From harwood@hgea.org Thu Oct 2 08:49:59 2003 Date: Wed, 01 Oct 2003 16:48:13 -1000 From: Jim or Pat Harwood To: denault@jeans.ifa.hawaii.edu Cc: tokunaga@ifa.hawaii.edu, tolles@ifa.hawaii.edu Subject: Correcting telescope position for index and collimation errors Tony, We talked at the Requirements meeting about the index and collimation error correction. Later, I was able to refresh my memory on the index and collimation correction facility that is currently provided to the TO's, so here's a rundown: Index Correction A pointing run is done with reference to a particular crosshair setting on a camera that is looking at the focal plane of the telescope. Then, when an instrument is used with its own crosshair or equivalent, the position of the instrument crosshair in the focal plane is usually different from the crosshair position of the camera when the pointing run was done. The result of this misalignment is that stars are all acquired off center in the instrument. They have to be guided back to the instrument center of view after every slew. The amount of misalignment is the same all over the sky. This misalignment is a coordinate index offset. Doing the Index-Collimation procedure (more on this below) with the instrument mounted will produce HA-Dec index correction offsets that are combined with the pointing correction so that slews will (or should) bring all objects to the instrument center, not the camera center. Collimation Correction Collimation is the alignment of the optic axes of the various components of an optical device. The optic axis of the camera that is used during a pointing run may not be exactly parallel with the telescope's optic axis. This collimation error of the camera is figured into the pointing error surface, so that slews will be corrected for the pointing camera collimation and the telescope's collimation state at the time of the pointing run. However, when an instrument is mounted, or when the telescope's internal collimation is changed, the collimation difference with respect to the camera during the pointing run will produce a collimation pointing error. The result of collimation pointing error is different for the two axes. The declination collimation error is just like index error, a constant coordinate offset. However, the HA collimation error changes by a factor of the secant of declination. So, as you go to higher and higher declinations, the HA pointing error due to collimation will get worse. Correcting index and collimation errors It would be very inconvenient to have to do a pointing run every time a new instrument is mounted or the telescope's collimation adjusted. At the same time, not correcting for index and collimation errors produces inconvenient pointing errors on all slews. We have developed a procedure to quickly measure and compute the current index and collimation errors and automatically apply them during slew. The telescope operators have a procedure associated with MCC-1 Pushbutton 6 that does this. First, they slew to a star near the celestial equator and, after positioning the target star at the center of the field of view, they press Pushbutton 6. Then, they choose a star at high declination and repeat the process. After pushing #6 for the second star, a calculation is done which determines index and collimation coefficients for HA and dec. These coefficients are then used for every slew to trim up the final slew position. Using stars widely separated in declination for the calculation of the correction coefficients gives an accurate calculation for the coefficient sensitive to declination, the HA collimation coefficient. With the HA and declination guiding corrections for the two stars, a set of simultaneous equations is solved for the four coefficients. Once this procedure is done, the index and collimation coefficients are used for all subsequent slews. The numbers are saved on the TCS floppy disk for future reboots. So, is this procedure in routine use? You will have to ask the T.O's, but probably not. As I intimated in the Requirements meeting, the current pointing performance of the telescope is poor enough that index and collimation errors are probably down in the noise, and probably, nobody bothers with the correction procedure. Note also that the index-collimation coefficient determination procedure depends on guiding the star to crosshairs after the two slews. If the majority of the pointing error for those two slews is due to general pointing inaccuracy instead of index and collimation error, the numbers computed for these coefficients will be very soft. I would expect TCS3 to produce much more accurate slews, so that the presence of index and collimation error will be very noticeable. I recommend that correcting for these errors be built into TCS3. -Jim ==================================== Jim and Pat Harwood, at home 1928 McKinley St. Honolulu, HI 96822 harwood@hgea.org ====================================