Machining After 3D Printing

For aerospace and defense hardware, additive manufacturing (AM) rarely ends when the build completes. Powder bed fusion (PBF)—including DMLS/SLM—can deliver near-net geometries, internal flow paths, and consolidated assemblies, but most flight- and mission-critical parts still require finish CNC machining to achieve controlled datums, tight tolerances, sealing surfaces, and predictable fits. The practical reality is hybrid manufacturing: AM is used to create geometry efficiently, and machining is used to lock in precision, repeatability, and inspection confidence.

This article walks through how machining after 3D printing works in real production environments—how teams plan machining stock, establish datums, fixture irregular geometries, set surface finish targets, and build an inspection approach that holds up under AS9100 quality systems and regulated procurement expectations (including ITAR and DFARS flows where applicable). The goal is to help engineers and sourcing teams avoid the most common late-stage surprises: insufficient stock, unstable workholding, distorted parts after stress relief or HIP, and inspection gaps that delay First Article Inspection (FAI).

Why machining is often required

PBF processes excel at complex shapes, but they are not inherently a “finished part” process. Machining is often required for six practical reasons:

1) Tolerance and geometric control. Even with well-tuned machines and parameter sets, PBF dimensional variation is influenced by scan strategy, local thermal gradients, support strategy, and powder/recoat behavior. Features like bores, bearing seats, dowel holes, flange faces, and precision interfaces typically require machining to hit tight limits and true position requirements consistently.

2) Establishing functional datums. AM surfaces are not ideal datum features because their as-built texture and slight waviness can compromise repeatable measurement and assembly. Machining creates clean, stable datums for downstream machining, inspection, and assembly, enabling predictable coordinate systems.

3) Surface finish and sealing requirements. Many aerospace/defense interfaces require controlled roughness for sealing, fatigue performance, and flow behavior. As-built PBF surfaces are typically too rough and anisotropic for sealing lands, o-ring glands, bearing journals, or aerodynamic surfaces. Finish machining (and sometimes grinding, honing, or lapping) is the common path to meet Ra and waviness targets.

4) Fit, form, and function features not suited to as-built PBF. Threads, sharp edges, precision grooves, spline-like features, and certain counterbores may be unreliable in as-built form. Even when threads can be printed, most regulated programs prefer machining or tapping for better control and inspectability.

5) Post-processing dimensional change (stress relief, HIP, heat treat). Stress relief and heat treatment can change shape slightly; Hot Isostatic Pressing (HIP) can also shift geometry while improving density and fatigue behavior. When these steps are part of the route, machining is planned after them to “final size” the part.

6) Certification and inspection practicality. Procurement and quality teams often need a clear path to verify conformance. Machined datums and machined critical features simplify CMM programming, gage design, and objective evidence in an AS9102 FAI package.

In practice, the decision is not “AM vs. machining,” but where AM adds value (complexity, consolidation, lead time) and where machining is essential (interfaces, precision, verification).

Adding machining stock

“Leave stock for machining” sounds simple until you are dealing with thin walls, lattice-backed skins, or irregular geometry coming off a build plate. Successful teams treat machining stock as a controlled design parameter—defined per feature, tied to the process route, and validated with build-to-build capability data.

Start with a process route. Stock requirements depend on what happens between printing and final machining. A typical regulated route might look like:

Step 1: PBF build (DMLS/SLM) with defined orientation, supports, and witness coupons.

Step 2: Stress relief (or in some cases in-situ plate heating + post build stress relief) to reduce residual stresses.

Step 3: Part removal from build plate (wire EDM is common for high-value builds), support removal, and initial cleanup.

Step 4: HIP when required by spec or performance needs (common for critical Ti-6Al-4V and Ni-based superalloys), followed by any required solution/aging cycles.

Step 5: Rough machining to establish datums and create stable reference features.

Step 6: Finish machining (often 5-axis), deburr, and controlled surface finishing.

Step 7: NDE and inspection (CMM, CT scanning, dye penetrant, etc.), documentation, and release with CoC and traceability records.

Define stock by feature criticality and expected distortion. Common starting points (which must be validated for your material, machine, and geometry) include leaving extra stock on:datum pads, flange faces, bores, and sealing lands. Thin walls can distort during support removal and stress relief, so they may need less stock (to avoid overcutting into distortion) but more robust fixturing strategy. Conversely, robust bosses and pads can tolerate more stock and provide a reliable path for establishing datums.

Plan for HIP and heat treat movement. If HIP is in the route, it is generally safer to not finish-machine critical interfaces before HIP. HIP can slightly change dimensions and round edges; finishing after HIP helps ensure final tolerances are met. If some features must be machined before HIP (e.g., to allow HIP capsule fit or to create temporary datum pads), treat them as pre-machined features with generous allowances for final finishing.

Use “machining stock” as a drawing-controlled or model-based definition. For regulated work, it is common to include stock notes on the drawing or in the model-based definition (MBD) to prevent misinterpretation at the supplier. A clear approach is to identify:(a) surfaces to be left “as-built,”(b) surfaces to be machined, and(c) required stock allowance or target pre-machined condition.

