UT 2021 Nov 04 - Split Night PETS/PEPSI - D/E

Night Info:

  • D/E unti UT 06:30 (23:30 UT)
    • Reconfigure for PEPSI by 06:00 UT (23hr local time) if using LBCs, 06:15 (23:15 local time) if using MODS or LUCIs
  • Sunset 17:32
  • Twlight Ends 18:50
  • Twilight Begins: 05:16
  • Sunrise: 06:35
  • Moon is nominally down most of night

Night Plan:

  • DONE LBTI AO testing until ~20:00hr
    • some clouds
    • seeing up to 1.5"
  • LUCI-2 AGW transform confirmation collection up to 1.5hr (20:15hr-21:15/21:45)
    • some clouds
    • seeing up to 1.2"
    • Possible lists (all visible during 1st half):
      • Stone_A3_LUCI.list
      • Stone_P1.list
      • Stone_O_0.list
  • mid-z ULIRGS MODS-bino - 21:15/45 until handover
    • after AGW-2 transform should stay with LUCI or MODS (~1hr lost to config time to go back and forth with LBCs)
    • FF0234-0139 - 1.25hr w/overheads
      • filler science program
      • OT: /home/lbto/Service_Observing/2021B_D_E/MODS/Science_Rothberg/midz_ULIRGs_V6.xml
      • seeing 1.2" or better
      • thin clouds
      • Feige 110 SpecStd
  • If completion of tasks leaves 30min or less before PEPSI handover - then reconfigure for PEPSI early and double check pointing.
  • Contingencies
    • DONE if AGW-2 can't be executed for whatever reason, then 2.5 hours could be allocated to TMS in good conditions
      • start at 20:30 after reconfigure from LBTI
      • finish at 23:00 in order to reconfigure to PEPSI/PETS in time.
  • Poor Weather Backup
    • if seeing too poor for LBTI and LUCI-2 AGW transform - LBC TMS testing -3hrs
    • LUCI-Bino or MODS-Bino Poor weather merger backups
      • Can be a filler in any conditions
      • any clouds
      • up to 2" seeing
      • if TMS/LBCs unavailable and conditions not clear
      • Targets:
        • OT File: /home/lbto/Service_Observing/2021B_D_E/PoorWeather/LocalMergers/COBandhead_Mergers_V22.xml
        • Center on nucleus of targets (finder charts provided - only a single nucleus for all sources)
        • MODS
          • Arp 230 -Dual Grating MODS 0.5hr-1hr minimum (visibile 19:30 thru handover)
          • SpecStd - Feige 110
        • LUCI
          • Arp 230 - LUCI - G210 0.5" slit CWL 2.27 - 0.5hr in script (double if thicker clouds and/or poorer seeing and/or mono) (visibile 19:30 thru handover)
          • UGC 11905 - LUCI - G210 0.5" slit CWL 2.30 - 0.5hr in script (double if thicker clouds and/or poorer seeing and/or mono) (twilight thru handover)
          • NGC 6598 - LUCI - G210 0.5" slit CWL 2.30 - 0.5hr in script (double if thicker clouds and/or poorer seeing and/or mono) (twilight thru 21hr)
          • NGC 7585 - LUCI - G210 0.5" slit CWL 2.27 - 0.5hr in script (double if thicker clouds and/or poorer seeing and/or mono) (visibile twilight thru handover)
    • Photometric Standards
      • any seeing, must be clear
      • Bino-LUCI
        • /home/lbto/Service_Observing/Restart_2021/LUCI/LUCI_2021_Restart_V1.63.xml
      • LBCs
      • MODS (lowest priority as a complete set has already been taken recently)

Night Log:

03:00 Steve is reconfiguring to the LBCs.

TMS Engineering Plan:

Andrew proposed the following plan, which "would both have TMS operating as it should in science support and put it through some elevation-extreme paces, and get us a maximum amount of FPIA information with a minimum consumption of observing time."

