UT20090624 Telescope Performance Verification
with LUCIFER at LFBG

Observers: W. Seifert, N. Ageorges, V. Knierim, J. Hill (LBT)
Telescope Operator: A. Ceranski
Software Support: T. Leibold, C. Biddick (Tucson)
Instrument Support: T. Shih
Telescope Support: J. Little

Summary

Weather wiped out the first 3.5 hours of the night. After that nearly every problem you can imagine wiped out the next 4 hours of the night (~3 hours time lost to telescope). We had problems with networks delays talking to VxWorks. We had multiple problems with mirror support air. We had serious problems with the LFBG rotator faulting. We had problems with a failed disk on a TCS server. We had problems with pointing. We had problems with condensation on the LUCIFER entrance window. We had problems with the left shutter door latch pins. After 09 UT, things settled down and we actually got some work done measuring the LUCIFER rotation zeropoint on-sky. We started on AGW transformation data on StoneO, but ran out of night.

Preparations

First as IRTC,

02:20 - 02:45 Lots of various startup problems before we open the dome. RA/DEC not updating via IIF, perhaps because MCSPU was off. Both primary mirrors were panic'd. OSS had to be restarted to get M3 to collimate. After some puzzling,

Then as LUCIFER, via IRAF

02:47 pointto BS9150 ACQUIRE

02:48 John L. found the sunset highly photogenic, and it is clearing to the west, but a small shower off to the south is keeping us from opening.

Then as actual LUCIFER

02:53 preset BS9152 - SX mirror chase fails to make tolerance.

03:16 Move to zenith to exchange LUCIFER masks

Open

03:25 Open with T=12.7 degC, D=5.3 degC Wind calm, Sky is about 70% covered with thin clouds.

03:30 preset to a pointing star 12:42:15 +45:52:37 TRACK

Wed Jun 24 03:33:18.164 2009  Preset alert for left LUCIFER mode TRACK: left PMC Mirror GoToPositionAbs Command Failed because Chase failed to make tolerance
Wed Jun 24 03:33:18.214 2009  Preset command failed for left LUCIFER mode TRACK from IIF
Chris thinks this error happened only two seconds after the move started. So there seems to be a crack in the timing that has appeared where code hasn't changed in more than a year. DX VxWorks was cycling FeedForward on and off about this same time, this is also consistent with network delays.

03:34 resend the preset and it succeeds, but we don't immediately find a star.

Close

03:42 Close, as another cell is building just east of Swift Trail Junction.

03:47 SX Mirror panics for lack of air pressure. It appears to be correlated with the dryer venting.

04:03 preset with dome closed. no problem with air or network delays.

04:10 J. Little gets to see for himself the SX Air Dryer problem. It appears that the SX dryer controller (ADD 401) is opening both valves when it should only be opening one.

04:30 Open bypass, Lower DX mirror, Raise SX mirror on DX compressor.

05:20 Sky has mostly cleared, but now we are above the humidity limit. Walter starts a series of LUCIFER flexure tests.

05:50 Chris installs a patch for PMC that makes it more clever about watching for non-IDLE mirror states before waiting for the IDLE state to return. This works OK under tests initiated from PSF GUI with no obvious delay under normal circumstances.

Open

06:06 Reopen to mostly clear sky with humidity 89% (LBTO) and 94% (SMT).

06:12 Preset to pointing star 16:30:15 +47:57:08 and we see a star for the first time in several days.

06:15 SX Mirror Panics even though running on the right compressor. The right humidity bypass valve is chattering even though ECS is reporting both humidities below -70 degC. We restart the Left compressor, but it still has the previous problem.

06:39 Notice that LFBG rotator has faulted. Lost 5 min.

06:40 Aaron has jury rigged the humidity bypass value. Lost 25 min.

06:41 Preset again

06:42 CA=54.08 IE=56.0 (not very precise)

06:44 Send active preset - LFBG rotator faulted again, and then again before it got back to tracking.

06:49 Stop and restart MCSPU software. Lost 10

06:52 Send active preset - failed with an error message that Torsten has never seen activated.
Wed Jun 24 06:52:32.539 2009  Preset alert for left LUCIFER mode ACTIVE: full CCD readout could not be saved for external analysis. Processing impossible.

06:55 Send active preset - failed again with same message. Torsten thinks that something is hosed up in the lbtmu304 filesystem so GCS is not able to save the acquisition image to be analyzed by sextractor. It seems that /tmp has changed to a read-only file system sometime in the last few hours. It was working when we tested it earlier at 2:58 UT as we can see the previous image there. This is Issue 2252.

07:07 Stop GCS and restart it on lbtmu303. Load the config file.

07:08 Send active preset - took image, but couldn't find star.

07:09 Send another parallactic mode preset IE=47.875 CA=105.33 Something goofy is going on with pointing!

07:12 Send active preset - guiding successfully starts this time. In a few iterations, the image is down to 0.7 arcsec FWHM.

07:19 Send active preset - off-axis - another rotator fault. Aaron recovered before preset ended. We started guiding on a very faint star, but it wasn't the intended guide star.

07:22 Move to zenith to put encoders back in strip mode before we reset the pointing again.

07:25 preset in parallactic mode - rotator faulted again Lost 4 min - Field is too crowded to identify the pointing star.

