Return to Previous Section

B. Reduction of Dark Frames


In order to study the behavior of the dark frames as a function of temperature and exposure time, it is first necessary to bias-correct them. The first step in the process is to see if the bias frames fall on the Bias-Temperature relation discussed in Section 2. In Figure 8 I show a Bias-Temperature plot similar to Fig. 1, but including the bias frames from one of the dark-test nights. Similar plots for the other dark-test nights showed the bias levels to be consistent with the Bias-Temperature relation determined in Section 2.

  • Figure 8: The Bias-Temperature relation from Fig. 1, including results from the bias frames obtained during the 08 March 2004 dark tests.

    The next step is to look for temporal variations in the bias level at a given detector temperature. A typical dark-test temperature sequence takes 1.5 to 2 hours. Fig. 2 shows an example of the bias level for one temperature sequence from 18 January 2004. As discussed in Section 2, although there is clearly bias-jitter at the +/-1 ADU level, there is no evidence for coherent temporal drift in the bias on 1-2 hour timescales. Given this, I median combined all 13 bias frames for each dark-test temperature sequence, and then subtracted the medianed bias frame from each of the raw dark frames. Figure 9 shows an example dark frame both before and after bias subtraction. Note that the top-left to bottom-right gradient is largely removed, as are several of the bad columns.

  • Figure 9: A 10-minute dark frame with T ~ -20 degrees C, obtained on 08 March 2004. Top: Before Bias-correction. Bottom: After Bias-correction.


    Procede to Next Section

    Return to Main Dark Page

    Return to Outline


    Updated: 2009 August 20 [pbe]