The candela (/kænˈdɛlə/ or /kænˈdlə/; symbol: cd) is the SI base unit of luminous intensity; that is, luminous power per unit solid angle emitted by a point light source in a particular direction. Luminous intensity is analogous to radiant intensity, but instead of simply adding up the contributions of every wavelength of light in the source's spectrum, the contribution of each wavelength is weighted by the standard luminosity function (a model of the sensitivity of the human eye to different wavelengths).[5] A common candle emits light with a luminous intensity of roughly one candela. If emission in some directions is blocked by an opaque barrier, the emission would still be approximately one candela in the directions that are not obscured.

The word candela means candle in Latin.


Like most other SI base units, the candela has an operational definition—it is defined by a description of a physical process that will produce one candela of luminous intensity. Since the 16th General Conference on Weights and Measures (CGPM) in 1979, the candela has been defined as:[6]

The candela is the luminous intensity, in a given direction, of a source that emits monochromatic radiation of frequency 540×1012 hertz and that has a radiant intensity in that direction of 1683 watt per steradian.

The definition describes how to produce a light source that (by definition) emits one candela. Such a source could then be used to calibrate instruments designed to measure luminous intensity.

The candela is sometimes still called by the old name candle,[7] such as in foot-candle and the modern definition of candlepower.


The frequency chosen is in the visible spectrum near green, corresponding to a wavelength of about 555 nanometres. The human eye is most sensitive to this frequency, when adapted for bright conditions. At other frequencies, more radiant intensity is required to achieve the same luminous intensity, according to the frequency response of the human eye. The luminous intensity for light of a particular wavelength λ is given by

where Iv(λ) is the luminous intensity in candelas, Ie(λ) is the radiant intensity in W/sr and is the standard luminosity function (photopic). If more than one wavelength is present (as is usually the case), one must sum or integrate over the spectrum of wavelengths present to get the total luminous intensity.


A common candle emits light with roughly 1 cd luminous intensity. A 25 W compact fluorescent light bulb puts out around 1700 lumens; if that light is radiated equally in all directions, it will have an intensity of around 135 cd (135 lumens/steradian). Focused into a 20° beam, it will have an intensity of around 18 000 cd.

The luminous intensity of light-emitting diodes is measured in millicandelas (mcd), or thousandths of a candela. Indicator LEDs are typically in the 50 mcd range; "ultra-bright" LEDs can reach 15 000 mcd (15 cd), or higher.


Prior to 1948, various standards for luminous intensity were in use in a number of countries. These were typically based on the brightness of the flame from a "standard candle" of defined composition, or the brightness of an incandescent filament of specific design. One of the best-known of these was the English standard of candlepower. One candlepower was the light produced by a pure spermaceti candle weighing one sixth of a pound and burning at a rate of 120 grains per hour. Germany, Austria and Scandinavia used the Hefnerkerze, a unit based on the output of a Hefner lamp.[8]

It became clear that a better-defined unit was needed. The Commission Internationale de l'Éclairage (International Commission on Illumination) and the CIPM proposed a “new candle” based on the luminance of a Planck radiator (a black body) at the temperature of freezing platinum. The value of the new unit was chosen to make it similar to the earlier unit candlepower. The decision was promulgated by the CIPM in 1946:

The value of the new candle is such that the brightness of the full radiator at the temperature of solidification of platinum is 60 new candles per square centimetre.[9]

It was then ratified in 1948 by the 9th CGPM which adopted a new name for this unit, the candela. In 1967 the 13th CGPM removed the term "new candle" and gave an amended version of the candela definition, specifying the atmospheric pressure applied to the freezing platinum:

The candela is the luminous intensity, in the perpendicular direction, of a surface of 1 / 600 000 square metre of a black body at the temperature of freezing platinum under a pressure of 101 325 newtons per square metre.[2]

In 1979, because of the difficulties in realizing a Planck radiator at high temperatures and the new possibilities offered by radiometry, the 16th CGPM adopted the modern definition of the candela.[2] The arbitrary (1/683) term was chosen so that the new definition would exactly match the old definition. Although the candela is now defined in terms of the second (an SI base unit) and the watt (a derived SI unit), the candela remains a base unit of the SI system, by definition.[2]

SI photometric light units

SI photometry quantities
Luminous energyQv lumen secondlm⋅sTJ Units are sometimes called talbots.
Luminous flux / luminous powerΦv lumen (= cd⋅sr)lmJ Luminous energy per unit time.
Luminous intensityIvcandela (= lm/sr)cdJ Luminous power per unit solid angle.
LuminanceLvcandela per square metrecd/m2L−2JLuminous power per unit solid angle per unit projected source area. Units are sometimes called nits.
IlluminanceEvlux (= lm/m2)lxL−2JLuminous power incident on a surface.
Luminous exitance / luminous emittanceMvluxlxL−2JLuminous power emitted from a surface.
Luminous exposureHvlux secondlx⋅sL−2TJ
Luminous energy densityωvlumen second per cubic metrelm⋅s⋅m−3L−3TJ
Luminous efficacyη lumen per wattlm/WM−1L−2T3JRatio of luminous flux to radiant flux or power consumption, depending on context.
Luminous efficiency / luminous coefficientV1
See also: SI · Photometry · Radiometry
  1. Standards organizations recommend that photometric quantities be denoted with a suffix "v" (for "visual") to avoid confusion with radiometric or photon quantities. For example: USA Standard Letter Symbols for Illuminating Engineering USAS Z7.1-1967, Y10.18-1967
  2. Alternative symbols sometimes seen: W for luminous energy, P or F for luminous flux, and ρ or K for luminous efficacy.
  3. "J" here is the symbol for the dimension of luminous intensity, not the symbol for the unit joules.

Relationships between luminous intensity, luminous flux, and illuminance

If a source emits a known luminous intensity Iv (in candelas) in a well-defined cone, the total luminous flux Φv in lumens is given by

Φv = Iv 2π [1 − cos(A/2)],

where A is the radiation angle of the lamp—the full vertex angle of the emission cone. For example, a lamp that emits 590 cd with a radiation angle of 40° emits about 224 lumens. See MR16 for emission angles of some common lamps.[2][2]

If the source emits light uniformly in all directions, the flux can be found by multiplying the intensity by 4π: a uniform 1 candela source emits 12.6 lumens.

For the purpose of measuring illumination, the candela is not a practical unit, as it only applies to idealized point light sources, each approximated by a source small compared to the distance from which its luminous radiation is measured, also assuming that it is done so in the absence of other light sources. What gets directly measured by a light meter is incident light on a sensor of finite area, i.e. illuminance in lm/m2 (lux). However, if designing illumination from many point light sources, like light bulbs, of known approximate omnidirectionally-uniform intensities, the contributions to illuminance from incoherent light being additive, it is mathematically estimated as follows. If ri is the position of the i-th source of uniform intensity Ii, and â is the unit vector normal to the illuminated elemental opaque area dA being measured, and provided that all light sources lie in the same half-space divided by the plane of this area,

In the case of a single point light source of intensity Iv, at a distance r and normally incident, this reduces to