Date: Wed, 09 Feb 2000 12:32:10 -0700 From: "s. a. snedden" Subject: 2.5-meter Obs Log (MJD 51583) TUESDAY, FEBRUARY 8, 2000 APACHE POINT OBSERVATORY SDSS 2.5M OBSERVING LOG (MJD 51583) Observing team: Ellyne Kinney (APO) -- first shift Dan Long (APO) -- second shift Stephanie Snedden (APO) -- third shift Bing Chen (JHU) -- PT Jon Brinkmann (APO) Visitors: Peregrine McGehee (LANL) Robert Lupton (PU) Jon Loveday (UC) Don Holmgren -- by phone (FNAL) Jim Annis -- by phone (FNAL) Software: watcher v2_6_1 switched to v2_6_2 at 02:00 iop v3_8_2 switched to v3_8_3 at 02:00 Plan for tonight: Large pointing model. Counterweight 2 has been fixed by French. Russell explained that the difference between commanded and actual M2 positions that was seen last night is to be expected and will decrease as the mirror moves in the collimation process. Ellyne noted that since the installation of the new spectrograph fill system, the watcher will issue warnings about spectrograph dewars being empty just after a fill. This apparently is caused by low pressure in the newly filled dewars, and clears soon. Steph noted that the sp2 dewar takes longer to fill (is emptier) and usually runs out of LN2 first. The 10-l dewar may have lost some vacuum. Disk space on /data/photo is low. The 07feb200 focus sweep data will be deleted (a tape of the data was sent to jeg). Hartmann data will also be deleted. Ellyne and Jon B. took spectro darks per Alan's instructions during the afternoon. Their notes were sent out in detail in a separate message to sdss-spcommish@astro.Princeton.EDU Ellyne tried Ron's feature that should gracefully handle data loss during bias drift scan 1128. She reported that the feature did not work. The early evening was cloudy, so Dan did some tests: He did some refining of force distance tables for the cartridge with corrector and the engineering camera. He also did some testing of fiducial repeatability. This yielded some interesting results. This all started when he was setting the fiducials in preparation for the night's run. The altitude fiducial was off by about 10 arcseconds. he had noticed it being off by between 5 and 15 arcseconds on a couple occasions when he was on early in the evening and setting them for the first time that day. Tonight as in previous times, he would set the fiducial again when he found it change by more than a couple arcseconds upon setting it. Consistently, subsequent settings of the fiducial would show only slight changes, on the order of a couple tenths of an arcsecond. This got him wondering why, if it is that repeatable, it ever gets 10 arcseconds off. What he found was that if there has been a lot of altitude motion between times of setting the fiducials the errors are larger, as if the encoder is loosing/gaining counts. Below is a list of moves and fiducial changes. Alt motion error from setting fiducial < 1 deg < .2" 2x 20 to 90 to 20 2.1" 10x 20 to 90 to 20 9.0" Running similar tests on the azimuth and rotator did not show such large errors. Az motion error from setting fiducial < 1 deg < .2" 5x 120 to 300 to 120 3.1" Rot motion error from setting fiducial ~5 deg < .1" 4x 0 to 150 to 0 .3" Some other things he observing while doing these tests. 1.) The altitude is in contact with the windscreen for a time when moving up about 5 degrees after the start of a long TCC TRACK command ( more than 40 deg ). The motor currents show something happening, rising to about 6000 for a few degrees, when otherwise they stay within 2000 of zero. 2.) The azimuth went into oscillation on about 2/3 of the moves ( note that alt was at 87 during this ). 3.) Dan had trouble finding rotator fiducials when moving at 250000 counts per second ( missed 3 out of 4 times ), but found it every time at 50000 counts per second. We opened at about 22:30, filled the spectrographs, homed the optics, and then proceeded to set up for taking instrument block data. The block wasn't far off. It took a while fighting clouds to get all the data, rot_inst_ang changed by -.34 deg to 89.61 rot_inst_xy changed by .003 and -.003 to .001100 , -.000150 image scale changed by 1.3% from 23722 to 24040 IMAGING: Bias scans were run for analysis at FNAL. They are runs 1124-1129. Steph had trouble archiving them to tape. The crates had to be rebooted, and it took some time to figure out how to get the tapes happily mounted & enabled. Talked with Don Holmgren. The tapes would not load properly...ArchLoad did not get through & archRewind did not get through. ArchInit was also unresponsive. Turns out the tapes were interrupted during a write ... SCSI bus re-set. Ultimately, the tapes in crate 1 had to be replaced. As a result of the troubleshooting, endNight was started late, and did not finish until after the daily FedEx arrival. The tapes will have to be shipped tomorrow. POINTING MODEL: The clouds cleared out enough for us to take a 70 point pointing model starting at 02:30. Because of the questions we have about cumulatively losing altitude encoder counts, we paused the model every 15 points and reset the altitude fiducial (and azimuth, just in case), then resumed data taking. The worst change was 2.5" in altitude on one fiducial reset. The others were 2" or less. Steph spent too much time on the archiver problem to permit running TPOINT before filing this log. It will have to be run later today. MIRROR NUMBERS Primary: setmir: 0 -3.9 3.9 213.5 213.5 galil: 4850 -2350 100 -9500 50 -100 linear gauges* 1 = 1.9 2 = ? 3 = 0.5 4 = 0.4 5 = ? 6 = ? Secondary: ( inst=engcam focus=-3100 alt=57.6 deg T=5.0 C) setmir: -3353 5 -5 50 -58.8 galil: -4221259 -4225407 -4235650 -91 -5237 linear gauges* 1 = 3.302 2 = 3.328 3 = 3.324 4 = 0.010 * Realtime display had all zeros for the migs. The plot showed values for the some of the primary but couldn't get accurate numbers. Secondary migs read through terminal server. No interlocks were bypassed. Tapes status: Will be shipped tomorrow. See section on imaging. LN2: The spectrograph dewars were filled at 22:30 and again at 07:30. There were no full dewars for the imager. When Norm came in in the morning, he proceeded to fill a half-full one from another half-full one. French and crew will install the dewar this afternoon after it pressurizes. The imager should be okay until then...it had about half of it's 180-l dewar left at 08:00. Weather: Skies were variably cloudy. Seeing was about 2arcsec, at times going to about 2.3 arcsec, throughout the pointing model. Winds were out of the SSW through the night. Wind speeds varied from 10-15 mph. Dust counts varied from 20-60 particles/ft^3. Relative humidity went from 5%-20%.