into a T and only one face needs to be flat, you fabricate and then machine that one face. Machining the whole thing from solid wastes material and machine time.

Very low volume, non-critical geometry can still be hand-finished. A one-off wooden template part for a repair in a remote yard may not justify programming. But once the same part is needed twice, the program pays for itself.

Parts that need internal cooling channels or lattice structures are usually better served by casting or additive processes first, then machining only the critical faces. Hybrid routes are common in marine work: cast the body, machine the sealing surfaces.

The honest test is this. If a dimension has to be right, or two features have to line up, machine it. If it just has to exist, there is probably a faster way.

Marine buyers do not just want a part. They want evidence that the part is what the drawing says. We inspect 100% of parts before shipment, with raw material check, in-process monitoring and final inspection. Reports are available on request, including dimensional data for critical features.

Material traceability is the other half. A 316L bracket without a mill certificate is a guess. When the part goes into a sea water system, that guess becomes a corrosion failure two years later. We keep material documentation with the job so it can be matched to the delivery.

Programs are held with revision control. If a drawing changes, the old program is retired rather than edited in place. That sounds bureaucratic until a replacement part has to match a ten-year-old vessel. Then it is the only thing that saves the fit.

We hold ISO 9001:2015, IATF 16949:2016, ISO 13485:2016 and ISO 27001:2022. The quality system matters less than the habit it enforces: check the part, record the check, keep the record.

Quotation and free DFM analysis come back within 12 hours. Production can start within 24 hours of a released order. Parts ship in 3–5 days for standard jobs. Historical late-delivery probability sits below 2%. Those numbers assume a complete drawing and a defined material.

There is no minimum order quantity. A single prototype bracket and a 10,000-part run go through the same quoting path. For marine repair work, the single prototype is often the whole job, and that is fine.

DFM feedback is where most marine jobs save money. A 0.5 mm corner radius where the drawing shows a sharp internal corner forces a small tool and a long cycle. Opening it to 2 mm can cut machining time by a third. The same applies to deep pockets and tolerance callouts that no one will measure.

Uploads are kept secure and confidential. An NDA is available on request if the drawings are sensitive. For new designs, the sample center shows what comes off the machines before you commit to a run.