20201227 UT --- Engineering plus PEPSI+LUCI transit

Summary:

  1. The MODS AGw alignment checks were performed and no adjustments; neither to the hotspot nor the on-axis stage position (agw.ini) file were made. For MODS1, the hotspot was very close to the rotator center, but for MODS2 it was off in Y (the stage can only travel from 0 to 204, and the rotator center is above Y=204 (see Com20180921)
  2. The SX-M3 alignment was completed (along with DX-M3 alignment). John summarized this as follows:
    "Tonight, 20201227 UT, we refined the M3 Tip-Tilt corrections
    for both LFBG and RFBG.  SX M3 was recently dismounted, and we refined DX M3 as well with better data on its peculiar elongated pattern
    
    On top of the previously applied SX selector correction of -1.800 deg,
    tonights correction were:
    
    SX relative corrections are Tip=+160 Tilt=-60. SX absolute Tip=-220 Tilt=-270 Sel=+28.000. New pupil position is 381.6,244.1.
    
    DX relative correction is Tip=+325  DX absolute Tip=-340 Tilt=80 Sel=-25.727.   New pupil position is 379,265.
    
    These were only applied to LFBG and RFBG, but the SX relative values might be
    a good guess for LBTI.
    
    OSS, GCSL, GCSR were restarted about 4 UT for the changes to take effect."
  3. The LUCI1 + PEPSI-DX transit observation of WASP-12 was observed.
  4. After the transit, we obtained no useful data as it was not long before we had to close for high winds.

The conditions were clear but the seeing was really bouncing around, 0.5-0.6" at the start of the night, but even then some excursions up to 2-3" at times. After reconfiguring to LUCI+PEPSI, it became less stable. Often the guider FWHM on SX was better than on DX. The winds were high and threatening during the transit observation, and, about an hour after it was over we closed for high winds.

MODS AGw alignment checks

00:47: Opening

00:56: MODS in observing mode and David is doing the pointing check.

01:07: On an 11th mag collimation star - too bright. Going to a 14th mag one.

Check guider hotspot

01:16: John and I are both examining the WFS images - ds9, GCSGUI and gcs_wfsc_display, /right (/left). The star looks pretty well centered in the lenslets..

The seeing has been bouncing around from 0.7 - almost 2"...

01:17 The verdict is that both sides look good. No adjustments are needed to the hotspot.

Is the hotspot on the rotator center?

01:20 Rotator center check:

  • Stop guiding on GCS GUI, and take a 'junk' image:
  • gcsclient readGuideCam left -e 1000 (and same for right)
  • Move rotators to 0 deg (lost preset as expected)
  • Offset telescope:
    • ptincrement {0,1} CA -5
  • gprobe 0 0 on MODS GUIs (it was far)
  • started a pair of guider images on left, right
    • gcslient readGuideCam -e 90000 left and (oops - did 2 left ones)
    • gcsclient readGuideCam -e 90000 right (360->0). and few seconds later,
  • initiated slew of the rotators from 0-> 360 (for right reverse)
  • analysis:
    • MODS2:
      • right_guider000001.fits DETSEC = [85:449,51:424]
X,Y = 181.5, 163.5 
X', Y' = 265.5, 213.5
hotspot: 267.0, 240.0… 

MODS2 off, but John said one of these is off on purpose. Com20180921: on MODS2, Ycenter must be between 0 and 204 "ERROR: Ycenter value: 204.36 invalid, must be 0..204"
    • MODS1:
      • left_guider000003.fits DETSEC = [76:447,22:379]
X,Y = 175.5, 177.2
X',Y' = 250.5, 198.2
hotspot: 254.8, 199.5 

MODS1 looks good

M3 alignment:Pupil Wobble Measurements for LFBG and RFBG - JMH/OPK/DLM

01:36 Reconfiguring from binoMODS->binoLUCI

01:50 LUCIs are ready and authorized.

Warning - problems with tip-tilt requests for left side - wrong pointing star.

Pointed and collimating now.

02:14 Collimated on the right side. John is stopping WFS on DX.

02:17 Collimated on the left side. John stopped WFS on DX.

FWHM = 0.8"

02:18 John is running pupil_collect, EXP=15.0 in ~lbto/data/20201227_lfbg (and rfbg for right)

02:27 First set done on both sides. 6 rotator angles. Note - plot labels (m3grace.py) both say LFBG...could be confusing.

SX-set1:

02:28 On SX - from analysis of first set,

+tip 150" needed (all tips, tilts in arcsec)

/lbt/lbto/supportscripts/MetrologySupport/m3tt.py -s SX -p 150 -t 0

02:32 John running pupil_collect on the left.