07:32 preset again to a different field

07:33 condensation increasing on LUCIFER window - go to zenith at rotator angle 341. D=8.1 degC T=10.6 degC Close dome for this activity.

Close

07:36 Chris and Torsten moved the remaining subsystems PSFL, LSS, SDB off of lbtmu304 and stopped its network server.

Open

07:40 Reopen - outboard latch pins on Left Shutter Doors trip their breakers. Aaaaaaaaaaaaaaaaaaaaaaarrrrrrrrrrrrrrrrrrrrrrggggggggggggghhhhhhhh! 4 breakers on the outboard left latch pins tripped. Lost 18 minutes.

07:55 preset to pointing star TRACK 0 but now we hit the XY limit in M1 collimation. Cleared the M2 offload entry and will recollimate.

07:59 another preset IE=15.25 CA=115.6 but wasn't a reproducible result. Was this because the rotator had already faulted?

08:03 send active preset - rotator fault again - GCS finds only a faint star.

08:06 couldn't find star on axis either

08:09 preset to a pointing star AGAIN IE=32.875 CA=80.85 - Still wrong by 3 arcsec in CA. CA=77.35

08:20 send active preset - found star

08:26 Move M1 and M2 both by dY=-2.0 mm Global Offset.

08:27 send active preset - image improves to 0.9 arcsec.

Open-loop offsets

08:32 preset in TRACK mode to do open-loop offsets - LEFTZEROPOINT is -187.703

08:44 start set of larger amplitude offsets to measure the needed rotation

08:51 LEFTZEROPOINT set to -186.703. Stop and restart PCS.

08:52 send active preset to recollimate - still about 1 arcsec seeing This time working on a fairly faint star - apparently too faint for WFSing. GCSGUI was complaining about the faint guide star, but not making any complaints about the WFS images. Jun 24 08:59:48 lbtmu303 LBT_left_GCS: WFS: maxPeak: 1067, bkgrd: 1014

09:00 another set of open-loop offsets - LEFTZEROPOINT is -186.703

09:09 preset active off-axis

09:15 active offsets

09:21 preset TRACK still at -186.703

09:23 open-loop offsets again

09:30 Set LEFTZEROPOINT to -187.776 based on the previous offsets

09:32 preset ACTIVE - found a really faint guide star, so pointing must have drifted.

09:36 preset TRACK - far off. Azencoders were off by 2 arcmin from absolute. Reset them with azload elload at AZ=249 EL=90.

09:58 IE=242.875 CA=15 but that didn't go where we expected, so retrench again with a new preset.

10:00 IE=33 CA=77 - We're offsetting around this field trying to confirm the identity of

10:03 CA=-123

10:05 IE=100.6 CA=-36.5

10:07 IE=107.6 CA=-43 then make an RADEC offset to the correct star

10:15 preset Track Parallactic

10:15 IE=247.6 CA=-223

10:21 IE=327.6 CA=-43, but of course this was the wrong sign

10:23 IE=-113 CA=-43 - at last we have the correct star near the center (more negative IE moves the star down)

10:24 IE=-111.4 CA=-13

10:25 preset verifies pointing

So our pointing problems seem to have been a combination of 2 arcminutes of strip encoders different from absolute, 1 arcminute of actual pointing model change, and 1 arcminute of three people doing math with insufficient oxygen.

10:28 preset ACTIVE to a star at 21:04 +30:30 Hit +Y limit again on M1. Increase the Global Offsets on M1 and M2 to -4.0. Pointing was off about 20 arcsec, but we decide not to correct at EL=85.

10:36 preset TRACK 22:27 +19:16 Update IE=-94.8 CA=-23.25

10:40 Rotator faulted again

10:45 preset ACTIVE - seeing is 0.6 arcsec.

More open-loop offset tests

10:49 preset TRACK

10:56 Rotator fault

11:05 Have confirmed that we like LEFTZEROPOINT at the last adjustment of -187.776 based on analysis of a few stars with +-100 arcsec offsets open-loop

measurements of center of rotation.

AGW Transformations

11:19 pointto StoneOagw TRACK PAM- Adjust IE=-106.8 CA=-12.5

11:26 pointto StoneOagw ACTIVE PA=0 gs=21 - star is way too faint (for the 1.4 arcsec uncollimated image)

11:28 pointto StoneOagw ACTIVE PA=0 gs=1 forgot getdat

11:32 PA=0 gs=11 forgot getdat

11:34 PA=0 gs=4 getdat0

11:36 PA=0 gs=8 getdat0

11:38 PA=0 gs=1 getdat0

11:40 PA=180 gs=3 getdat0

11:42 PA=180 gs=5 guiding didn't stabilize? Sky too bright so gradient fools the algorithm, but 5 has a faint companion - no LUCI data

11:46 PA=270 gs=1 getdat0 - rotator faulted - image is ugly (too much light in WFS)

Sky Flats

11:51 Telescope to zenith, AZ=270. Rotator=341

Close

12:30 Close T=12.8degC D=2.2, Sky mostly clear, wind 4 m/sec.

-- JohnHill - 24 Jun 2009
I Attachment Action Size Date Who Comment
daq.20090624.gifgif daq.20090624.gif manage 46 K 08 Jul 2009 - 18:15 JohnHill Plot of Telescope and Mirror Temperatures
Topic revision: r6 - 08 Jul 2009, JohnHill
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback