X-Ray and CT Inspection: Detecting Internal Porosity in Safety-Critical Castings

A single 200-micron gas pore sitting at the root of a fillet in a suspension knuckle can propagate into a fatigue crack within 40,000 load cycles. This is the reality that separates casting inspection for decorative or low-stress components from inspection for safety-critical parts — steering knuckles, brake calipers, turbocharger housings, hydraulic manifolds, and aerospace structural brackets. In these applications, internal porosity that is completely invisible to the naked eye, and often invisible to dye penetrant or magnetic particle testing, can sit undetected inside a wall section until the part fails in service. X-ray radiography and computed tomography (CT) are the only non-destructive methods capable of resolving this class of defect volumetrically, and understanding how to specify, interpret, and act on this data is a core competency for any engineer releasing castings into a regulated supply chain.

At Microns Hub, we run X-ray and CT inspection as a standard verification gate for aluminum and zinc die castings, sand castings, and investment castings destined for automotive, industrial, and off-highway applications. This guide covers the physics behind porosity formation, the technical parameters that separate a marketing-grade scan from a metrology-grade one, the ISO and ASTM standards that govern acceptance, and the cost drivers that determine whether 2D radiography or full 3D CT is the correct tool for a given part.

• Gas porosity and shrinkage porosity have different root causes and require different corrective actions in the casting process, not just different inspection thresholds.
• 2D X-ray radiography (per ASTM E505 / ISO 5579) is sufficient for wall thickness verification and gross porosity screening, but CT is required to size and locate individual pores in complex 3D geometries.
• Acceptance criteria must reference a specific standard and severity level — ASTM E155 reference radiographs or ISO 10042 quality levels — never a subjective visual call.
• Voxel resolution, not just machine wattage, determines the smallest detectable pore; a 450 kV system without the correct focal spot and detector pairing will still miss sub-100-micron defects.

Why Internal Porosity Forms in Metal Castings

Porosity is not a single defect type — it is a family of discontinuities with distinct metallurgical origins, and distinguishing between them is the first step in both root-cause correction and inspection strategy. Treating all porosity as equivalent leads to either over-rejection of harmless defects or, worse, acceptance of a defect type that behaves very differently under cyclic load.

Gas Porosity

Gas porosity forms when dissolved hydrogen (in aluminum alloys such as A356, A380, or Al 6061-T6 castings) or entrapped air/combustion gases (in high-pressure die casting) come out of solution during solidification and become trapped as spherical or near-spherical voids. Gas pores are typically 0,05 mm to 2 mm in diameter, smooth-walled, and distributed relatively uniformly through a section rather than concentrated at a single point. Because they are round, gas pores are less severe stress concentrators than angular shrinkage defects of the same volume, but in large clusters they reduce effective load-bearing cross-section and create leak paths in pressure-tight housings such as transmission cases and hydraulic blocks.

Shrinkage Porosity

Shrinkage porosity occurs when the liquid metal feeding a solidifying section cannot keep pace with volumetric contraction, leaving irregular, dendritic, interconnected voids concentrated at thermal centers — typically the last area to solidify, often near thick-to-thin transitions, ribs, or bosses. Shrinkage defects are far more dangerous under fatigue loading than gas porosity of equivalent size because their sharp, branching morphology acts as a stress riser, and their interconnected nature can create continuous leak paths through an otherwise sound wall.

Micro-Porosity in Thin Walls

In thin-wall die castings (below 2,5 mm), micro-porosity often forms as a fine dispersed network rather than discrete pores, driven by rapid solidification rates and turbulent gate filling. This defect type is notoriously difficult to detect with 2D radiography because the individual voids are below the resolution threshold when viewed in projection, which is precisely why CT reconstruction — where the part is rebuilt as a true 3D density map — has become the standard for safety-critical thin-wall aluminum housings.

X-Ray Radiography vs Computed Tomography: Choosing the Right Tool

Both methods use the same underlying physics — differential X-ray attenuation through varying material density and thickness — but they produce fundamentally different data. 2D radiography produces a single projected shadow image, compressing all internal features along the beam path into one flattened view. CT rotates the part (or the source/detector pair) through 360 degrees, capturing hundreds to thousands of projections that are mathematically reconstructed into a full 3D voxel model, allowing any internal plane to be sliced and examined independently.

For high-volume production of pressure die castings, 2D radiography remains the practical inline solution — it is fast, cheap, and adequate for catching gross defects above the reject threshold. CT is reserved for first-article inspection, PPAP submissions, periodic process capability audits, and any component where the internal geometry (undercuts, cored passages, complex rib intersections) makes projection-based imaging ambiguous. Many of our automotive and hydraulics clients run a hybrid model: 100% 2D screening on the production line, combined with CT sampling at a defined AQL frequency to validate that the 2D screening threshold is still catching what matters.

