Campbell Scientific Kipp and Zonen CMP-Series Pyranometers - Manual (Page 21)

CMP/CMA series manual
Page 20
5. Calibration
5.1. Calibration principle
An ideal radiometer gives voltage output that is proportional to the absolute irradiance level. This
relationship can be expressed as a constant ratio called ‘sensitivity’ (S
ensitivity
).
The sensitivity figure of a particular radiometer is unique. It is determined in the manufacturer's
laboratory by comparison against a reference radiometer of similar type. The reference radiometer is
calibrated outdoors regularly at the World Radiation Centre (WRC) at Davos, Switzerland. The spectral
content of the laboratory calibration lamp differs from the outdoor solar spectrum at the World
Radiation Centre. However, this has no consequences for the transfer of calibration, because standard
and test radiometers have the same black coating and domes.
The supplied sensitivity figure is valid for the following conditions:
ƒ
An ambient temperature of +20°C.
ƒ For a horizontal radiometer and for a tilted radiometer.
ƒ Normal incident radiation of 500 W/m
2
.
For any other condition the sensitivity figure can be used within uncertainty bands given in the
specifications for each model.
A summary of calibration methods is also found in the WMO guide of 1996.
5.2. Calibration procedure at Kipp & Zonen
5.2.1. The calibration facility
The indoor calibration procedure, according to ISO 9847 Appendix III, is based on a side-by-side
comparison with a reference radiometer under a stable artificial sun. Kipp & Zonen uses a 150 W
Metal-Halide high-pressure gas discharge lamp with voltage stabilisation. Behind the lamp is a reflector
with a diameter of 16.2 cm. The reflector is 110 cm above the radiometers producing a vertical beam.
The irradiance at the radiometers is approximately 500 W/m².
To minimise stray light from the walls and the operator, the light is restricted to a small cone around the
two radiometers. The unknown radiometer 'a' and the reference radiometer 'b' are placed side by side
on a small table. The table can rotate to interchange the positions (1 and 2) of the radiometers. The
lamp is centred on the rotating axis of this table. Actually there is no normal incidence of the radiation,
but the angle of incidence is the same for both radiometers (3°) so this cannot give rise to errors. The
two radiometers are not levelled with the screws, but placed on their bases. The effect of the small
beam tilt is negligible (compare cos. 3° = 0.9986 and cos. 4° = 0.9976).
5.2.2. Calibration procedure
After illuminating for 30 seconds, the output voltages of both radiometers are integrated over 30
seconds. Next, both radiometers are covered. After 30 seconds the zero offset signals of both
radiometers are integrated, again over a period of 30 seconds. The irradiance at position 1 (radiometer
'a') may be slightly different from that at position 2 (radiometer 'b') due to asymmetry in the lamp optics,
etc. Therefore the radiometers are interchanged by rotating the table and the whole procedure is
repeated.