DX-set1:

02:33 On DX - from analysis of first set,

John estimates right needs +100 tip.

m3tt.py -s DX -p 100 -t 0

02:37 John started pupil_collect on DX

SX-set2:

02:43 SX - second set -> need tilt -50

m3tt.py -s SX -p 0 -t -50

Total tip, tilt now +150, -50

02:44 John starting pupil_collect on SX

DX-set2:

DX - second set -> need +100 more tip

m3tt.py -s DX -p 100 -t 0

Total tip, tilt now +200, 0

02:51 John starting pupil_collect on DX

SX-set3:

03:00 SX - third set - diameter about 3.5 pix in Y and 1 in X.

m3tt.py -s SX -p 10 -t -10

Totals tip, tilt now 160, -60

03:02 John starting pupil_collect on SX

DX-set3:

03:05 DX - third -> need +100 more tip (RFBG still has an elongation issue)

m3tt.py -s DX -p 100 -t 0

Total tip, tilt now +300, 0

03:06 Starting pupil_collect on DX --- seeing bouncing up to >~1" now, from ~0.6-0.7".

SX-set4:

03:13 SX - fourth set -

m3tt.py -s SX -p 10 -t -10

Totals tip, tilt now 170, -70

03:13 John starting pupil collect on SX (5th set)

DX-set4:

03:19 DX - fourth set - needs another 50 tip

m3tt.py -s SX -p 50 -t 0

Totals tip, tilt now 350, 0

SX is done.

The fifth set looks a tad worse than the fourth. John will enter the tip, tilt values from the 4th into the config file. This result is dominated by measurement errors, although Doug is thinking of ways to make his pupil measurement algorithm better. SX relative corrections are Tip=+160 Tilt=-60. SX absolute Tip=-220 Tilt=-270 Sel=+28.000. New pupil position is 381.6,244.1.
DX - fifth set

The elongation pattern is rotated wrt the 4th set. Split the difference (bt 4th and 5th). This result is dominated by the unexplained elongation of the pupil wobble pattern. We split the difference where the pattern rotated 90 deg. DX relative correction is Tip=+325 DX absolute Tip=-340 Tilt=80 Sel=-25.727. New pupil position is 379,265.
Summary of M3 adjustments:
  SX M3 global offsets DX M3 global offsets
seq tip(cum) tilt(cum) tip(cum) tilt(cum)
1 0 0 0 0
2 150 0 100 0
3 150 -50 200 0
4 160 -60 300 0
5 170 -70 350 0

The final two sets for SX are below:

The final two sets of DX are below (yes, the plot says LFBG, too):

03:33 David is authorizing LUCI1 only for Doug's test.

03:35 John is entering the M3 tip, tilt values in the config file and will restart OSS after Doug's test. GCSs were also restarted for slightly modified on-axis pupil positions.

03:38 Doug running AO bin 3. trying to close the loop, the AdSec went into a failure.

Pupil & FS checks after M3 adjustments

03:49 David is pulling the guide probe out of the beam on DX and I'll check the pupil and FS alignment now that we adjusted M3 tip, tilt. DX M3 global offsets now 320, 0.

I took the LUCI2 images listed below, but did not use them to make any adjustments.

LUCI2 (these were probably not taken at 210)  
0005 N1.8 N30FS PVlens+J Pupil Viewer (double swing-arm)
0006 Jhi+J PV out
0007 N1.7 N30FS PVlens+K pupil viewer in, needs adjustment but I will focus on LUCI1 now.

04:04 David is restarting OSS on both sides and GCS on both.

04:06 - authorizing for LUCI1 + PEPSIPFU-DX

David moving LFBG to 341 deg

I took the LUCI1 images listed below.

SX pupil + FS alignment     dx, dy (pix)
13 N1.8 N30FS PVlens+K PV in

P: 1142, 1100.5

CS: 1126.5, 1097

-15.5, -3.5
14 N1.8 N30FS clear+K PV out 1023, 1027.5 +1, -3.5
15 Jhi+J    
16 PVlens + J    

Based on 13 & 14, PFSalign calculated:

FM1m1 -908, FM1m2 -1826, FM4m1 -815, FM4m2 1117

I entered these values in the MCU gui and took the pair of images 17 & 18:

17 Pupil PVlens+J tiny crescent at top
18 Pupil clear + J 1023, 1023

Looks pretty good. I could not see the screws but after adjustment the crescent is a bit smaller but shifted to the top more than the left side. Lots of moonlight.

LUCI1+PEPSIPFU-DX transit observation: 04:30 - 10:30 UT

Summary:

The filenames are: luci1.20201227.0NNN.fits where NNN = 028-605. These covered UT 05:25:49 to 10:34:55. A late start due to LUCI1 acquisition issues followed by a filter wheel 1 error, but hopefully it gives enough baseline pre-ingress. The PEPSI sequence started earlier. The seeing was variable throughout and counts were not as high as predicted by the ETC, but 11-15k at the peak.

