before the flare, the thin shutter is already in. at ~20:40 UT the thick shutter comes in. the jump at ~20:46:30 is caused by a decimation change (D1 to D4).
pointing better than for the X17 flare, but after the thick shutter moves in, there are still times when SAS does not see the total solar disk.
image shows solar limb in yellow; white line gives location of rotation axis as observed by SAS. the red crosses are averaged positions. time interval=['2003/10/29 20:44:55','2003/10/29 20:45:27'] after thick shutter moves in.
after the thick shutter moves (~20:40), the full set of 'limbs' is only available for some time intervals:
The flare is divided in 20 time intervals with good SAS solution (duration of intervals are ~40 seconds). for each interval CLEANED images were made with grid 3-9 (10" FWHM clean beam). this images are shown in contours on TRACE images.
TRACE observed with different filters (1600A, 195A, and a few with 284A). I picked TRACE images that are close to the 20 intervals. so some images show 195 A (green color table) others 1600A (black and white color table).
imaging at 15-20 keV does not work between 20:44 and 20:50 UT. I do not understand this. However, during this time interval livetime is below 60%.
separation of ribbons can be clearly seen.