Build-to-machine coupons and capability data. Mature programs do not guess. They correlate as-built dimensions, post-heat-treat/HIP dimensions, and post-machining outcomes across multiple builds. That data informs future stock allowances and reduces scrap. If your supply base provides build reports, request dimensional capability summaries on representative features—not just a one-off “it passed inspection.”

Fixturing and datums

Fixturing is where hybrid manufacturing succeeds or fails. AM parts can be asymmetric, thin-walled, and internally complex, which makes conventional workholding risky. The best practice is to design the part and the build with machining in mind, including datum strategy and temporary features for fixturing.

Establish datums intentionally, not opportunistically. Many machining problems start when teams pick an “easy” surface as a datum even though it is not functionally meaningful. For aerospace/defense parts, datums should align with assembly interfaces and inspection needs. A common approach is:

Primary datum: a machined plane (flange face or pad) that controls orientation.

Secondary datum: a machined bore or slot that controls lateral position.

Tertiary datum: a machined face or feature that locks rotation.

Use sacrificial/temporary datum pads and tabs. If the functional geometry does not provide good clamping surfaces, add build-in pads, tabs, or bosses that will be removed later. These features can:(a) improve rigidity during cutting,(b) provide consistent datum targets,(c) prevent damage to functional surfaces.In RFQ packages, call these out explicitly so everyone understands they are intentional and removable.

Plan the first operation to “create the part that can be fixtured.” Often the first machining operation is not about hitting final dimensions—it is about creating flatness, squareness, and a reliable hole pattern that allows repeatable re-fixturing. This can include:(1) skimming a datum face,(2) drilling/reaming a pair of tooling holes,(3) machining reference pads for probing.

Account for anisotropy and thin-wall dynamics. Printed metals often have directional properties and residual stress histories. Thin sections can “spring” during cutting, leading to taper, chatter, or out-of-tolerance bores. Mitigations include:(a) choosing conservative cutting parameters and sharp tooling suited to the alloy condition,(b) using distributed clamping to reduce local deformation,(c) machining in a sequence that balances material removal to minimize distortion.

5-axis probing and in-process metrology are not optional for complex AM geometry. When a part has few orthogonal surfaces, 5-axis probing to establish a best-fit coordinate system can dramatically reduce setup error. A practical pattern is to probe printed reference features (or pre-machined pads), apply a coordinate rotation/translation, then machine critical features in a single setup to reduce stack-up.

Think about support strategy and cut access. Supports can block tool access or create hardened/irregular regions after removal. If machining access is critical, coordinate the AM build orientation and support design with the machining plan. A design that prints successfully but cannot be machined efficiently is a common hybrid failure mode.

Surface finish targets

Surface finish in hybrid manufacturing is not just aesthetic—it drives sealing, fatigue life, friction, wear, and flow. The right approach is to assign finish targets based on function and inspection capability, not a blanket “make it smooth.”

Separate surfaces into categories. A pragmatic categorization used by many suppliers is:

Category A (critical interfaces): sealing lands, bearing journals, mating flanges, hydraulic ports, valve seats. These typically require machining, controlled roughness, and sometimes additional finishing (honing/lapping) depending on the spec.

Category B (secondary machined surfaces): bolt patterns, external profiles, datum faces used for inspection or assembly alignment. These are usually machined to achieve geometry and moderate finish.

Category C (as-built or minimally finished): internal passages, non-contact external surfaces, weight-reduction features. These may remain as-built if requirements allow, or be media blasted/shot peened for cosmetic uniformity and stress state control.

Make finish measurable and inspectable. Call out roughness with appropriate parameters (commonly Ra), but ensure the requirement is realistic for the geometry and tool access. For internal channels, be cautious: you may not be able to measure finish directly without destructive methods or specialized metrology. If internal roughness affects flow, consider validating by performance testing, CT-based surface characterization, or by designing internal features that can be finished (e.g., straight bores that can be reamed/honed).

Recognize AM-specific surface realities. As-built PBF surfaces can have partially sintered particles and directional texture. Media blasting may remove loose particles but does not guarantee a uniform roughness profile. For fatigue-sensitive parts, surface condition can be a major driver; machining removes the outer “skin” and any near-surface defects, while HIP addresses internal porosity. The combined route (print → HIP → machine critical surfaces) is a common strategy to improve performance and reduce variability.

Coordinate finish targets with coatings and downstream processes. If the part will receive plating, thermal spray, anodize, or other controlled finishing, machining finish targets may need to account for coating thickness and adhesion requirements. In NADCAP-governed special processes, the surface preparation method and final condition must be consistent with qualified process parameters and documented traveler steps.

Inspection approach

Inspection is where engineering intent becomes objective evidence. A strong inspection approach is planned alongside the machining strategy and includes what to inspect, how to inspect, when to inspect, and how to document.

1) Define critical characteristics early. Identify the features that drive fit, function, and safety—datums, true position of mounting holes, coaxiality of bores, sealing surface flatness, minimum wall thickness, and any internal features that impact flow or strength. Procurement teams should ensure these are explicit in the RFQ package to avoid assumptions.

