Process Manufacturing

UV Shadowing: The Defect That Passes Inspection and Fails in the Field

Shadowed UV cure is the rare defect that looks perfect at the inspection station and fails months later at the customer. The cause is geometry, and no amount of lamp power fixes it.

Ultraviolet curing has a failure mode that behaves differently from every other process defect: it produces parts that pass visual and often functional inspection, then fail in service. The mechanism is simple. Photoinitiator requires photons. Any surface the light does not reach does not cure, and an under-cured adhesive or coating can be tacky, soft, or merely weak while appearing entirely normal. If the shadowed region is structural, the part ships and fails later.

Why the radiometer says everything is fine

Standard dose verification runs a flat radiometer puck through the tunnel on the conveyor. It measures the dose delivered to a flat surface at conveyor height, facing the lamp. That is a valid measurement of lamp condition and a poor proxy for what a three-dimensional part receives. Recessed features, vertical walls, undersides of flanges, and anything a fixture occludes see a fraction of the reference dose, and the difference is not a small correction. A surface angled well away from the source can receive a small share of the flat-panel reading.

How to find it before the customer does

Four checks, cheapest first

• Solvent rub on the suspect geometry. Under-cured material softens or transfers to the cloth where fully cured material does not. Crude, fast, and surprisingly discriminating.
• Attach the radiometer to the actual part or a replica fixture rather than running it flat, so the measurement reflects the geometry you ship.
• Destructive section on a sample from each new part number, examining cure through the depth of the joint rather than at the surface.
• Accelerated ageing on shadowed samples, since under-cure most often manifests as premature failure rather than immediate defect.

Every other defect gets worse the longer you look at it. Shadowed cure looks perfect at the inspection station and only tells the truth at the customer.

The fixes are mechanical, not electrical

Because the problem is line-of-sight, increasing lamp power rarely helps and increases the risk of over-curing the surfaces that were already fine. The effective interventions are physical: reorienting the part, adding a second pass at a different angle, introducing mirrors or secondary reflectors, moving to a lamp geometry that surrounds rather than overhangs, or, where the design allows, switching to a dual-cure chemistry with a secondary mechanism that does not need light. The last of these is the only reliable answer for genuinely enclosed joints.

Price the risk at quote, not after

A part with shadowed cure-critical geometry carries a real and quantifiable risk premium: additional verification, higher scrap during process development, and field failure exposure. That belongs in the quote for the part, identified during design review rather than discovered during production. A geometry assessment at quoting stage costs an engineering hour. The same assessment after a field failure costs the recall.

Use the UV shadowing risk calculator to score a part's exposed and occluded cure-critical area. Assess the geometry

Published 2026-08-08.