  1. Select some high and some low fields.
  2. DOFPIA at first field and run TMS in normal way, active (note our old friend channel 11 is misaligned, setting the offset automatically ignores missing channels)
    1. slew to copointing field which will be in focus
    2. correct copointing
    3. set TMS reference, after which we can start TMS active loop
    4. slew to science field and observe for 30-min
  3. After 30 minutes, pause TMS and do vfastextra
    1. run preppupils.py to inject z4/z11
    2. run RB_fastextra4.ob (16-sec may need to use scale factor to increase)
  4. Start TMS active again (using original offset not a new offset).
  5. Go to second field (if first was high, second low).
  6. On obtaining track, wait for a couple cycles of TMS corrections and then pause TMS, do vfastextra, resume TMS, spend 30 mintues doing science like OB on field.
  7. after 30 minutes, pause TMS, do vfastextra, resume TMS.
  8. repeat 4-7 oscillating between high and low fields.

If image quality obviously degrades, interject with DOFPIA at any point in this sequence and set new offset.

TMS E Log:

03:15 Tried to copy reference files into /tmp, but permission denied. There were some files of the same name owned by LBTO. I deleted them and then copied the ones:
  • tms_{s/d}x.dat
  • telescope_{left/right}_tmsoffset.dat

03:16 Slewing to GD 2.

03:20 Run TMS active: tms_loop.py

**=> requesting a TMS vector over TCP...
timed out

Heejoo rebooted the software.

03:35 tms_loop.py ran for a few iterations before cntrl-C'ing. This populated Group Rotation values which were all 0's at the start, as expected.

03:37 dofpia. First set of Zernikes are:

          Computed Aberrations (nm)              |   (arcsec)   | channel 
 Defoc   AstX   AstY   ComaX  ComaY   Sph        |  Est.   Est. |         
 (Z4)    (Z5)   (Z6)   (Z7)   (Z8)   (Z11)  (Z22)| seeing  WIQ  |         
  4470   -165   -453    178   -402   -222    -35   0.71   1.79    blue
  2041     87     34   -565    234     58    -31   0.52   0.85     red

03:48 copointing

COPOINTING: B=34832 R=34828
Pointing updates: delta_IE = -20.60", delta_CA = 9.62"
  Mirror updates:   dX(mm)   dY(mm)   dRX(")   dRY(")
              SX:    0.40     2.86    61.31    -8.57
              DX:   -0.53     1.40    29.96    11.26

03:53 set reference and start tms loop (or did I start loop after starting the observations? It doesn't matter)

03:54 Starting observations at B & R

03:04 TMS is not sending any corrections to DX. The new reference I just took contains channel 11. But it's a 'dud' channel and John is going to set up the "set_reference" script to exclude channel 11.

04:06 setting a new reference: ./lbt_set_tms_offset.py -s 2

04:07 ./tms_loop.py -- now it's sending corrections to DX.

04:07 There was an error with the blue camera which caused a delay in the script. IT 8516. Eventually the image was taken (041340)

021/11/04 04:10:32.965000 E B CAMERA CCDCTRL bad input voltage line 9 ( +3.72V instead of +4.8V+-1.0V)

04:20 Back-face Z4 corrections are working.

04:26 Stopped the B & R only script and loaded GD2_UBVRI.xml to cycle through filters as well as dither. On the last B & R images the FWHM was 1", 0.8"

04:35 GD2 crossed the meridian at 89.25 deg and the images are elongated. They don't show signs of rotational trailing but instead the elongation is the same across the focal plane, and different betwen blue and red.

04:37 double-images on blue 043741 but the next image has pretty round PSFs. On the red image 044019, PSF is finally looking more round, but around zenith passage it was very elongated. Now we're about 2 deg from zenith.

04:38 DX measurements are failing, but then they came back. Then SX failed. This 'see-saw' is something that has been seen before.

04:47 3.5 on blue and 3.0 red

05:11 paused TMS, ran preppupil.py to inject spherical on both sides. And then took a pupil image. SX hardpoint error (IT 8739) and mirrors hit limits. 051118 and 051120 are a pair of pupil images but with the mirrors at limits. The script had sent not only Z4/Z11 adjustments but also a -0.8 mm adjustment to Z. I need to fix this. Some discussion about how to get out of this. Steve resolved the SX hardpoint error.

