SOLAR RADIATION AND ITS MEASUREMENT - Environmental UV Radiation: Impact on Ecosystems and Human Health and Predictive Models - page 29

A great problem in monochromators is stray light. Even tiny fractions of
radiation scattered in the wrong direction by imperfections on the grating or by dust
particles give rise to false readings especially in the UV-B due to the enormous
difference in intensity between wavelengths below and above 300 nm. A tandem
arrangement of two monochromators (double monochromator) reduces stray light
considerably. A concise review on the main aspects for choosing a monochromator can
be found in the article by Domanchi and Gilchrist 8 . The monography by Kostkowski 9
gives a very comprehensive presentation of the subject.
Input optics . A very important component of a radiometer is the input optics. In
order to measure irradiance it is necessary that the angular response of the input optic
follows the cosine law as close as possible because at oblique angles of incidence the
projected sensor area is reduced by the cosine of the angle of incidence (Fig. 7). In order
to maintain cosine response usually quartz or Teflon ® diffusers or sometimes
integrating spheres are employed. Unfortunately all practical devices show more or less
large deviations from the ideal cosine response. This can lead to large measurement
errors especially at low solar elevations, if the data are not corrected for the cosine
mismatch adequately 10 . The direct and the diffuse irradiance components produce
different cosine errors. Therefore, their relative portions have to be determined
independently by ancillary instruments (e.g. sun tracker, shadow band radiometer, etc.)
or model calculations and the correction algorithm has to treat them differently.
Azimuth and polarisation dependency of the response may also cause measurement
errors. For easy handling the input optics is coupled to the entrance slit by means of a
quartz fibre bundle.

D

aperture area

F

A

projected area

Figure 7. Cosine response of the measurement aperture, A = F·cos(D).

Detector . After passing input optics, fibre bundle, entrance slit relay optics and
the dispersive grating a narrow waveband leaves the monochromator through the exit
slit and impinges on to the photodetector. At about 300 nm the absolute values of
terrestrial solar irradiance in a 1 nm spectral band are very small (|10 -5 W -2 m 2 nm -1 ).
Therefore, photomultiplier tubes are usually employed because of their excellent noise
characteristics. The selection of the photocathode type mainly depends on the waveband
of interest. Bi-alkali cathodes are first choice if only UV and blue regions are of primary
interest, a trialkali (S20) type covers the UV, visible and near infrared band but it has a
somewhat poorer performance than the bialkali cathode. The photomultiplier signal is