Is a 25% time margin the same as a 25% dose margin?

Here, yes, and that is worth being explicit about because the earlier version was not. Dose is irradiance times time, so at a fixed irradiance multiplying the time by 1.25 multiplies the dose by 1.25. They stop being the same number the moment irradiance changes, which is exactly what happens as a lamp ages, and why the margin the line actually has is worth tracking separately from the margin you set.

How much margin should I set?

Enough to cover the lamp's decline between measurements. Mercury lamps lose output continuously rather than in steps, so if you measure quarterly you need enough margin to survive a quarter of decline; 20–30% is the usual answer. A line with no margin is undercured the day after commissioning and nobody finds out until a customer does.

What if I cannot slow the line down?

Then the other two terms have to move. More irradiance means a closer working distance, a cleaner reflector and quartz, or a newer lamp, all cheap before they mean a bigger lamp. More cure zone means another lamp station, which the conveyor speed page shows scales the achievable speed directly. Failing both, the material has to change, and a formulation with a lower dose requirement is a real option people forget is available.

Why does the page ask what my line does today?

Because the requirement alone does not tell you to do anything. Everyone opening this page has a line with a fixed cure zone and a belt speed, so the exposure available is already decided; what they need is the gap between that and the requirement. the earlier version computed the requirement and stopped, which meant the user had to do the comparison themselves, and doing it in your head is where the sign errors live.

My exposure is fine but parts are still failing. What now?

Check the band first: a correct-looking dose in the wrong band is the most common cause. Then check depth: if the coating is pigmented or thick, the surface can pass while the bottom never cured, and more time at the surface will not fix attenuation. Then check the surface itself for tack, which is oxygen inhibition rather than dose. Those three account for most cure failures on lines whose numbers look right.

Does the margin apply to the upper limit too?

No, and that is a real gap in this page. The margin here is one-sided: it protects against undercure. Overcure is a real limit for some coatings. Embrittlement, yellowing, and on heat-sensitive substrates the infrared that accompanies a mercury lamp distorts parts long before the UV harms them. The cure window page handles the two-sided version and the heat load page sizes the thermal side.

UV Curing calculator

UV Exposure Time Calculator with Undercure Margin

Work out how long a material needs under UV, and whether your line already gives it. Enter the cure dose from the datasheet, the irradiance a radiometer reads at the part, the safety margin you want above the bare requirement, and the exposure the line delivers today. The calculator returns the exposure needed with margin, the bare exposure without it, the dose the line actually delivers, the margin it actually has, and the shortfall if there is one. The requirement on its own is only half an answer; the gap is the part you can act on.

What this calculator does

Work out the exposure a material needs at the measured irradiance, and compare it against the exposure the line currently gives.
Use it for deciding whether a new material will run on an existing line, converting a datasheet requirement into a line setting, tracking how much margin a line has lost as its lamp aged, checking a proposed speed increase against the cure requirement, diagnosing an adhesion complaint by reconstructing the exposure the part got.
Work out the exposure a material needs at the measured irradiance, and compare it against the exposure the line currently gives.

Formula used

Bare exposure (s) = required dose ÷ measured irradiance
Exposure needed with margin = bare exposure × (1 + safety margin ÷ 100)
Dose the line delivers = measured irradiance × exposure available
Margin the line has = (exposure available ÷ bare exposure − 1) × 100
Exposure still needed = max(exposure with margin − exposure available, 0)

Inputs explained

Required cure dose from datasheet: The dose the material specification calls for, in a named spectral band. A dose figure without a band is not actionable.
Measured irradiance at part surface: What a radiometer reads at the part, at working distance, in the same band as the dose target.
Undercure safety margin: Headroom above the bare requirement. At fixed irradiance this is also the dose margin, because dose is proportional to time.
Exposure the line gives today: Seconds a point on the part is actually illuminated at the current setting. Cure-zone length divided by belt speed, or the timer on a batch cell.

How to use the result

Best suited to deciding whether a new material will run on an existing line, converting a datasheet requirement into a line setting, tracking how much margin a line has lost as its lamp aged, checking a proposed speed increase against the cure requirement, diagnosing an adhesion complaint by reconstructing the exposure the part got.
Surface dose only. Nothing here describes what reaches the bottom of a pigmented or thick film. Does not model oxygen inhibition, which can leave a tacky surface at a dose the bulk cured at. Assumes a single pass; a multi-lamp tunnel accumulates dose across stations and belongs on the multi-lamp page. A process setting, not cure evidence. Gel content or a rub test is evidence. Says nothing about the upper limit of the window: overcure, yellowing and substrate heat are separate constraints.

Common questions

Is a 25% time margin the same as a 25% dose margin? Here, yes, and that is worth being explicit about because the earlier version was not. Dose is irradiance times time, so at a fixed irradiance multiplying the time by 1.25 multiplies the dose by 1.25. They stop being the same number the moment irradiance changes, which is exactly what happens as a lamp ages, and why the margin the line actually has is worth tracking separately from the margin you set.
How much margin should I set? Enough to cover the lamp's decline between measurements. Mercury lamps lose output continuously rather than in steps, so if you measure quarterly you need enough margin to survive a quarter of decline; 20–30% is the usual answer. A line with no margin is undercured the day after commissioning and nobody finds out until a customer does.
What if I cannot slow the line down? Then the other two terms have to move. More irradiance means a closer working distance, a cleaner reflector and quartz, or a newer lamp, all cheap before they mean a bigger lamp. More cure zone means another lamp station, which the conveyor speed page shows scales the achievable speed directly. Failing both, the material has to change, and a formulation with a lower dose requirement is a real option people forget is available.
Why does the page ask what my line does today? Because the requirement alone does not tell you to do anything. Everyone opening this page has a line with a fixed cure zone and a belt speed, so the exposure available is already decided; what they need is the gap between that and the requirement. the earlier version computed the requirement and stopped, which meant the user had to do the comparison themselves, and doing it in your head is where the sign errors live.
My exposure is fine but parts are still failing. What now? Check the band first: a correct-looking dose in the wrong band is the most common cause. Then check depth: if the coating is pigmented or thick, the surface can pass while the bottom never cured, and more time at the surface will not fix attenuation. Then check the surface itself for tack, which is oxygen inhibition rather than dose. Those three account for most cure failures on lines whose numbers look right.

Last reviewed 2026-08-25.