The temperature of the NICMOS detectors is an important calibration parameter. The detectors show a number of effects which are dependent on the thermal environment. Most of these effects can, however, be removed through standard calibration, making it important to maintain the temperature as stable as possible over long time scales.
The temperature of the NICMOS dewar and its detectors is presently monitored using the NIC1 mounting cup sensor. This sensor is located behind the NIC1 camera array and is essentially a diode, with an operating temperature limit of 78K. The temperature is read-out every 30 seconds and is part of the normal HST engineering telemetry.
NICMOS thermal history can be divided into two distinct epochs. It was initially cooled to ~61K by a cryogenic dewar containing solid nitrogen. The cryogenically cooled dewar suffered a shorter lifetime than expected due to a thermal short, leading to a faster sublimation of the nitrogen. Following nitrogen exhaustion in January 1999, the NICMOS instrument warmed up to a temperature of around 260K. As this was much too warm for any useful science observations, NICMOS stayed dormant until the third Servicing Mission in March 2002 when a mechanical cooling machine, using a closed-loop reverse-Brayton cycle, was installed. The dewar temperature is regulated by the new NICMOS Cooling System (NCS) and is used to maintain the detectors at 77.15K, about 15K warmer than with the solid nitrogen cooling.
Operations during the first epoch, with solid nitrogen as the coolant, the detector temperature slowly increased from 61K to 62.5K from the start until right before nitrogen exhaustion. This variation in operating temperature made it necessary to obtain calibration data, i.e. darks, flats and photometric calibration, quite frequently to keep up with the changing thermal environment. During the NCS era, the dewar temperature is higher but also significantly more stable. Temperature variations over the time period March 2002 until the present is less than ~0.1K. Occasionally, there are temperature fluctuations with larger amplitudes, mostly associated with unusually long orbit day time, i.e. the time of the orbit that HST spends in sunlight, combined with the overall solar insolation (maximum in January and minimum in July). More information can be found in the NICMOS ISR 2006-004.
Below we give the temperature, as measured with the NIC1 mounting cup sensor, in monthly and yearly periods for both epochs, starting in March 1997.