LFBG Re-Commissioning with LUCI: 15 September 2010 UT

Observer: DThompson, JHill (LBTO)
Telescope Operator: SAllanson and GBechetti (LBTO)
Telescope Support: JUrban (LBTO)
Instrument Support: DOfficer (LBTO)

Summary

Lost 3 hours at the beginning due to problems starting PMCL. Things settled down after that and we got a bunch of data. We adjusted the pointing model zeropoint. We took two short sets of field aberration data, and 44 stars of transformation data. Conditions were superb with average seeing on the guider about 0.4 arcsec for the second half of the night. Dave had some problems with LUCI images that seem to be the LUCI optics not updating properly. We made a precise adjustment of LEFTZEROPOINT. We took some data for SX primary mirror verification.

Preparations

Tried to open all the A/C vents in the control room to reduce the hissing noise. More work is still needed as there are presently two sets of vents directly above the operator. (good thing that we've switched to the Canadian operator.)

Tony has turned on telemetry, so we need to restart some (all?) subsystems.

Michael was in the SX cell working on the remaining unconnected jet ejectors.

Jeff says we should keep the vent doors closed since the shields are still off the oil pads.

01:42 Steve and Geno are restarting all the subsystems to be sure we've got telemetry.

Details

It has been cloudy all afternoon, but clouds are dissipating as the sun sets.

Open without starting

01:46 Open shutter doors. Sky is mostly clear, but with scattered low clouds about. T=12.1 degC D=6.9 degC. Light wind from NW.

01:50 Steve having trouble starting PMCL. It makes telemetry events, but never finishes starting.

01:58 Reboot SX VxWorks

02:01 Try PMCL again.

Sep 15 02:01:29 tcs3 LBT_networkserver: PMCL died unexpectedly
Sep 15 02:01:29 tcs3 LBT_networkserver: PMCL is not active

02:03 Try PMCL on tcs2, same results.

Call Michele who contacts Tony, but we can't reach Chris.

Close

02:36 Close shutter doors to work on Right Hardpoint2 and Stow pins. That only took a couple minutes to sort out. The hardpoint needed to torque switch ring wiggled. And the Stow Pins were accidentally turned off at the SCP.

Wasn't all this stuff supposed to happen last night?

03:00 We've loaned half-an-hour of zenith time to Rick to check MODS fibers. Michele and Tony are still scratching their heads. Rolling back to the previous version of Telemetry didn't help at all. In the meantime, Steve and Geno are reconfiguring the right side for LBC-Red observations.

Balance Issues - Nothing is Simple

The telescope didn't come out correctly balanced after reconfiguring the right swing arms. DAshby quickly figured out that this was because MCSPU thought the AdSec was mounted on its swing arm. It gave Steve and Geno lots of experience in configuring swing arms and balancing.

Open

LBC-Red Turn-On

03:45 Reopening for some LBC-Red checkout observing. Sky is mostly clear.

03:52 Power on LBC-Red.

03:58 SA113_grizY Unfortunately the BOTH preset failed.
Wed Sep 15 03:58:16.656 2010 Preset alert for both LBC mode TRACK: PMC left not running Wed Sep 15 03:58:16.682 2010 Preset command failed for both LBC mode TRACK from IIF

04:05 Register LBC at 'prime right' using IRAF. Then send a preset with pointto to BS9177 on-axis.

04:12 dofpia, /redonly
Couldn't find images at first because lbcobs had /Repository instead of /newdata. (nobody has run dofpia as lbcobs in a while.) I fixed this in .cshrc.

04:25 wierd NVIDIA error, maybe because I was running IDL as user lbcobs on user LBTOs display

04:28 dofpia, /redonly as LBTO, but no pupils of any kind visible

04:33 pointto M5_1568

04:36 dofpia, /redonly still no pupils

04:39 10images5secV (full science array)

add 3000 nm Z4

04:49 10images5secV (full science array)

Too bad, the LBC-Red front cover is still on.

But in the meantime, CBiddick and friends have succeeded in starting the SX Primary by some trick. We are expecting an explanation to emerge later in the light of day. Note: It turned out to be a bug in telemetry r10. See Issue 2832 for details.

05:03 25 biases with LBC-Blue and Red

Initial Pointing with AGw3 and LUCI

Had to reconnect LUCI to TCS since IIF was restarted.

05:11 Preset to ACT0255 with LUCI in TRACK mode

05:15 Operator restarts MCSPU to get the rotator woken up.

05:16 Preset to ACT0255 with LUCI in TRACK mode

05:19 Ran ptcorr 2 to do the pointing correction.

   Current IE = -84.5313
   Current CA = -61.8785
   Corrected IE = -74.8701
   Corrected CA = -25.7513

Since these offsets are quite similar to last night's, we've installed them permanently in the pointing model LUCIFERSX.20100915.ptmod and updated PCSInstrument.conf accordingly. PCS was not restarted at this time.

05:20 Preset BS9183 ACTIVE

05:28 Put arbitrator in Primary mode, and clear active optics. Seeing is 0.6 arcsec on Guider.

