Date: Sat, 15 May 1999 06:41:46 -0600
From: long@apo.nmsu.edu (Dan Long)
Subject: 2.5m obslog 990514 ( MJD 51313 )
                          FRIDAY MAY 14, 1999
                     APACHE POINT OBSERVATORY SDSS 2.5m
                              OBSERVING  LOG
   Tonight we saw that the recollimation of the secondary did wonders for
the image quality.  Images looked round rather than triangular.  We had
less success when it came to tracking.  There were rotator oscillations at
a large number of angles.  We will need the rotary encoder back for
Saturday night's operation.
   The imager was on all night tonight.  There were plans to put the
cartridge on during second half, but when rotator oscillations were
discovered, tracking was no longer an option, so we parked on the equator
and continued imaging.
   There was trouble with alt fiducials not working at the start of the
night.  It turns out someone had unplugged the reader for some reason, and
forgotten to put it back.
   First we set up on the equator.  Images look much nicer now.  There is
only a small amount of coma.  All other image artifacts are gone.  The rate
was checked on the equator.
   We then set up on on a great circle ahead of M13. Slew command below.
track 247.925,35.544 fk5 /rotty=obj /rota=-.22 /scanvel=(.002759,.003138)
   Robert wrote a proc to take the rotation info and apply it via a series
of small calibration offsets in two second intervals.  The TCC was
decidedly unhappy with this.  Below is the TCC output after a series of
calibration offsets in rotation.  About 7 of them in .03 deg increments.
0 2 I CalibOff = 0.000000, 0.000000, 4433471672.38000, 0.000000, 0.000000,
4433471672.38000, -0.0300
00, 0.000000, 4433471672.38000
0 2 :
0 0 ! Facil="SYSTEM"; VMSMsg="HPARITH"; EText="high performance arithmetic
trap, Imask=00000000, Fma
sk=00002000, summary=04, PC=000D17A0, PS=0000001B"
0 0 F DeadProc="T_TRACK"
0 0 W Modu="axe_i_Stop"; Text="no axes specified; nothing done"
0 0 F Modu="axe_i_SchMove"; Text="tracking failed; axes halted"
0 0 I AxisStop="TTT"
0 3 F Text="move failed; axes halted"
   At around 1:00 while Connie was filling the spectrograph dewars, she
noticed a low frequency hum and vibration from the rotator.  The rotator
oscillation ( ~20 Hz ) was noted at mount angles ranging between 120 and
205 degrees at elevations between 57 and 86 degrees.  The oscillation was
large enough to make image quality poor.  It was worrying enough with
spectro and imager optics mounted that we decided not to do any more
tracking.  We again set up on the meridian and equator and parked.
   I attempted to run the tpm tonight, but got the following when I tried.
        TPM>start data
        TPM>[TPM,tShell,3,E] Looking for v3 format, found v1. Need to recompile.
        stop
        ERROR: Data is not currently being logged.
        TPM>
   Out of focus images showed some coma, so we decided to adjust the
primary.  We first homed the secondary to assure a proper starting point.
Excerpts of the home are below, and seem to show rather large errors.
:XQ#HOME
: 0000.0,  0000.0,  0000.0,  0000.0,  0000.0 sec to set up motion
 0400.2,  0400.2,  0400.2,  0006.5,  0006.5 max sec to find reverse limit switch
 0030.1,  0030.1,  0030.1,  0030.1,  0030.1 max sec to find home switch
 0010.2,  0010.2,  0010.2,  0000.5,  0000.5 sec to move away from home switch
 0002.3,  0002.3,  0002.3,  0000.1,  0000.1 max sec to find index
 036,  002.1 microsteps, sec to find full step
-000040000, -000030000,  000000000,  000001050,  000000350 position error
 1,  1,  1,  1,  1 axis homed
   There was also a rogue COLLIMATE process which did not die when STOP
COLLIMATE was issued.  A SHOW SYSTEM revealed it and its PID, and it was
stopped.
   The telescope was taken to 87 deg once the collimate process was dead
and the primary was set using setmir to all zeros ( the nominal mechanical
collimation position ).  It was then relaxed and taken back down to the
equator.
   The secondary was placed in the proper position for 57.22 deg altitude
             1577.3, -180.0, -160.0, 0.0,110.9
   We then applied tilts to eliminate coma.  The numbers which did this were
                0, -2, 30,  0,  0
Next we used equations to back out the proper tilt and translation values
from the tilt only values.  The equations were from Pat Waddell in a
message at the end of February excerpted here
        Assuming that M1 is "mechanically" collimated
        Using SETMIR, find tx0 and ty0 (arcseconds) to remove coma
        This implies that the M1 center is displaced (in microns) from the
         rotator
        axis and may be found at
                cx =  ty0 * 42.7
                cy = -tx0 * 42.7
        Thus M1 must be moved to position dx = -cx, dy = -cy, or
                dx = -ty0 * 42.7
                dy =  tx0 * 42.7
        And to keep proper axis alignment M1 must be tilted (from the
        original attitude)
                tx = -dy * 0.0183
                ty =  dx * 0.0183
        final SETMIR command:  0 (piston), tx, ty, dx, dy
This gave the final setmir command
                0, 1.6, -23.4, -1281, -85
  These were implemented, and the primary relaxed at 87 deg.  When we
slewed back to the equator, it was obvious that these numbers were not
right.  The images were quite comatic.  I hope someone can figure out what
was wrong and take the tilt-only values we found which eliminated coma and
generate the correct tilt and translation values.
   The telescope was put away at 05:20.
                                Dan