De-Risking Additive Manufacturing with NDT

Why Additive Manufacturing Demands a Different Quality Mindset

Additive manufacturing builds parts by melting material layer by layer. In laser powder bed fusion (LPBF), a focused laser creates roughly 10,000 to 100,000 individual melt pools per cubic centimeter. Each pool solidifies under slightly different thermal conditions depending on geometry, scan strategy, and local gas flow. The result is that every AM build is unique, even when the same STL file runs on the same machine with the same powder lot.

Mature processes like casting or forging rely on decades of data that have established tight process-property relationships, and their inspection methods are well-understood.

The consequence for quality assurance is significant. Conventional manufacturing can rely on statistical process control built on large historical datasets and periodic destructive sampling. Additive manufacturing cannot.

Build-to-build variability from factors like powder reuse degradation, inert gas flow patterns, laser calibration drift, and recoater blade wear means that even a “validated” process can produce defective parts on any given build. Non-destructive testing is essential for any AM production application where part failure carries consequences.

Key takeaway: AM defects are process-dependent and vary between builds, unlike the predictable failure modes of conventional manufacturing. This build-to-build variability makes 100% non-destructive inspection essential rather than optional.

The Defect Landscape in Metal AM

Understanding which defects to look for is the first step toward effective inspection. Metal AM processes, particularly LPBF and electron beam melting (EBM), produce a characteristic set of flaws that differ from those found in wrought or cast metals.

Lack-of-fusion porosity occurs when insufficient energy input leaves unfused powder particles trapped between layers. These irregularly shaped voids act as stress concentrators and are the most damaging AM defect class, reducing fatigue life by 50-90% depending on size and location.

Gas porosity arises from dissolved gas in the powder or shielding atmosphere, producing spherical pores typically 10-100 μm in diameter. While less harmful individually than lack-of-fusion defects, high gas porosity density degrades both stiffness and fatigue performance.

Keyhole porosity forms at excessive energy densities when the melt pool collapses, trapping vapor in deep, narrow pores at the base of the keyhole.

Delamination, separation between successive layers, results from residual stress accumulation exceeding the interlayer bond strength, and often signals problems with build parameters.

Balling produces rough, spherical surface features when surface tension dominates melt pool dynamics, typically at excessive scan speeds or insufficient energy input.

Each of these defect types affects the mechanical response of the part in measurable ways: reducing effective stiffness, increasing internal friction, or both. This is precisely what makes resonance-based inspection so well-suited to AM quality control.

IET as a Post-Build Screening Method

Impulse Excitation Technique measures two independent physical quantities simultaneously: resonant frequency and damping (internal friction). This dual-indicator capability gives IET high sensitivity to the range of defects found in AM parts.

Resonant frequency is governed by elastic modulus, geometry, and density. A part with internal porosity has lower effective stiffness and density, producing a measurable frequency drop. Damping captures energy dissipation from internal surfaces such as crack faces, pore walls, and unbonded interfaces. A part can have normal frequency but elevated damping, or reduced frequency but normal damping, and each pattern points to a different defect mechanism.

The measurement itself takes seconds. A small mechanical impulse excites the part’s natural vibration modes, a microphone or laser vibrometer captures the acoustic response, and signal processing extracts the resonant frequencies and damping values. No couplant, no scanning, no skilled operator interpretation: the result is a pair of numbers that can be compared directly against acceptance criteria.

For production environments, this translates to throughput rates of hundreds of parts per hour at a cost below $1 per test, compared to $500-2,000 per part for computed tomography (CT) scanning with cycle times of 30-90 minutes.

The practical approach in AM facilities combines IET screening with targeted CT validation. Every part receives an IET measurement, a 100% inspection rate that would be economically impossible with CT alone. Parts that fall within established frequency and damping windows pass directly to the next production step. Parts that deviate beyond threshold values, typically 2-5% of production, are routed to CT for detailed volumetric defect analysis.

This tiered strategy captures the speed and cost advantages of IET while preserving the diagnostic detail of CT where it matters most.

Integrating NDT into the AM Workflow

Effective quality assurance in additive manufacturing extends beyond a single inspection gate at the end of production. It is a continuous thread running through the entire workflow. The most robust AM operations embed NDT at three distinct stages, each serving a different purpose.

Pre-build verification uses test coupons printed alongside production parts to qualify powder lots and machine conditions. Simple rectangular bar specimens (per ASTM E1876) printed at the start of each build provide a baseline frequency and damping fingerprint for that specific machine-powder-parameter combination.

Deviations from historical baselines flag process drift before it affects production parts, catching issues like powder degradation from excessive reuse or subtle shifts in laser power output.

Post-build screening applies IET directly to finished parts or dedicated witness specimens. For geometries that lend themselves to resonance measurement (prismatic bars, cylinders, discs, and even lattice structures with repeating unit cells), direct part measurement provides the most relevant quality data.

For complex geometries where modal analysis becomes impractical, witness coupons printed in the same build at representative locations serve as proxies for local build quality. The key insight is that IET measures bulk volumetric properties: a single measurement integrates information across the entire specimen volume, making it sensitive to distributed defect populations that point-based methods might miss.

Process monitoring over time tracks frequency and damping trends across builds to detect gradual process degradation. A slowly drifting mean frequency across successive builds may indicate powder chemistry evolution, while increasing damping scatter could signal growing process instability.

Engineers can use this statistical perspective to move NDT beyond pass/fail gating and into continuous process improvement.

The Standards Landscape for AM NDT

The standards framework for additive manufacturing inspection is maturing rapidly. ASTM F3122 provides guidance on evaluating mechanical properties of AM metallic materials, including recommendations for non-destructive evaluation methods. ASTM E1876, the foundational standard for IET measurement, defines the test method for dynamic Young’s modulus, shear modulus, and Poisson’s ratio by impulse excitation of vibration, and applies directly to AM test specimens. NASA-STD-6030 establishes requirements for additively manufactured spaceflight hardware, mandating NDT for all critical structural applications, while NASA-STD-6033 specifies fracture control requirements.

For organizations building quality management systems around AM production, these standards provide the framework for establishing acceptance criteria, defining inspection intervals, and documenting traceability. IET measurements align naturally with these requirements because they produce quantitative, repeatable data that can be logged, trended, and audited. These are essential attributes for regulated industries like aerospace, medical devices, and energy.

The Economics of AM Quality Assurance

The cost argument for IET-based screening in AM production is straightforward. A typical aerospace LPBF production run might produce 50-200 parts per build. CT scanning every part at $500-2,000 each adds $25,000-400,000 in inspection costs per build, often exceeding the manufacturing cost of the parts themselves.

IET screening at under $1 per part reduces that to $50-200 per build for 100% inspection, with CT reserved only for the small fraction of flagged parts. For a facility running 200 builds per year, the annual savings can reach $1-5 million depending on part volumes and CT costs, while actually improving inspection coverage from sample-based to 100%.

Beyond direct cost savings, faster inspection throughput reduces work-in-progress inventory and accelerates time-to-delivery. Parts cleared by IET within seconds move directly to post-processing, heat treatment, or assembly rather than waiting days in a CT queue.

In industries where AM is adopted specifically for its speed advantage (rapid prototyping, spare parts on demand, short-run production), lengthy inspection bottlenecks undermine the core value proposition. IET-based screening preserves the agility that makes additive manufacturing compelling in the first place.