5.5 Detector Response to the Internal Calibration Source

The measured response to an internal stimulator flash is used to calculate a gain factor for each of the detector blocks 2, 3, 4, 5 and 6. Up to revolution 64 (still within the Performance Verification phase) band 1 also received an internal illuminator bright enough for these detectors (InSb). However, the stimulator setting was too high for bands 2, 4, 5 and 6 and introduced transient effects in these bands. Band 1 detectors were also quite stable, within 2% (Heras et al. 2000, [15]), and the decision was made not to use the high stimulator setting after revolution 64.

Figure 5.28: Detector gain plotted against the phase of the orbit. The general trend is for a decay of the response with time in the revolution, with bands 2 and 4 showing the increase in the spread of points near the end of each revolution (approaching perigee), consistent with the increasing number of radiation hits.

Figure 5.28 shows a sample (roughly 1000 observations) of the photometric gains based on the internal illuminator scans, plotted as a function of phase (time after perige in a revolution). For each of the bands 2 (Si:Ga), 3 (Si:As), and 4 (Ge:Be) the response of the detectors changes systematically throughout the revolution. This is the same trend as seen by Heras et al. 2000, [15], but the gains are now calculated from all 12 detectors. Note that bands 3 and 4 see very similar wavelengths of the illuminator and yet record extremely different variations in their responses. This fact is taken as an indication that the responsivity of the different detectors is changing during a revolution and not the illuminator itself.

The general trend is for a decay of the response with time in the revolution and is related to the long term transient drifts caused by the extreme radiation hits received during perigee. Both bands 2 and 4 also show increase in the spread of points near the end of each revolution (approaching perigee), which is consistent with the increasing number of radiation hits.

Nominally, each observation contains at least two scans of the internal calibrator. Studies show that the 2nd internal calibration scan agrees with the first within the uncertainty of the measurement.

The average response to a stimulator flash is contained in the Cal-G 41 file. Stimulator flash data obtained during a science observation are compared to the average data in this file to determine the `photometric' gain correction for the science observation.