Technical Parameters That Determine Detection Sensitivity

The single biggest misconception in casting inspection procurement is that a higher-wattage machine automatically finds smaller defects. In practice, detection sensitivity is governed by the interaction of four parameters: tube voltage (kV), focal spot size, detector pixel pitch, and geometric magnification — and getting the combination wrong produces a scan that looks technically impressive but misses the defects that matter.

Geometric magnification — achieved by positioning the part closer to the source and further from the detector — is what allows microfocus CT systems to resolve sub-50-micron porosity in thin-wall aluminum parts even though the raw detector pixel pitch is coarser than that. This is why specifying a scan purely by machine wattage in a purchase order is insufficient; the specification must reference the required voxel resolution and the defect size that resolution must reliably resolve, tied to a statistical detection probability (POD) curve where possible.

Governing Standards and Acceptance Criteria

Acceptance decisions must be traceable to a published standard, not an inspector's judgment call, particularly for parts flowing into automotive PPAP or aerospace first-article inspection packages. The most commonly referenced standards in European casting supply chains are summarized below.

Severity level acceptance (Level 1 through Level 4 under ASTM E505, for example) is typically negotiated between the OEM and the foundry at the drawing stage, based on the criticality zone of the casting. A high-stress fillet radius near a bolt hole in a steering component will carry a far stricter severity level — often Level 1, permitting only minimal isolated porosity — than a low-stress cosmetic boss on the same part, which might be accepted at Level 3.

Applying Inspection Across Casting Processes and Complementary Fabrication

The porosity risk profile changes significantly depending on the casting process itself. High-pressure die casting, gravity die casting, sand casting, and investment casting each introduce characteristic defect signatures that CT operators learn to recognize quickly. Components produced withdie casting insert pins embedding steel threads in aluminum housingsrequire particular attention around the insert-to-aluminum interface, where differential shrinkage between the steel pin and the surrounding cast alloy frequently generates localized shrinkage porosity that 2D radiography can misread as a shadow artifact rather than a true void — CT resolves this ambiguity by isolating the interface plane in 3D.

Similarly, parts produced viacentrifugal casting for seamless tubes and cylindrical bushingspresent a different porosity signature: because centrifugal force pushes denser material outward during rotation, any residual gas porosity concentrates toward the inner bore, which is exactly the surface most critical for sealing and bearing fit in hydraulic bushings. CT sectioning along the radial axis is the only reliable way to quantify this gradient. In assemblies where cast housings are paired with formed enclosures or brackets, oursheet metal fabrication servicesteam coordinates directly with the casting inspection group so that mating tolerances and porosity-driven wall thickness variance are reconciled before final assembly, avoiding fit-up issues discovered only at the end of the line.

Defect Classification and Severity Levels in Practice

Critical stress zones — defined through FEA and mapped onto the casting drawing — typically carry a zero-porosity-above-threshold requirement regardless of overall casting quality level, while non-structural bosses and cosmetic surfaces are held to a more permissive standard. This zoned acceptance approach, increasingly standard in automotive and off-highway supply chains, requires the CT operator to have the part's stress map available at inspection time, not just the raw drawing.

For high-precision results,Receive a detailed quote within 24 hoursfrom Microns Hub.

Cost Drivers for X-Ray and CT Inspection Services

Inspection cost is driven by a combination of part size, required resolution, material density, and batch quantity, and understanding these drivers helps engineers specify inspection scope proportionate to actual risk rather than defaulting to maximum resolution on every part.

When ordering from Microns Hub, you benefit from direct manufacturer relationships that ensure superior quality control and competitive pricing compared to marketplace platforms that outsource inspection to unverified third parties. Our engineering team works directly with your drawing package and stress analysis to define zoned acceptance criteria before the first part is scanned, and our personalized technical support means porosity findings come with root-cause context from our own casting process data rather than a generic pass/fail report. This integration acrossour manufacturing services— from casting through inspection to secondary machining — is what allows us to close the loop between a detected defect and a corrected process parameter on the next production run.

Integrating CT Inspection into the Production Workflow

Effective use of X-ray and CT inspection is not a one-time gate at final acceptance — it is most valuable when embedded across three points in the production lifecycle. First, at process qualification, full CT scans of initial sample parts (typically 3-5 units per cavity) establish a porosity baseline and validate that gating, venting, and thermal balance in the die are producing acceptable internal quality before serial production begins. Second, at defined production intervals — often tied to statistical process control limits or a fixed AQL sampling plan per ISO 2859-1 — periodic CT or 2D X-ray sampling confirms the process remains in control as tooling wears and shot parameters drift. Third, at final customer acceptance, either 100% X-ray screening (for high-criticality automotive components) or a documented sampling plan with full traceability satisfies PPAP and IATF 16949 documentation requirements.

