Scale a measured direct contribution by distance
When a single source behaves as a point, its directional output and receiving incidence angle stay fixed, and reflected or ambient light has been excluded, a compatible measured contribution can be scaled without first inferring candela:
E₂ = E₁(r₁/r₂)², with signed distance loss 2 log₂(r₂/r₁) stops. Positive loss means less direct light at the new distance; negative loss means a gain. Moving the receiver can change its angle or expose a different part of the beam, so those conditions must be checked separately.
For an artificial 800 lx direct contribution at 2 m, moving to 4 m on the same basis gives 800 × (2/4)² = 200.00 lx, a 2.0000-stop loss. The formula also scales a measured zero to zero, but that observation does not validate the point-source approximation. Nearby panels, strips, reflections and a total room reading generally do not fit this single-contribution model.
Directional intensity is the starting point
This estimate answers a specific question: how much direct illuminance does one point-like source contribute at one receiving point? The source’s luminous intensity must be known in the direction of that point, in candela (cd). A product’s total lumens do not supply that directional value. Peak candela is usable only when the receiving point lies in the direction to which that peak applies.
For an unobstructed ray, the model is E = I × cos α ÷ r², where E is illuminance in lux, I is the directional intensity in candela, r is the straight-line source-to-point distance in metres, and α is the angle between the incident ray and the surface normal. The normal is an imaginary line perpendicular to the receiving surface.
The inverse-square relationship spreads the source contribution with distance. The cosine projection accounts for the receiving surface being tilted relative to the ray. That reference discusses radiometric irradiance; the area projection also applies to photometric flux, without converting watts into lux.
If you have a supported manufacturer file, IES Candela Lookup can read or interpolate intensity inside its recorded angle range. Match the file coordinates to the installed luminaire and receiving point before using the value here.
Distance and tilt affect the result differently
These artificial cases hold directional intensity at 1,600 cd. Each case assumes the point-source model is appropriate and the path is unobstructed.
Doubling distance quarters the contribution because distance is squared. Tilting the receiving surface to 60° halves it because cos 60° = 0.5. Display values round to two decimal places; the calculation uses unrounded values.
For a point directly below a light above a horizontal bench, r can equal the vertical source-to-bench distance. For an offset point, r is the sloping distance. If moving that point also changes the ray through the luminaire’s distribution, use the candela for the new direction rather than holding peak intensity constant.
Incidence angle is not beam angle
At 0° incidence, the ray strikes squarely onto the surface. As incidence approaches 90°, the projected contribution approaches zero. The supported range stops below 90°; a ray striking the back of a surface belongs to a different geometry. A manufacturer’s full beam angle describes its distribution and must not be entered as this incidence angle. The beam geometry tool handles an ideal angle-defined circular outline separately.
Add direct contributions at the same receiving point
Multiple-source mode evaluates Σ Iᵢ cos αᵢ/rᵢ². Each row needs its own directional intensity, ray distance and incidence angle, all referring to the same point and receiving plane. Adding values at different locations would not describe the illuminance at either location.
For an artificial pair with normal incidence and 4 m ray distances, 1,600 cd contributes 100 lx and 800 cd contributes 50 lx, giving 150.00 lx. If the second ray instead arrives at 60° from the normal, its contribution is 25 lx and the sum is 125.00 lx. The first contribution does not change. This sum includes no reflected or daylight contribution unless separately represented by a model that actually supports it.
Where a point-source estimate stops being useful
A real luminaire has size. At distances where that size and its near-field distribution matter, a single point with directional intensity may be inadequate. A nearby luminous panel or strip needs an appropriate extended-source method or applicable photometric data; increasing the number of decimal places does not fix the model.
The result includes this source’s direct contribution only. Other lights, daylight, reflections and obstructions can change measured illuminance. It does not establish uniformity across the surface, glare, maintained lighting or compliance. To summarise suitably collected readings from an actual surface, use equal-area illuminance sample statistics.