Radiometers, photometers, spectrometers and spectroradiometers are all used to measure light, but they operate in fundamentally different ways and have distinct advantages and disadvantages. This technical note describes the differences between these types of light meter. A note about nomenclature: we define a light meter as comprising both a photodetector and the electrical interface, whether that is a PC or a dedicated meter with its own display. Broadly, we can separate these instruments into two categories: radiometers and photometers; and spectrometers and spectroradiometers.
Radiometers and Photometers
Radiometers and photometers consist of a photodetector, an electrical readout and a calibration. The spectral response of a photodiode varies with the wavelength of the incident radiation. This means that 1 W/m2 of irradiance at 350nm in the UV will produce a different electrical current compared to 1 W/m2 of red light at 650nm.
The photodetector would typically be fitted with an optical filter which modifies the inherent spectral response of the sensor. For a radiometer to measure the amount of monochromatic light, you only need to know the calibrated spectral response of the photodiode at the relevant wavelengths. If the light source is anything other than monochromatic (which is the case for all light sources except lasers), a radiometer with bare photodiode cannot yield absolute radiometric measurements. A “radiometric” filter can be fitted to the photodiode that normalises the response over a limited wavelength range, for example ±10% from 450-950nm, which provides a partial solution.
A radiometer can also be equipped with a filter that limits or adapts the spectral response and these are typically used in safety or medical applications where certain wavelengths are more dangerous or more efficacious than others. Examples of this are measuring the safety of UVC light sources, measuring the erythemal effectiveness of a UVB source and measuring blue phototherapy lights used for the treatment of bilirubin (jaundice in new-borns).
A photometer is similar to a radiometer, but the photodiode is equipped with a special filter that modifies the spectral sensitivity of the detector/filter combination so as to match as closely as possible that of the human vision system, defined as the CIE spectral luminous efficiency for photopic vision (we refer to the “photopic” response). The photopic response of the human eye strongly favours green coloured light, with blue and red light being perceived as less intense. Thus, a photopic sensor ranks the brightness of light sources in close agreement to how the human vision system would perceive them.
For completeness, we should also mention colorimeters. A colorimeter is a type of photometer that combines three or four photodiodes each with a filter that closely matches the tristimulus response of the eye (the XYZ or XRXBYZ tristimulus colour matching functions). A colorimeter measures the emitted colour of the light source under test, and reports this as tristimulus values, as CIE chromaticity coordinates and as correlated colour temperature (CCT) values or as a dominant wavelength. The Y (green) filter of a colorimeter is the same as the photopic filter in a photometer, hence a “tristimulus” colorimeter can also function as a photometer.
Spectrometers and Spectroradiometers
Whereas radiometers, photometers and colorimeters employ photodiodes with special filters, a spectroradiometer measures what is called the spectral power of the light source, which is the amount of light at each wavelength. The spectral response of the spectroradiometer is calibrated at each wavelength, which therefore avoids the error that arises from measuring the irradiance of broadband light sources with unfiltered radiometers (or even radiometers equipped with normalising filters).
From the measured spectral power, the desired radiometric, photometric or colorimetric metrics can be calculated. A spectroradiometer avoids the potentially significant errors that photometers and colorimeters suffer from where the spectral response of the filtered sensor doesn’t exactly match that of the target observer, for example the photopic response of the eye. This is an important consideration; a high-quality photometer may have an average spectral mismatch (the f1’ factor) to the photopic observer of 5% over the 380-780nm visible light band, but can still yield errors of 50% or more when used to measure blue or red LEDs, for example.
A spectroradiometer is the name given to a spectrometer that is equipped with appropriate collection optics and an absolute calibration. A spectrometer is an optical instrument that employs a diffraction grating that physically separates the incident light into its component wavelengths. Each discrete wavelength is imaged onto an array photodetector, which allows for the recording of a spectral power distribution instantaneously.