05:15 Resent the preset to GD2 to see if that would help move the mirrors of the limits.

05:19 Realized that preppupils.py put in -0.8mm. Backed it out before taking the pupil images: New set of images: 051940, 051928

05:23 Remove z4/z11 injection. Run tms_loop.py for a few iterations (but it wasn't doing anything because Z was at -0.8mm).

05:33 Slew to J2052 (elev 40). DX vector incompatible with reference because we are missing channel 10. There was a pair of images taken, but no TMS corrections were made on DX. On SX, corrections were being made but image has elongated stars.

05:35 Back to GD2 copointing field with tms running in active mode. SX and DX corrections being sent.

05:42 lbc_set_tms_offset.py -s 2 now without channels 10 and 11.

05:42 tms_loop.py Slewing back to J2052 copointing field. Let the exposure be taken. SX is trailed. DX is in place. Took another pair of images: 054851 and 054 --- round

05:52 Inject z4/z11. Take RB_fastextra4

05:56 Then remove the z4/z11 corrections and play the science OB, J2052_BR. (elevation 35). The images have FWHM 4.8 pix and 4.35 pix and look round.

06:05 Turned off TMS laser. Let the exposures continue. 06:06 Stopped the OB.


TMS worked well to maintain collimation over 2 hrs and a large elevation change. The telescope and mirrors were in equilibrium and we had stable conditions, however, so the thermal Z4 correction did not get challenged very much.

We started at a relatively high elevation field, <~80 deg, (Clem & Landolt photometric standard star field, GD 2). We used a reference from 20211027 and ran TMS in active mode for a few cycles until it applied corrections (Group Rotation).

Then we ran dofpia. The first pair of pupil images were not that bad and collimation converged in 5 iterations. That was around 03:45 and we did not run dofpia again and observed until 06:00. We sent the copointing script, copointed and took a new reference. Then we started the active TMS loop.

We collected a series of 60-sec dithered images at B (LBCB) and R (LBCR) from about 03:45 to 04:26. Around 04:30, I switched to a script which cycled through UBV on LBCB and VRI on LBCR at each of 5 dither positions.

During this time the object crossed the meridian at an elevation of 89.3 deg. There were some images with elongated stars (but no arcs) just before and after meridian passage, but away from zenith the image quality improved. We ran the script to completion at 05:10.

At this position, we paused TMS and took a set of pupil images with Z4/Z11 injected so that they can be analyzed by lbcfpia offline. Some excitement when the script to inject Z4/Z11 also injected a -0.8mm Z offset and both mirrors ran into their negative Z limits when the script to take pupil images was run. I need to fix the script error.

After taking the pupil images, we started TMS again and slewed to a low elevation field, elevation <~40 deg (Clem & Landolt photometric standard star field, J2052). Once again we injected Z4/Z11 and took a set of pupil images to be analyzed by lbcfpia offline. We restored Z4/Z11 and then started a script to take a series of 60-sec B and R dithered images on this field and ended it just after 06:00 as it was time to reconfigure to PEPSI.

The code correctly read the face-back temperature difference and make calculations, but they were pretty small. Some of the last back-face corrections:

Back-face Z4 corrections:  SX=    -79.20  DX=    -22.40 (nm) from 5.7188 to 6.0869 hours UT
Back-face dZ corrections:  SX= -0.003002  DX= -0.000849 (mm) from 5.7188 to 6.0869 hours UT

Two operational "lessons":

  1. We learned, or just recalled, that the observer needs to wait before starting to take data after a slew, since the mirror might still be moving.
  2. Twice we had to adapt to a mismatch between the channels in the reference and the reliable channels: the initial reference did not contain channel 11, yet it was sometimes working today; and when we slewed to the low elevation field, channel 10 stopped working. In both cases, John edited the lbc_set_tms_offset.py script to remove the channel(s) and we took another reference.

-- BarryRothberg - 31 Oct 2021
Topic revision: r7 - 04 Nov 2021, BarryRothberg
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