2) Use staged inspection: as-built, post-HIP/heat treat, post-machining. Staged inspection reduces risk by catching issues before value-added steps. A realistic staged plan might include:

As-built / post-support removal: visual inspection, basic dimensional checks on accessible reference features, and documentation of build parameters and powder lot.

Post-HIP / heat treat: dimensional sampling of key features that are known to move, material verification as required (chemistry/mechanical testing via coupons), and NDE where specified.

Post-machining: full dimensional inspection to drawing requirements, surface finish verification on critical surfaces, and any final NDE/pressure/leak testing.

3) Choose metrology tools that match the geometry. Common tools and where they fit:

CMM: best for machined datums and GD&T-controlled features; requires stable datum strategy.

CT scanning: valuable for complex internal passages, lattice, wall thickness verification, and detecting internal anomalies. CT is also useful when you need objective evidence of internal geometry that cannot be probed. CT should be planned with acceptance criteria (voxel size, artifact control, and what constitutes a nonconformance).

NDE (e.g., dye penetrant, ultrasonic, radiography): used per material and specification requirements. When special processes or NDE are required under customer flowdowns, ensure the provider’s accreditations and written practices are compatible with contract requirements (often NADCAP in aerospace contexts).

4) Build a documentation pack that procurement can use. For defense/aerospace sourcing, the deliverable is not only the part but the paperwork. A typical certification pack may include:(a) material traceability (powder heat/lot, incoming inspection records),(b) build report and traveler with process steps,(c) HIP/heat treat charts and certifications (when applicable),(d) NDE reports,(e) dimensional inspection report / CMM output,(f) Certificate of Conformance (CoC),(g) AS9102 First Article Inspection (when required),(h) ITAR/DFARS compliance statements if contractually required (e.g., controlled technical data handling, specialty metals clauses as applicable).The key is that records should be internally consistent: part serial numbers, traveler IDs, coupon IDs, and inspection report identifiers must tie together cleanly.

5) Control configuration and revisions. Hybrid programs can change quickly (orientation tweaks, support changes, updated stock). Ensure the inspection plan and CNC programs align to the correct revision of the model/drawing, and that any deviations are documented through a controlled process (e.g., MRB disposition). Under AS9100, configuration management discipline is not optional.

Common workflow pitfalls

Most hybrid manufacturing problems are not “AM problems” or “machining problems” alone—they are interface problems between teams, suppliers, and process steps. The following pitfalls show up repeatedly in aerospace and defense programs, especially during transition from prototype to production.

Pitfall 1: Treating stock allowance as a single global number. Different features behave differently through stress relief and HIP. A global “+0.5 mm stock everywhere” approach can create thin sections that are unmachinable or leave insufficient stock on critical pads. Fix: define feature-based stock and validate with capability data.

Pitfall 2: Establishing datums on as-built surfaces. As-built texture and waviness increase setup variation and make CMM results noisy. Fix: create machined datum pads early and reference everything from them.

Pitfall 3: Underestimating part distortion after support removal or heat treatment. Removing supports can release residual stress; HIP/heat treat can move geometry. Fix: include distortion risk in DFMEA, plan staged inspection, and time machining steps after dimensional stabilization.

Pitfall 4: Designing for printability but not machinability. Engineers may optimize for build success and forget tool access, workholding, and chip evacuation. Fix: conduct joint AM + CNC design reviews; add temporary tabs/pads; adjust orientation for machining access when needed.

Pitfall 5: Inadequate fixturing rigidity for thin-walled AM structures. Chatter and deflection can destroy tolerances and surface finish. Fix: use distributed clamping, custom soft jaws, sacrificial supports, and toolpaths that minimize cutting forces; consider leaving ribs or temporary stiffeners to be removed after finishing.

Pitfall 6: Misaligned inspection requirements and actual measurability. Calling out tight tolerances on internal features that cannot be measured reliably creates schedule and acceptance risk. Fix: align GD&T and finish callouts with available metrology (CMM/CT/NDE) and define acceptance criteria up front.

Pitfall 7: Documentation gaps that delay acceptance. A part may be dimensionally correct, but missing traceability, HIP charts, or NDE certifications can stop shipment. Fix: treat the cert pack as a deliverable with its own checklist and internal audit. Ensure ITAR handling and customer flowdowns are addressed in the traveler and supplier qualifications.

Pitfall 8: RFQs that do not specify the hybrid route. If a buyer sends only a CAD file without clarifying HIP, machining, NDE, and FAI expectations, suppliers will quote different assumptions, and the program will pay later. Fix: RFQs should include drawing/MBD, process expectations (e.g., “PBF + HIP + 5-axis machining”), required certifications (AS9100, NADCAP special processes where applicable), inspection deliverables, and packaging/marking requirements.

Hybrid manufacturing is most successful when AM, HIP/heat treat, machining, and inspection are treated as a single integrated process—not disconnected operations. When the route is defined early, stock and datums are designed intentionally, and inspection is planned for real measurability, machining after 3D printing becomes a predictable, production-ready method for delivering complex, high-performance parts to aerospace and defense programs.