While a spectrometer forms the basis of a spectroradiometer, the part that is at the heart of a spectrometer is the diffraction grating. Broadband light that reflects from, or is transmitted through, a diffraction grating is separated into its constituent wavelengths as a function of angle, according to Braggs law. This property can be exploited in one of two ways. In a monochromator, the diffraction grating is rotated, such that one wavelength at a time is directed towards a single element detector.
In a spectrometer, the diffraction grating is fixed and rather than use a single element detector, the dispersed light is imaged onto a linear or 2D photodiode array. Each element in the detector array is called a “pixel” and receives one narrow band of wavelengths. Whereas a monochromator samples the spectrum of light received one wavelength at a time, a spectrometer captures all wavelengths at the same time.
In general, spectrometers tend to be lower cost instruments compared to monochromators, and because they don’t have any moving parts, are easier to maintain. Spectrometers also perform measurements more quickly, and they can be much more compact, which is ideal for use in portable, handheld applications. Monochromators can utilise significantly more sensitive photodetectors (for example, photomultiplier tubes or PMTs) so can measure lower light levels. They also provide a wider dynamic range and can be designed to minimise errors such as stray light and detector noise, making a monochromator potentially a more accurate solution than a compact spectrometer.
Input Optics
The optical system that the photodetector or spectrometer is connected to defines the type of measurements that can be made. There are four measurement types that you can make with your light meter: the total radiant flux; the flux per unit area received, called the irradiance; the flux emitted per unit solid angle, called the radiant intensity; and the flux emitted per unit area per unit solid angle, called the radiance. For photometric measurements, the equivalent parameters are luminous flux, illuminance, luminous intensity and luminance.
Radiant flux is the total light emitted by the light source in all directions, expressed in Watts. The photometric equivalent is luminous flux, the unit of which is the lumen. The input optics used to measure total flux is the integrating sphere. An integrating sphere is a hollow spherical chamber with a diffuse, high reflectance coating on the interior. Any light entering the sphere is reflected from the wall of the sphere with equal radiance in all directions. A photodetector, or optical fibre coupling from the sphere into a spectrometer, receives a defined proportion of the total light and allows for the measurement of the total radiant or luminous flux.
Irradiance is the radiant flux received at a surface per unit area, with the unit of Watt/m2. The photometric equivalent is illuminance, which is the luminous flux received on a surface per unit area. The unit of illuminance in the lux (lx), which is equivalent to a lumen/m2. The input optic used for irradiance or illuminance measurements is a cosine diffusor placed directly in front of the photodetector or at the input to a spectrometer (or an optical fibre coupling into a spectrometer).
Radiant intensity is the radiant flux emitted per unit solid angle, with unit of Watt/steradian. The photometric equivalent is luminous intensity which is the luminous flux emitted per unit solid angle, the unit being a lumen/steradian, which is simplified to the candela (cd). Unlike all the other parameters, you cannot directly measure intensity. Rather, you first measure irradiance or illuminance at a specified distance and then calculate the intensity by invoking the inverse squared law. This calculation assumes that the measurement is performed in what is known as the far-field.
Finally, radiance is the radiant flux emitted from a source per unit solid angle, per unit area. The unit is Watt/steradian.m2. The photometric equivalent is luminance with the unit of lumen/sr.m2, which is simplified to candela/m2 (cd/m2). The input optics for luminance and radiance measurements are either a lens (non-contact) or tube system (contact) to ensure the measurement is being made on a known spot on the light emitting surface.
The Pros and Cons of Radiometers Versus Spectroradiometers
A light meter based upon a spectrometer is a much more versatile instrument and has the distinct advantage of applying the desired photometric or colorimetric observer function perfectly, whereas filtered photodetectors give a more approximate match. On the other hand, spectroradiometers tend to be more expensive and bulkier, and can require the use of a computer. However, some spectroradiometers can still be relatively affordable and some are also designed to not need a computer.
The main drawback of spectroradiometers can be their more limited dynamic range compared to a radiometer or photometer. This explains why filtered photometers and radiometers are relied upon as the primary sensor in nearly all professional, accredited lighting test laboratories and in National Measurement Institutes (NMIs).
