|Solar irradiance spectrum above atmosphere and at surface. Extreme UV and X-rays are produced (at left of range shown) but comprise very small amounts of the Sun's total output power|
At Earth's distance from the sun, an average of about 1353 W/m2 of power reaches the top of the Earth/atmosphere system and is called the solar irradiance constant. About 30% is reflected back to space, mostly by clouds. About 70% is absorbed by Earth's surface and atmosphere and is then re-radiated in the form of thermal radiation. This interaction keeps the Earth in radiative balance, as required by basic physical laws, and maintains the Earth's average surface temperature at about 16°C.
The amount of sunlight reaching a horizontal surface at the Earth's surface is called insolation and is measured in units of watts per square meter (W/m2). Insolation (power), or insolation accumulated over some amount of time (energy), is affected by several factors, including time of day, time of year, latitude, cloud cover, moisture in the air, and air quality. Around midday on a summer day in temperate climates, roughly 1 kW/m2 of power reaches Earth's surface.
Similarly with solar illuminance, approximately 127 500 lm/m2 of illuminance reaches the top of the Earth/atmosphere system and is called the solar illuminance constant. Solar illuminance at a location is generally obtained from the product of the luminous efficacy and irradiance for that location. Luminous efficacy on the surface of the earth is a function of atmospheric conditions including air mass, elevation of the sun, cloud cover, precipitable water, and atmospheric turbidity. The terrestrial efficacy used as a base reference which applies on the outskirts of the earth’s atmosphere may be calculated using the terrestrial irradiance constant of 1353 W/m2 and the terrestrial solar illuminance constant of 127 500 lm/m2. The resulting terrestrial efficacy is 94.23 lm/W, and would apply at all points on the planet if there were no atmosphere. While luminous efficacy is a function of the elevation of the sun and atmospheric conditions and will vary daily, levels of the order of 100 to 130 lm/W have been published. Published measured levels of global illuminance include 110 000 lm/m2 in Athens, and 120 000 lm/m2 at an altitude of 1715 m at Johannesburg, in South Africa.
The objective of the detector is to gather solar illumination data for meteorological purposes and not solar radiation data for solar energy purpose. Therefore, whilst absolute values for illuminance are not essential a uniform response to illuminance over the range of solar zenith angles is.
For the purposes of solar power measurement the solar radiation is measured and reported in Watts per square metre (w/m2 and is covered in more detail in the section on Solar Energy. However for meteorological purposes there are serveral weather features that may be derived from a measurement of solar illumination which is measured and reported in Lux (lx) or Lumen per square metre (lm/m2).
What is the light level in Lux i.e. how bright is it? In 2003, WMO defined sunshine duration as
the period during which direct solar irradiance exceeds a threshold value of 120 watts per square meter
(W/m2). This value is equivalent to the level of solar irradiance shortly after sunrise or shortly before
sunset in cloud-free conditions.
This is generally reported as hours of bright sunshine per day. Sunshine duration is the length of time that the ground surface is irradiated by direct solar radiation (i.e.,sunlight reaching the earth's surface directly from the sun) and is of particlar interest to gardeners and of course holidaymakers. The hours of sunshine are conventionally measured using a Campbell-Stokes sunshine recorder which concentrates sunlight through a glass sphere onto a recording card placed at its focal point. The length of the burn trace left on the card represents the sunshine duration. However, the correct interpretation of these traces is notoriously difficult, especially interpreting the results from a day of intermittent clouds.
Radiation data recorded at intervals of a minute or so provide a record of cloud cover during the day and it has been shown, in peer-reviewed
[Duchon and O'Malley, 1999], that it is possible to use radiation data to classify cloud amounts and types even when individual measurements of
radiation are not highly accurate.
Instruments used to measure insolation are called pyranometers (from the Greek "pyr" (fire) and "ano" (sky). Ideally, pyranometers should measure all the radiation from the sun, across the entire electromagnetic spectrum (broadband radiation). Pyranometers based on thermopile detectors, which are collections of thermocouples embedded in special materials, closely approach this ideal. However, such instruments cost several thousand dollars. Much less expensive pyranometers use miniature silicon solar cell detectors. Commercial versions of these instruments still cost from $200 to several hundred dollars.
The downside of solar cell-based pyranometers is that their response to solar radiation is strongly peaked in the near infrared and does not extend across the entire solar. The graph at the right shows the response of their solar cell-based PYR pyranometer as a function of wavelength. Light visible to humans falls in the range from 400 to 700 nanometers. The near-IR peak is near a strong atmospheric water vapor absorption band. So, these devices are sensitive to changes in water vapor as well as sunlight.
Despite their shortcomings, solar-cell based pyranometers are widely used for meteorological, environmental, and agricultural monitoring and to characterize the potential of sites to produce solar power.
For the hobbyist a low cost alternative to a pyranometer is a photo-diode such as the SFH203P. As can be seen from the chart to the right, this diode has a much broader spectral response making it emminantly suitable for the measurement of solar illumination. This is the sensor used in the HobbyBoards radiation sensor.
There is one other important design consideration for even the simplest pyranometer. Ideally, pyranometer detectors should respond to direct sunlight in proportion to the cosine of the zenith angle of the sun. When the sun is directly overhead we define the "normalized" response as 1. As the zenith angle increases, the normalized detector response to direct sunlight should decrease as the cosine of the zenith angle. Real detectors do not have a perfect cosine response so some kind of sunlight diffuser can be placed over the detector to improve their response, Teflon® is often used because it has good spectral transmission properties, is very stable, and is relatively unaffected by long-term exposure to sunlight. For my own system I use a celluloid table-tennis ball as a diffuser.
The HobbyBoards radiation detector measures the current produced by a SFH203P photo-diode using the current sensing channel of the DS2438 battery monitoring slave device with a 390Ω.load resistor. If the DS2438 status register is set appropriately the current from the photo-diode will be integrated into the current accumultor and stored in the eprom memory. This may prove useful in the analysis of cloud types. I noticed that the current channel of the DS2438 slave device used in the HobbyBoards anamometer controller board was not being used and so I modified circuit to measure radiation as shown in the schematic below.
This modification enabled the radiation sensor to be installed adjacent to the anamometer and lightning detector all of which need to be installed in an exposed location.
Finding a suitable location for the installation of a pyranometer is very difficult since such locations must have day long sun exposure with no shadows. This is very difficult to find in an urban environment but the best chance of success is to mount the pyranometer at the the highest accessible point. In my case the radiation sensor is mounted on top of the lightning detector attached to the anemometer mast.
In order to maintain the cosine response care must be taken during installation to ensure that the sensing element of the detector is horizontal. This is no easy matter since the data sheet of the SFH 203P specifically states that the base of the detector is not flat.