Ensuring Uniform Properties During Additive Manufacturing with IET

Key Takeaways

• Additive manufacturing’s layer-by-layer process creates elastic property gradients driven by porosity, residual stress, and microstructural variation — all of which shift resonance frequency in ways that IET resolves non-destructively.
• Printing reference bars alongside production parts and measuring them with IET converts the build volume into a spatial property map, exposing machine drift and process inconsistency before downstream operations add further value to potentially defective parts.
• IET measures Young’s modulus, shear modulus, Poisson’s ratio, and internal friction from a single tap in seconds — providing the same elastic characterization that would otherwise require multiple destructive tests on sacrificial specimens.
• Frequency and damping together detect both the stiffness reduction from porosity and the frictional dissipation from micro-cracks, giving AM quality teams a two-channel signal that is more discriminating than either metric alone.
• At scale, IET-based 100% screening of AM parts generates the statistical density needed to distinguish part-level outliers from batch-level process drift — a distinction that sampled destructive testing structurally cannot make.

The Uniformity Problem That AM Teams Under-Estimate

Additive manufacturing offers geometric freedom that no subtractive process can match. But that freedom carries a less-discussed cost: the same complexity that enables intricate lattices and consolidated assemblies also makes the property field of a finished part extraordinarily sensitive to process conditions that conventional manufacturing simply does not face.

In a machined metal component, the bulk material properties are established before the part is shaped. In an additively manufactured component, the material and the part become simultaneously — layer by layer, scan vector by scan vector, with each local thermal event influencing the grain structure, porosity content, and residual stress state of the material directly beneath it.

This means two parts from the same STL file, built in the same machine on consecutive days, can have measurably different elastic moduli — not because the design changed, but because a subtle shift in laser power, powder layer thickness, or build chamber atmosphere altered the local solidification conditions at the scale of individual melt pools. Summed across hundreds of layers, this microscopic variability compounds into macroscopic property differences.

Detecting those differences requires a measurement sensitive enough to resolve them, fast enough to apply to every part, and non-destructive enough not to consume the parts it inspects. Impulse excitation technique provides all three.

What IET Measures and Why It Captures AM Property Variation

The physical relationship underlying IET is straightforward: a specimen’s resonance frequency is proportional to the square root of its elastic modulus divided by its density (f ∝ √(E/ρ)). Because a single tap excites this frequency in seconds, and because elastic modulus is a direct function of atomic bond stiffness and microstructural continuity, the measurement is inherently sensitive to precisely the features that vary in additive manufacturing:

Porosity acts as a compliance field — each void removes load-bearing material and locally softens the matrix. A part with 2% porosity has a measurably lower elastic modulus than a fully dense counterpart of the same geometry, and therefore a measurably lower resonance frequency. The relationship is consistent enough that GrindoSonic’s correlation studies between IET measurements and CT-validated porosity have established quantitative relationships for specific alloy systems, enabling porosity estimation directly from frequency data.

Micro-cracks at layer interfaces, along grain boundaries, or at inclusion sites contribute a frictional damping component that IET captures independently of the frequency shift. The damping ratio — the rate at which vibration energy decays — rises as micro-crack density increases. This dual signal (frequency for stiffness, damping for crack-related dissipation) means that two parts with identical resonance frequencies but different damping ratios carry different damage states — a distinction invisible to frequency-only screening.

Anisotropy from preferred grain orientation and columnar microstructure — characteristic of laser powder bed fusion builds along the build direction — manifests as different resonance frequencies for flexural versus torsional vibration modes. Measuring both modes, per ASTM E1876, yields Young’s modulus and shear modulus independently. The ratio between them reveals Poisson’s ratio and, by extension, the degree of elastic anisotropy that build direction introduces.

Residual stress from rapid thermal cycling during each layer affects both the frequency and the non-linear acoustic response of the part. While stress alone produces smaller frequency shifts than porosity, its interaction with micro-crack propagation under service loading is critical — and IET damping data provides an indirect window into the stress state that complements direct diffraction or hole-drilling measurements.

The Reference Bar Strategy: Mapping the Build Volume

The most powerful application of IET in additive manufacturing is not part-level screening in isolation — it is systematic mapping of property variation across the entire build volume using purpose-printed reference specimens.

