Date: Fri, 25 Jun 1999 03:13:44 -0600 From: long@apo.nmsu.edu (Dan Long) Subject: 2.5m obslog 990624 ( MJD 51355 ) THURSDAY JUNE 24, 1999 APACHE POINT OBSERVATORY SDSS 2.5m OBSERVING LOG Observering team: Jeffrey Hayes (APO) Stephanie Snedden (APO) Dan Long (APO) Tonight we tested the azimuth after some work during the day, and ran the EC in centroiding /pointing tests. -We found the same behavior as before regarding azimuth amp overcurrenting at velocities in the negative direction which would be fine in the positive direction. -The azimuth only "hummed" once briefly tonight ( el was 78 deg ). -There was a curious lurching behavior in azimuth for the first dozen slews which decreased in frequency as time went on. -There was one rotator runaway on the first slew of the night. -The centroiding algorithm worked fine for FK5 stars. I noticed nothing different in its behavior. -A couple elevation sweeps worth of pointing data yielded 2.0 arcsec rms with no extra fiddling. -A full model was less successful do to a problem cropping up which would not allow calibration offsets to complete. -The 15 deg stop still does not get out of the way when the altitude is moving briskly. Starting about an hour before sunset, we took the enclosure off, and got set up to make some azimuth tests. This included considerable moth gut removal from the azimuth drive disk. We observed the same behavior reported in the log a couple nights ago, that in the negative direction the azimuth would overcurrent at velocities of 1.45 deg/sec or more, but 1.4 deg/sec was fine. Also as before, in the positive direction 2.0 deg sec worked as well as lower speeds in terms of motor current. Note that TPM logging was active all night tonight. The data file is called data_990624 and is in the usual directory, and hopefully is filled with useful data. A curious thing was noted about azimuth motion. Occasionally during the first several slews the telescope would abruptly speed up then slow down. This usually happened during the steady state max velocity of the slew. These episodes occured quickly enough that I never really could see the current change much on the MCP readout. I'm pretty sure it was the telescope rather than the windscreen. Watching closely I would see the telescope move with respect to the windscreen _then_ I would hear the pitch of the windscreen motors change as it tried to recenter, rather than the reverse. This behavior became less frequent and ceased altogether as the evening progressed. There was only one incidence of azimuth oscillation. It "hummed" once while taking pointing data when the altitude was at about 78 degrees. Sometimes while crossing fiducials we would see errors like the following AXIS 0:ERR=8OFF latched pos0=31024674. pos1=19.0 On the first slew of the night the rotator ran away, hitting the negative hard limit before we realized it wasn't doing what was asked of it ( one of the drawbacks to higher velocities ). There was no damage, the limit behaved as it should. We just backed off the rotator a few degrees by hand, and reset the amplifier. It did not happen again. The following limits are now in the axelim block. The only change is the azimuth velocity limit was raised to 1.4 deg/sec from 1.25 deg/sec. Az El Rot ! Maximum |velocity| (deg*/sec) Vel 1.4 1.25 2.0 ! Mmaximum |acceleration| (deg*/sec^2) Accel 0.1 0.1 0.1 ! Maximum |jerk| (deg*/sec^3) (ignored for uncomputed offsets). Jerk 0.4 0.4 0.2 We then proceeded to setup for taking pointing data. There was some trouble talking to the EC which Craig traced to some recent rewiring work in the computer room ( the perils of not observing for over a week... things change :) After setting up on an FK5 star, we focused, and noticed those odd triangular star shapes near focus ( see /data/spectro/guider/gimg0009.fits ). Homing the secondary did not change the shape. However, relaxing the primary got rid of it ( see number 16 I think ). The numbers from the migs Mirror Numbers: Primary: 1 = 1.932000 2 = 0.556500 3 = 0.555500 5 = 1.903000 6 = 0.403000 setmir: 0.0, 0.0, 6.3, 341.6, 0.0 galil: 4600, -2350, 350, -12900, 4400, 4150 Secondary: 4 = 2.799000 7 = 2.816000 8 = 2.814000 setmir: -1420.6, -227.0, -90.0, 0.0, 0.0 galil: 1617750, -1842900, -1503800, -12000, -135337 We then proceeded to take a couple elevation sweeps of pointing data. One at ~120 deg az, and the other at ~0 deg. When this data was run through tpoint we got a sky rms of 2.0 arcseconds without having to remove any data points! Only the main terms were used ( ie ia pzz1 an aw ca ). Sadly, when I tried to take a full model, the telescope started making an odd complaint I've never seen before. The result is that most of the time calibration offsets would not be implemented. It wouldn't even save the offset for the next slew, it would simply report the following in TELRUN and leave the calibration offset unchanged. offset calib/pabs 0 0 F Modu="prt_ReadReplySet"; Text="too many replies on port LTA19:" 0 0 I Modu="prt_ReadReplySet"; Text="the following replies were read:" 0 0 I Modu="prt_ReadReplySet"; Text="ERR: bump active" 0 0 F Modu="prt_ReadReplySet"; Text="too many replies on port LTA19:" 0 0 I Modu="prt_ReadReplySet"; Text="the following replies were read:" 0 0 I Modu="prt_ReadReplySet"; Text="ERR: bump active" 0 0 F Modu="axe_o_Move"; Text="command failed for TEL1, TEL2" 0 0 I TCCStatus="HHH","NNN"; TCCPos=NaN,NaN,NaN; AxePos=-3.86,35.45,-3.42 0 0 I SpiderInstAng = -3.499064, -0.004300, 4437010134.46000 0 0 I AxisStop="TTF" 0 3 F Text="move failed; axes halted" Without being able to consistently issue corrections, stars would stop showing up in the field. That problem combined with patchy clouds caused me to stop the model only a dozen points in. I have not implemented the model generated by the two elevation sweeps. The EC is still connected to AC power via an extension cord, so be careful making moves with the telescope that the cord doesn't catch on anything. Dan