From: Subject: 3.5m Night Log 20090411 Date: Sun, 12 Apr 2009 12:26:05 +0000 (UTC) Apache Point Observatory 3.5m Telescope Night Log Saturday, April 11, 2009 ACTUAL ASTRONOMER OBSERVER(S) INSTRUMENT START FINISH ------------------------------------------------------------------- S.Hawley R.McMillan Echelle,Tspec 19:40 20:40 N.Zakamska R.Reyes,X.Liu Tspec 23:45 00:10 T.Murphy R.McMillan APOLLO --:-- --:-- J.Greene J.Greene DIS --:-- --:-- T.Murphy R.McMillan APOLLO --:-- --:-- INST SCHED(TWILIGHT) WEATH EQUIP OBS DAY NOT USED I/ROW/DN START FINISH D B D B D B USE D B ---------------------------------------------------------------------- UW12/W 19:39 20:39 0 0 0 0 0.2 0 0.8 0 0 PU08/R 20:39 01:05 1.3 2.7 0 0.2 0 0.2 0 0 0 DD01/O 01:05 02:35 0 1.5 0 0 0 0 0 0 0 PU04/R 02:35 05:05 0 2.5 0 0 0 0 0 0 0 DD01/O 05:05 06:29 0 0.6 0 0 0 0 0 0 0 note: scheduled times listed include instrument change time ------------- ACTIVITY LOG -------------- Observing Specialist: R.McMillan This was a challenging night in many ways. The weather was extremely variable, with occasional patches of clear sky (and poor seeing) alternating with clouds, high humidity, high winds, and snow. Some of the precipitation was not apparent on radar, but it was hard to call the clear patches "sucker holes" when we had some programs that could perform meaningful science in short openings. Snow and graupel when I drove in gave way to clear skies at sunset, so I opened and got some data through clouds with both Echelle and Triplespec. When the clouds thickened and radar showed something approaching, I closed. I started up a closed-dome pointing model while we waited on the clouds. This was to serve two purposes: I was exercising the rotator, which has recently had a small change to its controller software. And I was also looking at the telemetry for the azimuth, trying to get a handle on some of the extra noise we've been seeing. The rotator failed the first init attempt with "servo error too large," but after a second init it seemed to behave normally. For the azimuth, I made a modest change to one of the PID loop parameters which reduced the noise noticeably (PE.RATE.SCALE changed from 0.1 to 0.06). Past experience suggests that this change should not have a strong impact on tracking, but I was hoping to confirm that on sky. Clouds and radar echoes improved towards midnight, so I stopped the in-dome pointing model in order to open up. When the observers slewed to their first star, it appeared that offsets weren't quite behaving correctly. Finally I noticed that the rotator was not at its commanded position, and was saying "servo error too large." Fritz and I later reconstructed that the rotator had probably been in an unbalanced position (with Triplespec mounted) when I stopped the motors in order to open the dome. The unbalanced rotator drifted and produced the error. But that wasn't quite the end of the story. When I tried to re-initialize the rotator at 05:51 UT, I first got a series of failures messages including "read timed out on LTA5". The second init seemed to work, but the rotator still wouldn't move -- it kept claiming "motor control buffer empty." I eventually got to the point where the observers were able to center their target and observe, just by telling the TCC to put the rotator at the position where it appeared to be stuck. In the mean time, I called Fritz for help. Fritz established that an incorrect flag was being set, causing the false "motor control buffer empty" flag. Once that was corrected and the controller was reset, the rotator appeared to behave normally. The reasons for the init failure at 05:51 are still unclear, and the unbalanced drifting problem seems to suggest that Triplespec should always be parked at 0 mount even if the motors are only stopped briefly. While this was going on, one of the inbound APOLLO plane spotters called to report that it was snowing along the Sunspot highway. Clouds were closing in again and humidity was rising, so I closed the dome. A few minutes later, spots of snowflakes started to appear on the cloudcam. Second half the humidity remained high. I did some more in-dome pointing to exercise the rotator. DIS was mounted even though I neglected to change the set inst, so it still said TSPEC. The rotator behaved well, and the azimuth dataq traces also looked better... until I stopped the pointing model and moved the telescope to park position, when I saw two strong spikes. So further checks on the useful range of parameters may be needed. ------- FAILURE LOG ------- PROG INST FAILURE MODE TIME (SEDFNVOG) TI/SHU START FINISH DESCRIPTION ---------------------------------------------------------------------- PU08 T T/H 23:52 00:04 rot misbehavior ---------------- FOCUS LOG --------------- Time Instrument Focus Az El Temp Strc Prim Sec Air filt FWHM ------------------------------------------------------------------------- 19:52 ECHELLE 150( -890) 239 52 -1.6 1.0 1.6 -0.3 -1.1 ND3 2.15 20:27 TSPEC 190( 90) 239 46 -2.1 -0.5 0.1 -1.4 -1.8 - 2.00 ------------- WEATHER LOG ---------------- Wind Precip.or Time Temp F Dewp F MPH Direction Dust Sky Conditions ------------------------------------------------------------------- 16:00 34 20 20 236 (SW) 2934 cloudy lt snow 17:00 34 20 19 259 (W) 710 mstly cldy none 18:00 34 19 27 268 (W) 2528 mstly cldy none 19:00 33 16 23 271 (W) 361 ptly cldy none 20:00 29 17 19 264 (W) 569 ptly cldy none 21:00 28 20 22 264 (W) 495 cloudy none 22:00 27 22 13 250 (WSW) 518 cloudy none 23:00 27 20 18 267 (W) 618 ptly cldy none 00:00 25 21 12 265 (W) 605 ptly cldy lt snow 01:00 24 22 12 291 (WNW) 387 mstly cldy none 02:00 23 22 10 303 (WNW) 363 mstly cldy none 03:00 22 22 5 328 (NNW) 345 mstly cldy fog 04:00 23 21 13 324 (NW) 272 mstly cldy none 05:00 22 20 13 303 (WNW) 334 mstly cldy none 06:00 21 21 15 311 (NW) 426 mstly cldy fog Note: the wind was coming from the azimuth listed. The convention used is north=0 degrees, east=90 degrees. The dust count is particles > 1u per 0.1 cubic feet. TELESCOPE STATUS Telescope drives operational. Current TCC version: 2.10.0 Current TUI version: TUI 1.5.1