Effect on SCI_PHOT data from a loss of synchronization

For some time during June 2011, there have been reports of scans flagged in MIA for SCI_PHOT data for the last third of a fringe track or so. This is (for a test observation) the appearence of the MIA VIS_GUI for such a case.


This problem can be corrected in the data reduction using MyMidiGui 1.7.5, specifically replacing xmidispvisi.pro in the MIA software by the version distributed with MyMidiGui. The reduction of the above test obseervation would then look like this.

The problem was traced by Walter Jaffe to a slight change in the time it takes to read out a window when the readout windows were adjusted for the SCI_PHOT mode last year. It turned out that the software seems to use the experimentally determined readout time to set the chopping frequency and flag the individual frames.

Citing Walter and Thomas Rivinius (MIDI IS), in the older files the chopping rate (CHOP FREQ) was 0.544767. The shutter open exposure time per frame was 0.18 s. The readout time per frame was 0.00494464s, and thus the total frame time dt=(exposure+readout) was 0.022945117s. The PVRATIO was 1.0 and the number of frames/scan was 40, so a full chop cycle of target+sky is 80 frames. A full observation is 8000 frames, or 100 chopping cycles. Doing some math on these numbers, the end of 8000 frames the flagging of frames is out of sync by 0.16 frames. 

8000*dt     100*tchop      8000*dt-100*tchop    (8000*dt-100*tchop)/dt
183.56103s   185.56472s     -.00368s               -0.16 frames
For the data since the change of the window, the chopping rate is unchanged, and the live exposure time is still 0.018s, but the readout time is 0.004934535s, thus dt=.00229344s, slightly faster. Thus at the end of 8000 frames the flagging of frames is now out of sync by 3.9 frames. 
8000*dt     100*tchop      8000*dt-100*tchop    (8000*dt-100*tchop)/dt
183.47529   185.56472s     -.089428s               -3.90 frames
This fix has been applied as of September 2011.

Christian A. Hummel

European Southern Observatory
Karl-Schwarzschild-Str. 2
85748 Garching
Germany