At EL=60.8 Xtotal=-0.536 Ytotal=+1.302 Ztotal=1.848 RX=+43.16 RY=-19.91
At EL=60.5 Xactive=-0.684 Yactive =2.196 Zactive=+0.186 RX=+23.73 RY=+6.83

Planning to add those to the collimation model, but haven't done it yet.

LUCI 00005 before instrument initialize (not very impressive images)
LUCI 00006 after instrument intialize (focus looks better)

Trying Transformation Data on BS9183

05:48 Resend the BS9183 preset in GUIDE mode - seeing 0.5 - 0.7, then collimating with agw_collimate_loop

% Compiled module: SXDELPAR.
% Expression must be a scalar or 1 element array in this context: RETVAL.
% Execution halted at: AGW_COLLIMATE_LOOP   74
  /home/LBTO/idl/wfsc/agw/agw_collimate_loop.pro
%                      $MAIN$          

IDL> agw_collimate
% Compiled module: AGW_COLLIMATE.
/OldRepository/AGW_Data/100915

Acquiring WFSC Image: should take about 30 seconds....
    ...Done: Filename = /OldRepository/AGW_Data/100915/wfsc000002.fits
% Expression must be a scalar or 1 element array in this context: RETVAL.
% Execution halted at: AGW_COLLIMATE      79
  /home/LBTO/idl/wfsc/agw/agw_collimate.pro
%                      AGW_COLLIMATE_LOOP   74
  /home/LBTO/idl/wfsc/agw/agw_collimate_loop.pro
%                      $MAIN$          
% Program caused arithmetic error: Floating underflow

It turns out that Doug changed the names of his IDL commands, but he didn't update the instructions. New command is collimate_multiple, 10, /send

GCS version of field data

This is GCSL sending wfs corrections at all positions, so you have to integrate the differences.

06:02 pointto BS9183 gs=0 (all these are ACTIVE)

06:05 pointto BS9183 gs=60 PA=0 (first correction was 524 nm)

06:08 pointto BS9183 gs=0 PA=0 (first correction was 657 nm)

06:12 pointto BS9183 gs=60 PA=39.56 (first correction was 417 nm)

06:16 pointto BS9183 gs=0 PA=0 (first correction was ??)

06:21 pointto BS9183 gs=51 PA=161.45 (first correction was ?? )

06:26 pointto BS9183 gs=0 PA=0 (first correction was 268 nm)

06:31 pointto BS9183 gs=17 PA=311.36 (preset failed star out of range)

06:33 pointto BS9183 gs=14 PA=297.55 (first correction was 448.58 nm)

06:40 pointto BS9183 gs=0 PA=0 (first correction was 578 nm )

06:43:47 Stop WFSing

06:44 collimate_multiple, 2, /send (worked OK)

LUCIFER Orientation check. Images are all 6x10.0s Ks

06:50 BS9183 PA=0 0014, 0015=sky sources elongated
06:54 BS9185 PA=180 0016, 0017=sky ~2.8 pixels FWHM
07:02 BS9186 PA=0 0018, 0019=sky
07:09 BS9188 PA=180 0020, 0021=sky 0022=init (collim looked bad on 0021, definite improvement after init)

15:07:16 Error on LUCIFER while trying to compensate flexure at EL=84.33 FC turned off.

07:24 BS9107 PA=0 0023, 0024=sky, 0025=after init EL=43.6

mean change of zero point is -0.071 degrees over various fits to the two densest fields (BS9107 and BS9183). This differs from last night's zero offset, but with more data. Check again later.

More Field Data with IDL

/OldRepository/AGW_Data/100915/

07:37 pointto BS9101 gs=0 ACTIVE - seeing is 0.5 arcsec on guider

Persson's Star = BS9101
Start time = 07:38

for the off-axis points, collimate_multiple, 3

  => StepFocus sent to PSF
% Compiled module: IRC_STEPFOCUS.
===> Error in irc_stepfocus
        err = Usage:         err =      StepFocus RELATIVE-POSITION[mm] OPE SIDE
        err = 
        err = StepFocus moves the focus position by moving the OPE (M1 or M2 or M3 or M1M2)
        err = the given distance in the direction of the telescope's Z axis.
        err =          err =    i.e StepFocus 1.0001 M1 left         err = 
===> Error in irc_stepfocus
% Program caused arithmetic error: Floating underflow

Guide Star	wfscimage000*
0	05-08
4	is a galaxy
8	09-11

Don't use images 05-11. Stopped data collection to fix a missing space bug in irc_stepfocus.pro

Start time = 08:02 - Seeing is better than 0.4 arcsec on guider, wfs noise less than 80.
This is IDL sending corrections only on-axis in the traditional way.

Guide Star	wfscimage000*
0	12-15
8	16-18
0	19-22
7	23-25
0	26-28
6	29-31
0	32-34
9	35-37
0	38-40
10	41-43
0	44-46
3	47-49
0	50-52

End time = 08:47

LUCIFER Orientation check. Images are all 6x10.0s Ks

08:46 BS9107 PA=0 6x10.0s Ks for all data. EL=60
Files 0026 on source, 0027 = sky, 0028 after init, 0029 again, all offset Net offset of -0.05 degrees, applying +0.05 to LEFTZEROPOINT and restarting PCS.