A frequent engineering mistake is treating inspection frequency as a fixed cost to be minimized rather than a risk-management lever tied to failure consequence. A hydraulic manifold operating at 350 bar and a decorative aluminum bezel share the same base alloy but warrant entirely different inspection intensity — conflating the two either wastes inspection budget on low-risk parts or, more dangerously, under-inspects a part where a missed shrinkage cavity could result in field failure. Establishing this risk-based inspection plan at the design review stage, before tooling is cut, avoids costly re-negotiation of acceptance criteria after parts are already in production.

Common Pitfalls and Best Practices

The most common technical error we see from external suppliers is specifying CT resolution in terms of machine capability rather than required defect detection size — a 225 kV system can produce a scan advertised as "high resolution" while still missing the 0,08 mm micro-porosity that matters in a thin-wall transmission housing, because the part-to-detector geometry was not optimized for magnification. A second frequent error is applying a single acceptance severity level across the entire part rather than zoning acceptance by actual stress distribution, which either drives unnecessary scrap in low-stress zones or masks risk in high-stress zones. Finally, treating CT data purely as a pass/fail gate — rather than feeding porosity location and volume data back into casting process parameters (gate velocity, venting, cooling channel layout) — wastes the single most valuable output of the inspection: the ability to close the loop and reduce defect rates at the source, rather than simply sorting good parts from bad ones after the fact.

Frequently Asked Questions

What is the smallest porosity size that X-ray or CT inspection can reliably detect?

Standard 2D radiography reliably detects porosity down to roughly 0,3-0,5 mm depending on wall thickness and material density, while microfocus CT with proper magnification setup can resolve individual pores as small as 0,02-0,05 mm in thin-wall aluminum sections. The actual detectable size depends heavily on voxel resolution, focal spot size, and the contrast between the void and surrounding material density, not solely on machine wattage.

Is CT scanning required for every safety-critical casting, or is 2D X-ray sufficient?

Most production programs use a hybrid approach: 2D X-ray for 100% or high-frequency screening on the production line, combined with periodic CT sampling for first-article qualification, PPAP submission, and process capability validation. Full CT on every part is typically reserved for very high-consequence applications such as aerospace structural castings or complex geometries where 2D projection cannot reliably localize internal defects.

Which standard should I reference for porosity acceptance criteria on aluminum die castings?

ASTM E505 is the most widely used reference radiograph standard for aluminum and magnesium die castings in automotive and industrial applications, while ASTM E155 applies to sand and permanent mold castings. European supply chains frequently reference EN 12681-1 as the baseline radiographic testing standard, with acceptance severity levels negotiated and documented on the engineering drawing.

Can X-ray inspection detect porosity near a steel insert embedded in an aluminum casting?

Yes, but with important caveats — the density difference between steel and aluminum creates strong contrast that can mask or mimic porosity artifacts at the interface in 2D projection. CT scanning is strongly recommended for insert-to-casting interfaces because it isolates the interface plane in three dimensions, distinguishing true shrinkage porosity from geometric shadow effects common with embedded threaded inserts.

How does porosity affect the pressure-tightness of cast hydraulic components?

Interconnected shrinkage porosity or clustered gas porosity can create continuous leak paths through a nominally solid wall section, causing hydraulic components to fail leak testing even when wall thickness and dimensions are within ISO 2768 tolerance. This is why pressure-tight castings such as hydraulic manifolds and transmission housings typically carry stricter zero-tolerance porosity zones around sealing surfaces and bolt bosses compared to non-pressurized structural areas of the same part.

What is the typical cost difference between 2D X-ray and full CT inspection per part?

2D X-ray inspection typically runs from €3 to €15 per part for standard automotive-scale aluminum castings, while full CT scanning ranges from €25 to over €150 per part depending on required resolution, part size, and reporting depth. Batch sampling per an AQL plan such as ISO 2859-1, rather than 100% CT scanning, is the most common way to control inspection cost while still maintaining statistical confidence in process quality.

Does CT scanning also verify dimensional accuracy, or only porosity?

Modern industrial CT systems, guided by VDI/VDE 2630 metrology guidelines, can simultaneously generate a full 3D point cloud for GD&T dimensional verification against the CAD model while also mapping internal porosity in the same scan. This dual-purpose capability makes CT particularly valuable for first-article inspection, where both internal soundness and external dimensional conformance to ISO 2768 tolerances must be confirmed from a single data set.