Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.naic.edu/~phil/pnt/tiltandtdmotionJul02.html
Дата изменения: Fri Oct 19 19:36:35 2007
Дата индексирования: Tue Oct 2 13:30:12 2012
Кодировка:
Поисковые слова: m 65
platform_tilt_when_td_move_together

Platform tilt when tiedown tracking on

july,2002

     The average height of the platform is kept constant using the distomats (laser rangers) and the tiedowns. Every two minutes the distomats remeasure the average height of the platform. The tiedowns will then pull or release evenly on all 3 corners of the platform to keep the height close to optimal. The tiedown motion is a gradual one that takes 45 seconds to accelerate up, coast, and then decelerate. The tiedowns are moving in program track mode using a 24 bit optical encoder that is accurate to fractions of a mil.
    During some of the pointing runs there were source tracks that had jumps and then recoveries  in the pointing errors. One possibility was that the tiedowns could have been tilting the platform as they moved from one position to the next. Maybe the cables were stretching differently. Data was taken on 29jul02 with the azimuth at 242.87 degrees and the dome at 10 degrees za. The tiedowns made two motions of .2 inches and .16 inches while the tilt sensors where running.
The plot shows the pitch and roll of the tilt sensors during the tiedown motion (.ps) (.pdf).
  •     The top figure is the relative pitch of the dome (i did not put it on the absolute scale determined by the theodolite).
  •     The bottom figure is the relative roll of the dome. Positive roll is clockwise looking uphill.

    • The red line is the motion of the tiedowns during this period. The tilt is not affected by the tiedown motion so they must be moving at the same rate. There is a small downward drift in the roll over the entire period of time. The oscillations at the end of the measurement is me walking onto the turret floor to disable the tilt sensors (maybe i need to go on a diet!).
    From previous computations, a pitch error of .01 degrees at za=10 is a pointing error of 4.3 arc seconds (assuming you have pulled the telescope back down into focus).  The motions we see in the tilt sensors would map into small pointing errors.
        The jumps that were seen in the source tracks were probably caused by the gregorian wheels sliding across the rolling surface (since the side rollers are not tight).
        processing: x101/020729/doittilt.pro
     home_~phil