Appendix b. details about using the heater, B.1 general information – Campbell Scientific CNR1 Net Radiometer User Manual
Page 31

Appendix B. Details about Using the 
Heater 
Whenever the heater is used, the heating may cause errors in the 
measurement of the sensor temperature, see chapter 1.1.2.3. in 
the Kipp and Zonen CNR1 manual 
(
wo degrees typical, and
zero offsets in the CM3 (10 Watts per square meter typical). 
Under most conditions the accuracy that is gained by heating will 
be larger than the errors that are introduced by heating. 
NOTE
In both the CM3 and CG3, thermal sensors are used. These sensors in principle 
measure a heat flow. For optimal performance, these sensors should be at 
equilibrium with the ambient air. Heating the sensor disturbs this equilibrium. 
This will give rise to zero offsets particularly in the CM3. A typical value 
under conditions with no wind is 10 Watts per square meter. Therefore heating 
should only be done if absolutely necessary. CG3 is less sensitive to this. 
Offset values for CG3 cannot be determined, and therefore are not specified. 
B.1 General Information
There is one major reason for heating: avoiding water deposition on the CG3 
window and on the CM3 dome. In case of dew deposition on the CG3 
window, the dew will ultimately obstruct the Far Infrared radiation completely, 
causing a signal close to zero. In the case of rain, this will probably not lead to 
significant errors, because with an overcast sky the signal is close to zero 
anyway. The case of dew deposition is far more significant. Dew deposition 
will probably take place under conditions with large Far Infrared irradiation 
from the pyrgeometer detector to the clear sky, typically minus (-) 100 Watts 
per meter square. Under conditions where clouds or wind are present, dew is 
less likely to occur. A situation where dew depositing on the window of the 
CG3, causing a signal of -100 Watts per square meter to drop to zero, is 
certainly significant. If heating can avoid this, it certainly should be done 
because all errors that are described above are much smaller than the gain in 
accuracy of 100 Watts per meter square. 
B-1
