Why replace the risk score with geometry?

Because the score could not be checked, compared, or acted on. It was a weighted average of three numbers the user invented, published to one decimal. Nothing anchored 55 rather than 60, two engineers would not agree, and 37.5 corresponded to no physical quantity. Averaging guesses does not cancel their error; it hides it behind a precise-looking figure. Opening width and depth are measurable with a rule, and the view factor between them is standard radiative geometry.

How accurate is the view-factor estimate?

It is the right order of magnitude and the right shape, not a measurement. It is exact for its stated case. A point below a circular opening under a diffuse hemispherical source with absorbing walls, and real recesses depart from that in both directions: a single off-axis lamp does worse, reflective walls do better. Use it to decide whether a feature is comfortable, marginal or hopeless, then measure the marginal ones.

Can I fix a shadowed recess with more lamp power?

Only where the shortfall is small. Because dose at the floor scales with the view factor, doubling the floor's dose means doubling it everywhere, so a 27% shortfall costs 37% more on every open face: often tolerable. A feature at 2:1 depth-to-width needs something like twenty times the surface dose, which is not available and would destroy the rest of the part long before it arrived.

What actually works for deep features?

Change the geometry or change the chemistry. Cure before the feature is enclosed; tilt or rotate the part so the floor gets a view of the source; add a second head at a complementary angle; or use a dual-cure or moisture-cure adhesive in the recess and let UV handle the open faces. Reflective fixturing helps at the margins. All of these beat trying to force light down a hole.

Why does depth matter so much more than width?

The view factor is R² ÷ (R² + h²), so depth enters squared in the denominator while the opening enters squared in both. Once depth exceeds the opening radius the denominator is dominated by depth and the fraction falls roughly as 1/h². Opening a feature up helps, but not nearly as fast as making it shallower.

Does oxygen inhibition make this worse?

Usually, and the two compound. A recess holds still air, so oxygen at the surface of the resin is replenished rather than consumed, and surface cure competes with it. That effectively raises the dose the floor needs at the same time as the geometry cuts the dose it receives. Where a recess must cure, inerting it or flooding with nitrogen attacks the half of the problem that lamp power cannot.

UV Curing calculator

UV Shadowing Calculator: Does the Bottom of the Recess Cure?

Estimate how much UV actually reaches the bottom of a pocket, slot or bore, and whether it is enough to cure. Enter the opening width, the depth down to the surface being cured, the dose measured on a flat face of the same part, and the material's requirement. The calculator applies the standard view factor from a point to the opening above it and returns the fraction of the surface dose that reaches the floor, the dose that implies, its margin against requirement, and what the flat surface would have to receive before the recess cleared. This replaces the earlier page, which asked for three subjective scores out of 100 and averaged them.

What this calculator does

Estimate the UV dose reaching the bottom of a recess from its geometry, using the standard view factor, and say whether it clears the material's requirement.
Use it for deciding at design review whether a pocket will cure, explaining why a bore or slot fails while the rest of the part passes, testing whether more lamp power can fix a shadowed feature, setting a depth-to-width limit for UV-cured features on a family of parts, screening parts before committing to a dose-mapping trial.
Estimate the UV dose reaching the bottom of a recess from its geometry, using the standard view factor, and say whether it clears the material's requirement.

Formula used

Opening radius R = opening width ÷ 2
View factor F = R² ÷ (R² + depth²)
Dose reaching the recess floor = flat-surface dose × F
Recess margin (%) = (floor dose − required) ÷ required × 100
Flat-surface dose needed = required ÷ F

Inputs explained

Recess opening width: Clear width of the opening at the surface. For a slot use the narrow dimension; for a hole, the diameter.
Depth to the surface being cured: Distance from the opening plane down to the surface that has to cure: the floor of the pocket, not the full part thickness.
Dose on the flat surface: Measured dose on an unobstructed face of the same part in the same pass. This is what the recess floor is a fraction of.
Required cure dose: What the coating or adhesive needs, in the band being measured.

How to use the result

Best suited to deciding at design review whether a pocket will cure, explaining why a bore or slot fails while the rest of the part passes, testing whether more lamp power can fix a shadowed feature, setting a depth-to-width limit for UV-cured features on a family of parts, screening parts before committing to a dose-mapping trial.
This is a geometric ESTIMATE, not a measurement. A radiometer probe or a dose-indicating strip in the actual recess is the measurement, and it is worth taking before anything expensive is decided. Assumes a diffuse hemispherical source. A collimated array or a single lamp position off to one side can deliver far less than the view factor suggests, since it fills only part of the hemisphere. Ignores wall reflections, which work the other way: a bright, shallow recess can beat this estimate. Says nothing about oxygen inhibition, which is usually worse in a still recess than on an open face and can prevent surface cure even where the dose arrives. Covers one simple opening over one point. Complex geometries, undercuts and stacked features need a real map.

Common questions

Why replace the risk score with geometry? Because the score could not be checked, compared, or acted on. It was a weighted average of three numbers the user invented, published to one decimal. Nothing anchored 55 rather than 60, two engineers would not agree, and 37.5 corresponded to no physical quantity. Averaging guesses does not cancel their error; it hides it behind a precise-looking figure. Opening width and depth are measurable with a rule, and the view factor between them is standard radiative geometry.
How accurate is the view-factor estimate? It is the right order of magnitude and the right shape, not a measurement. It is exact for its stated case. A point below a circular opening under a diffuse hemispherical source with absorbing walls, and real recesses depart from that in both directions: a single off-axis lamp does worse, reflective walls do better. Use it to decide whether a feature is comfortable, marginal or hopeless, then measure the marginal ones.
Can I fix a shadowed recess with more lamp power? Only where the shortfall is small. Because dose at the floor scales with the view factor, doubling the floor's dose means doubling it everywhere, so a 27% shortfall costs 37% more on every open face: often tolerable. A feature at 2:1 depth-to-width needs something like twenty times the surface dose, which is not available and would destroy the rest of the part long before it arrived.
What actually works for deep features? Change the geometry or change the chemistry. Cure before the feature is enclosed; tilt or rotate the part so the floor gets a view of the source; add a second head at a complementary angle; or use a dual-cure or moisture-cure adhesive in the recess and let UV handle the open faces. Reflective fixturing helps at the margins. All of these beat trying to force light down a hole.
Why does depth matter so much more than width? The view factor is R² ÷ (R² + h²), so depth enters squared in the denominator while the opening enters squared in both. Once depth exceeds the opening radius the denominator is dominated by depth and the fraction falls roughly as 1/h². Opening a feature up helps, but not nearly as fast as making it shallower.

Last reviewed 2026-08-25.