ASPH 611 Term Project
A University of Calgary Department of Physics and Astronomy
Graduate Course in Radio Astronomy
Temperature Conversion To Kelvins
Dave Gibson, December 12, 2005
The output data from our system gives us arbitrary units of power on the y-axis. This raises the problem of finding a conversion relation from these arbitrary units to something more useful such as temperature. This would allow us to compare our acquired values with those of other observations that have been made. This requires the observation of a minimum of two objects of known temperature. The difference in temperature should be as wide as possible to minimize the possibility of irregularities in the conversion factor. From there a relation can be made and applied to future data sets before band shape corrections are made. This way any errors in the shape correction do not effect the temperature conversion.
bed was left at ambient temperature (hot load) while the other was doused with liquid nitrogen (cold load). Each absorber was then held in front of the feed horn (special thanks to Fred Babott) to eliminate all other signals. Temperature measurements were taken before and after data acquisition to account for temperature change during the observing period (A 77K temperature was assumed to the absorber doused in liquid nitrogen). This process was repeated 3 times for each load.
Graph 1 - Hot and Cold Load Bands
The acquired data represented the band shape of our system, translated according to the temperature of the load. Looking at the graphs is apparent that, although the relationship between the arbitrary units and temperature is linear, the slope of the relations changes for each of the channels in the band therefore must be calculated for each one. The equation for the correction factor of a single channel is:
where delta T and delta P are the changes in temperature and power respectively giving us units of K/(arb. unit). The correction factor was then calculated for each channel and for each of the 3 sets of loads that were observed. The results were then averaged for each individual channel and implemented in the DATACORR program.
Graph 2 - Correction factor per Channel
Due to a recent discovery as to the averaging technique used in the reduction software, it is possible that this correction factor may be off by some factor. Because these loads were observed at 100ms it may not give the correct temperatures to observations made at anything else.
Back to observations.
Last modified: 10:22 am July 17, 2014