The concept is simple and experimentally robust. Before the production build, a set of bar-shaped specimens is added to the build file at predetermined positions: corners, center, different heights within the stack. These bars are printed in parallel with the production parts, experiencing the same local thermal environment, the same powder layer, the same atmosphere, and the same post-build heat treatment.

After the build, each reference bar is measured by IET. The resulting frequency and damping values are mapped onto the build volume geometry, producing a spatial property distribution: regions of the build platform where elastic modulus is high and uniform (well-controlled process conditions) and regions where frequency or damping deviate from the target band (local process anomalies). Production parts that occupied anomalous zones are flagged for additional inspection or rejection before machining, coating, or shipping adds further cost.

This strategy converts IET from a per-part measurement into a machine calibration and process audit tool. It answers the question that simple end-of-line screening cannot: not just which parts are out of specification, but where in the build volume and under which process conditions the deviation originated. That diagnostic specificity is what makes it possible to correct the machine or process parameter rather than merely screen out its consequences.

Applying IET Across the AM Process Chain

Property uniformity is not a single-point concern in additive manufacturing. It is a cumulative outcome of decisions and conditions that span the full process chain — from powder qualification through post-processing — and IET adds diagnostic value at multiple stages.

Powder Feedstock Qualification

Powder reuse is economically necessary in metal AM but introduces progressive changes in particle morphology, oxide content, and flowability that affect melt pool dynamics and therefore consolidated part properties. Printing a standard test geometry from each powder batch and measuring its elastic modulus by IET provides a fast, sensitive check on feedstock consistency that weight-based density or sieve analysis alone cannot deliver.

Post-Build Heat Treatment Verification

Stress relief, hot isostatic pressing (HIP), and aging treatments alter grain structure, residual stress, and phase composition — all of which shift elastic modulus and damping. IET measurement before and after each thermal treatment stage provides a quantitative check that the treatment achieved its intended microstructural effect. A HIP cycle that fails to fully close sub-surface porosity leaves a detectable frequency deficit relative to the baseline for fully dense, HIPed material.

Final Part Screening

At the end of the process chain, GrindoSonic’s automated measurement system enables 100% inspection of production parts against the property distribution established during process qualification. Every part receives a resonance frequency and damping measurement; the system applies GO/NOGO logic based on the qualified property band and flags outliers for review or rejection — at throughputs that make 100% screening economically viable where sampled destructive testing is not.

Real-World Example: Validating LPBF Process Consistency Across a Build Stack

A manufacturer producing safety-critical aerospace brackets in a titanium alloy qualifies their laser powder bed fusion process using a combination of tensile testing, CT scanning, and microstructural analysis. The qualification establishes the property range for conforming parts. But the qualification itself was performed on a single build, under stable conditions, with fresh powder.

Three months into production, subtle changes have accumulated: the powder has been recycled several times, the laser has drifted slightly in power, and the build volume has been fully utilized with denser part packing than the qualification build. None of these changes individually crosses a visible threshold — but their combined effect on part elastic modulus is measurable.

IET screening of reference bars from each build reveals a gradual downward trend in mean Young’s modulus across successive builds — small (under 1%) but statistically consistent across multiple builds, signaling that the cumulative process drift has moved the operating point away from the qualification center. Corrective action — laser recalibration, powder refresh — is taken and confirmed in the subsequent build’s reference bar data before any non-conforming production parts have shipped.

This is the value of continuous, quantitative process monitoring rather than periodic requalification: the drift is caught while it is a small, correctable signal, not after it has propagated into a field failure investigation.

From Process Insurance to Process Intelligence

Quality control in additive manufacturing is often framed as insurance — a cost incurred to prevent bad parts from reaching customers. IET reframes it as intelligence — a measurement that generates actionable information about process state with every build.

The frequency and damping distributions from a fleet of production builds encode the history of machine performance, powder quality, and parameter stability over time. Analyzed as a time series rather than a per-build compliance check, this data reveals seasonal atmospheric effects on build quality, correlates property shifts with maintenance events, and identifies which process parameters carry the highest sensitivity to part performance — insights that no amount of end-of-line dimensional inspection or destructive sampling can provide.

For AM teams operating under pressure to reduce qualification costs, accelerate material development cycles, and demonstrate process control to customers and certification bodies, IET measurement is not an additional burden on the quality program. It is the measurement infrastructure that makes the rest of the program faster, cheaper, and more defensible.