Photometric vs Radiometric LED Measurement: Lumens, Candela and Watts
Quick Answer: Match the Quantity to the Receiver
Use photometric quantities when the receiver is a human eye looking at the light, and radiometric quantities when the receiver is a surface, a material or a detector that responds to energy.
The rule is about the receiving element rather than the emitter. A red indicator judged by an operator belongs in photometric units even though the emitter is the same device that would be specified radiometrically if it were delivering energy into a curing process or onto a photodiode.

Two Ways of Weighting the Same Optical Power
Both systems start from the same physical quantity, which is the optical power the emitter radiates. What differs is what happens next.
Radiometric quantities report that power directly, in watts, and they treat all wavelengths as equivalent. A measurement of radiant flux counts the energy that crosses a surface regardless of whether a person could see it.
Photometric quantities apply a weighting function before reporting, which describes how strongly a standard observer responds to each wavelength and which peaks in the green part of the visible spectrum. Two emitters with identical radiant power can therefore have very different luminous values, which is the entire point of the system.
Because the weighting function is defined over the visible band, it loses meaning outside it. That is the practical boundary between the two systems, and it is the reason a UV emitter is never specified in lumens.
The Unit Map: What Each Quantity Measures
The table below pairs the quantities that appear most often in reading LED specifications, together with the receiver each one assumes.
| Photometric | Radiometric counterpart | What the quantity describes |
|---|---|---|
| Luminous flux, in lumens | Radiant flux, in watts | Total output leaving the emitter in all directions |
| Luminous intensity, in candela | Radiant intensity, in watts per steradian | Output concentrated into a direction |
| Illuminance, in lux | Irradiance, in watts per square metre | Output arriving at a surface |
| Luminance, in candela per square metre | Radiance, in watts per steradian per square metre | Output leaving or arriving at a surface in a direction |
Reading the table downwards gives the geometry of each pair, and reading it across gives the two systems. Most confusion in LED datasheets comes from mixing rows rather than systems, for example comparing a directional figure from one supplier with a total-output figure from another.
Within the photometric side there is a second distinction that catches teams out regularly, and it is worth separating intensity from flux before comparing suppliers at all. Our explanation of luminous intensity compared with luminous flux sets out where that difference comes from.
Why Lumens Stop Being Useful Outside the Visible Band
The weighting function is close to zero in the ultraviolet and in the near infrared, so a luminous value for those emitters either stays near zero or depends on a tiny amount of visible spill light that has nothing to do with the intended function.
This is not a technicality. A UV curing system is judged by the energy delivered to the coating, and an infrared emitter for sensing is judged by the radiant intensity that reaches a detector. In both cases the right specification is radiometric, even though the same production line may also make visible indicators that are correctly specified in photometric units.
For invisible emitters the useful figures are radiant flux and radiant intensity, and for a system the useful figure is the irradiance arriving at the working plane. A worked example of how one of those is obtained in practice is set out in our guide to IR LED radiant intensity, which shows why the measurement geometry matters as much as the emitter.

Converting Between Systems: What Is and Is Not Valid
A conversion between the two systems is possible only when the spectrum is known in detail, because the weighting function has to be applied to the actual emission rather than to a nominal colour. Two emitters with the same nominal colour can hold different spectral distributions and therefore produce different luminous values for identical radiant power.
Two consequences follow. A general conversion factor quoted without a spectrum should not be trusted for a specification, and a radiometric figure cannot be turned into a luminous claim for a UV emitter at all, because the weighting does not apply in that band.
The reverse direction is more forgiving but still conditional. A visible emitter can be specified photometrically and then checked radiometrically for a non-visual purpose, provided the spectrum and the measurement condition are recorded. Our guide to measuring UV LED optical power shows the same principle applied inside a single system, where geometry and condition determine whether two readings can be compared.
The correct engineering move is therefore to keep the two systems separate in the document and to state each number with its own condition, rather than to convert one into the other and publish a single figure.
Writing the Right Quantity Into a Specification
The specification follows four questions in sequence, and each answer constrains the next.
- Identify the receiver: an operator looking at the light, a surface receiving energy, or a detector converting it into a signal.
- Choose the system from the receiver, and choose the row from the geometry the requirement describes.
- State the measurement condition: drive current, temperature, distance and acceptance angle, because a quantity without a condition cannot be verified.
- Name the instrument class and the acceptance rule, so that a disputed lot can be re-measured rather than argued about.
Two common failures appear at this stage. The first is a specification that quotes a luminous value for a device whose function is invisible, which produces a number nobody can verify. The second is a specification that quotes the right system but omits the geometry, so that two readings taken under different conditions are treated as comparable.
Under the second failure, a lot can be rejected for a difference created by the measurement rather than by the part, which is expensive in both directions. Keeping the condition attached to the number removes that risk, and it also makes the specification usable by a different laboratory later.
Frequently Asked Questions
How do photometric vs radiometric LED measurement options compare for OEM designs?
Photometric quantities describe the visible effect of light on a standard observer, while radiometric quantities describe the physical power in the beam. A visible indicator is normally specified photometrically, and an emitter whose job is to deliver energy is normally specified radiometrically.
Which photometric vs radiometric LED measurement choice fits an industrial application?
If the output is judged by a person looking at it, use photometric quantities such as luminous intensity and luminous flux. If the output drives a reaction, a material or a detector outside the visible band, use radiometric quantities such as radiant flux and radiant intensity.
How should engineers validate a photometric vs radiometric LED measurement decision?
Check what the receiving element actually responds to, then confirm that the specification quotes a quantity that element can use. Where both matter, quote each one separately rather than converting between them.
Can lumens be converted to watts for a UV LED?
No, because a UV emitter sits far outside the region where the photometric weighting applies, so a meaningful luminous value does not exist. Specify the UV output in radiant power or irradiance instead.
Why do two datasheets use different units for the same part?
Different suppliers target different readers, and a visible indicator is easier to present in candela or millicandela. When two documents disagree, compare the underlying test condition rather than the headline number.

Conclusion: A Specification You Can Defend
Photometric vs radiometric LED measurement is settled by one question: what receives the light. Answer that first and the system, the quantity and the unit follow from it, and the specification stops containing numbers that cannot be compared or verified.
The approach also makes Optical Parameters & Measurement work repeatable, because LED optical measurement and LED datasheet parameters are then recorded with their conditions attached. That is what allows a second laboratory, a second supplier or a second production site to reproduce the same result rather than reinterpret it.
If you are writing an optical requirement for a visible, UV LED products or infrared emitter and want the measurement condition confirmed before the specification is issued, send the requirement and the intended receiver through the inquiry page.
LED Moisture Sensitivity Level MSL: Storage, Baking and Reflow Basics