09:10 BS9107 check after updating LEFTZEROPOINT 0030 - 0032

09:18 LEFTZEROPOINT now -187.764 degrees in PCSInstrument.conf.

Transformation Data

09:40 pointto StoneCagw gs=0 ACTIVE PA=0 - seeing 0.4 arcsec

GS	PA	UT	LUCI#	LUCI Xcen Ycen
20	0	09:46	0040	1026.06 1040.80
19	0	09:48	0041	1028.84 1040.25
16	0	09:50	0042	1030.28 1045.04
17	0	09:52	0043	1035.58 1043.47
18	0	09:55	0044	1034.51 1041.46
5	0	09:57	0045	1032.77 1049.62
2	0	09:59	0046	1027.86 1051.68
21	90	10:01	0047	1023.63 1035.79
20	90	10:02	0048	1025.34 1044.06
21	110	10:04	0049	1025.35 1038.21
20	110	10:06	0050	1024.13 1045.25
13	180	10:08	0051	1033.34 1040.30
7	180	10:10	0052	1029.35 1034.35
13	200	10:11	0053	1036.20 1043.25
10	200	10:13	0054	1037.48 1044.08
11	200	10:14	0055	1036.09 1044.61
12	200	10:16	0056	1035.71 1043.27
14	200	10:18	0057	1026.80 1040.52
21	200	10:20	0058	1032.66 1046.80
7	200	10:22	0059	1033.42 1037.10
8	200	10:24	0060	1037.74 1041.72
3	200	10:25	0061	1026.80 1034.42
8	270	10:27	0062	1035.41 1050.26
9	270	10:29	0063	1033.28 1051.06
12	270	10:31	0064	1032.68 1049.12
13	270	10:32	0065	1033.30 1049.37
1	270	10:34	0066	1039.43 1042.31
2	270	10:36	0067	1038.30 1040.59
3	270	10:37	0068	1038.42 1040.87
6	270	10:39	0069	1039.61 1045.56
7	270	10:41	0070	1039.38 1046.17
4	270	10:43	0071	1039.22 1037.68
5	270	10:45	0072	1036.85 1036.73
14	270	10:46	0073	1032.84 1042.36
16	270	10:49	0074	1030.51 1037.87
18	315	10:51	0075	1032.68 1034.87
17	315	10:52	0076	1035.88 1035.54
4	315	10:54	0077	1041.80 1045.13
5	315	10:56	0078	1039.40 1042.28
1	315	10:58	0079	1036.81 1048.05
2	315	11:00	0080	1037.33 1045.88
3	315	11:02	0081	1037.34 1046.62
14	315	11:04	0082	1030.94 1043.50
16	315	11:06	0083	1033.96 1039.99

GS1 might be a binary

SX cell checkout

Still on StoneC at EL~58

Reference is 00083 above
10 arcsec Tip 00084
10 arcsec Tilt 00085

Reference is 00086
300# Z-bias is 00087
0# is 00088
100# Z-bias is 00089
0# is 00090
300# Y-bias is 00091 (image looks strange) See Issue #2838
0# is 00092
300# Y-bias is 00093 (image looks strange) See Issue #2838
0# is 00094
10000 MX-bias is 00095
0# is 00096
10000 MY-bias is 00097
0# is 00098

Guiding Tests

11:35 Guiding resumed, but showing a systematic bias to one side on the GCSGUI. Why did it just start doing it now?

11:43 Changed guide correction factor from 0.25 to 0.5 -- not much difference.

11:50 Stop sending guide updates to PCS from GCS, still taking images. We don't see any appreciable drift. In fact, it is hard to tell that we aren't guiding.

11:57 Resume sending guide updates to PCS

12:00 Moved IE from -75 to -77, Guider responded correctly to move it back, halfway at a time. But, it still shows a bias to the lower left.

12:04 Changed guide correction factor from 0.5 to 0.9. This makes things a bit jumpier, but it still shows a bias.

12:11 Resend the preset to star 16 at PA=315 x=36.985 y=542.351 This seems to have recentered the guiding. Maybe this is how it is supposed to work when the guide probe hasn't moved for an hour? Perhaps the position of the green dot should get replotted, but we don't think GCS knows the moving spot position.

Verification of hotspot alignment (modulo the wierd guiding, and off-axis star)

See the tadpole WFS images starting at 12:10. More obvious at 12:14 with good guiding. The right 2/3 of the pupil has well centered spots. The left third of the pupil has the spots displaced toward the right edge of the tadpole illumination. I believe this is because we are guiding off-axis, and there is some residual aberration on the well-corrected pupil. Allowing for the expected off-axis aberration, we find that the spots are well-centered in the apertures.

Wavefront RMS is now about 70 nm - a nice stable night.

Close

12:20 Close for impending dawn. T=11.2 degC D=3.1 degC. Sky is clear.

12:46 Changed guide correction factor from 0.9 to 0.5 in AIP_L.cfg. Proposing 0.5 to be the new standard value.

-- JohnHill - 14 Sep 2010