Detector Electronics Temperature

Prior to delivery to XRCF, the ACIS integrated flight instrument underwent testing at MIT Lincoln Laboratories. Some of these measurements involved operating ACIS with the Detector Electronics Assembly (DEA) at temperatures of -17 C$\hbox{$^\circ$ }$ and 20 C$\hbox{$^\circ$ }$. These tests were used to map the gain dependence on the electronics temperature. A detailed description and analysis of the gain calibrations are reported in ACIS Memo #143 [Pivovaroff1997c]. The main points are summarized below:

                 -The DEA was stabilized at temperature Ti, (Ti = -17 C$\hbox{$^\circ$ }$,20  C$\hbox{$^\circ$ }$)

                 -Each chip was independently illuminated with an Al K$\alpha$ source  and an 55Fe source

                 -The focal plane (FP) temperature during these measurements ranged between -116 and -119 C$\hbox{$^\circ$ }$

                 -Gains were calculated by fitting a linear relationship to a gaussian centroid for the Al K$\alpha$ and the Mn K$\alpha$ line

Table 4.65 lists the mean gain for each chip for DEA temperatures of 20 C$\hbox{$^\circ$ }$ (hereafter LLTV_hot) and -17 C$\hbox{$^\circ$ }$(hereafter LLTV_cold). Table 4.65 also lists the average chip gains calculated from the XRCF Flat Field (XRCF_FF), Phase I data. See Section 4.3.1 for the details of these measurements.
 

Table 4.65: Mean gains for ACIS at different operating temperatures

FP Location	Mean Gain (eV/ADU)
	LLTVcold	XRCFFF	LLTVhot
	DEA=-17 C$\hbox{$^\circ$ }$ FP=<-117.5> C$\hbox{$^\circ$ }$	DEA=13 C$\hbox{$^\circ$ }$ FP=-110 C$\hbox{$^\circ$ }$	DEA=20 C$\hbox{$^\circ$ }$ FP=<-117.5>
I0	3.812	3.827	3.840
I1	3.809	3.849	3.857
I2	3.563	3.592	3.610
I3	3.839	3.882	3.891
S0	3.689	3.797	3.811
S1(BI)	4.405	4.413	4.455
S2	3.724	3.758	3.771
S3(BI)	4.492	4.507	4.574
S4	4.256	4.306	4.319
S5	4.075	4.101	4.119

It is clear from the data that the gains increase with increasing DEA temperature. In order to quantify this relationship, we have made a linear fit of the gains as a function of temperature. Figure 4.97 shows the linear best-fit plotted against the measured gain values. The hash marks simply indicate the data point and are not representative of errors. The LLTV_cold data for I0 and S0 are known to have a problem with drifting biases, so these two data points are suspect^4.7. With only three data points to fit, the incorrect LLTV_cold gains greatly influence the fits and skew the slope of the expected lines. Excluding I0 and S0, the data seem to follow the linear relationship quite reasonably. The small deviations from linearity are expected, as only two data were used to define the LLTV gains.
 

Figure 4.97: ACIS gain variation as a function of DEA temperature.