04:32 We're reconfigured for LUCI1 + PEPSIPFU-DX and David is slewing to the pointing star. Moving the WASP-12 mask into the FPU.

04:39 Pointing done, collimating now. The seeing is very poor (what happened?!)

04:45 Sending sync presets to field. SX guide star FWHM is about 1.5-3".

04:49 The PEPSI observations have started

The first set of acq images are 19, 20 and 21. The LUCI 1 FS is off. It was very well centered with the N1.8 camera, so I don't understand what happened.

I put the mask to turnout to check the field stop alignment with the N3.75 camera.

22: is the first test image - the field is off to the left, adjusted it via the AFC tab on the RTD.

23: OK - this image looks fine.



Now, retaking the set of acquisition images:

24: -5 -5 sky image - done manually to avoid sending a new preset

25: from 2nd item of script (skip to item and go from there).

26 will be the mask image

The alignment was difficult because the seeing was so poor. Used stars 30, 31 and 34 as these gave a better result than the 5, but only a minimal solution with 3 stars … (Note: I clicked send 2x so the offsets are twice in the observation panel, but I skipped to the 2nd set and ran the script from there so as not to double-up).

27 is the confirmatory through slit image. RTD slowdown and difficulties with incomplete subtracted images. Finally restarted the RTD and subtracted 27-24.

The bottom object is not very well centered up in the slit, but we're losing time, so I started the spectroscopic series. Called the PI to ask. The bottom object (northmost) is about 6 pixels (0.72") to the left of the slit center. The top object is centered and the stars look fairly well centered in the 4" boxes.



Below is the difference image, 27-24:

04:32 We're reconfigured for LUCI1 + PEPSIPFU-DX and David is slewing to the pointing star. Moving the WASP-12 mask into the FPU.

04:39 Pointing done, collimating now. The seeing is very poor (what happened?!)

04:45 Sending sync presets to field. SX guide star FWHM is about 1.5-3".

04:49 The PEPSI observations have started

The first set of acq images are 19, 20 and 21. The LUCI 1 FS is off. It was very well centered with the N1.8 camera, so I don't understand what happened.

I put the mask to turnout to check the field stop alignment with the N3.75 camera.

22: is the first test image - the field is off to the left, adjusted it via the AFC tab on the RTD.

23: OK - this image looks fine.

Now, retaking the set of acquisition images:

24: -5 -5 sky image - done manually to avoid sending a new preset

25: from 2nd item of script (skip to item and go from there).

26 will be the mask image

The alignment was difficult because the seeing was so poor. Used stars 30, 31 and 34 as these gave a better result than the 5, but only a minimal solution with 3 stars … (Note: I clicked send 2x so the offsets are twice in the observation panel, but I skipped to the 2nd set and ran the script from there so as not to double-up).

27 is the confirmatory through slit image. RTD slowdown and difficulties with incomplete subtracted images. Finally restarted the RTD and subtracted 27-24.

The bottom object is not very well centered up in the slit, but we're losing time, so I started the spectroscopic series. Called the PI to ask. The bottom object (northmost) is about 6 pixels (0.72") to the left of the slit center. The top object is centered and the stars look fairly well centered in the 4" boxes.

Below is the difference image, 27-24:

We had a LUCI1 FW1 error which I had not noticed during the phone call. It showed OH1060 and I initialized on the FW1 gui. Showed OH1060 after init. I right-clicked to continue.

05:26 Having recovered the error, starting the spectroscopic series now.

In image 0042, the counts peak around 10,000 ADU. The seeing has been ~2" however.

06:00 SX/DX FWHM 1.05, but still bouncing around a lot between 1-2". It seems to be consistently better on DX than on SX.

Image 0131: counts peak around 12k. The peak is in the H band, southern-most star.

Image 0143: There are two traces for each star. Was there a jump in TCS or LUCI1? Nothing unusual on the right side. The LUCI slit traces are rock solid - an SX mirror jump?

The left_guiderimage001400.fits shows the star in the upper left corner - UT 06:27:11.457.

Image 0213: 15k in the peak. Seeing ~1".

Just before 07:30: Crossed the meridian. Seeing SX/DX <~1" now. Winds have been trending up, but are still within limits, in the upper teens and a few gusts to 20 m/s.

07:34 SX/DX guide FWHM = 1.3"/1.2" but still varying from 1-2".

08:19 seeing spiked to 3".

09:49 seeing 0.8", down to 0.6" for ~1-2 sec. On image 520, 15k in the peak.

10:25 on the last PEPSI exposure.

The seeing has been gotten better but it is still not that stable. I don't see excursions above 1.5" often and mostly it is <~1". On SX/DX, avg FWHM ~ 1.2/1". At 10:31, it went back to >1.5 for a while.

10:34: the PEPSI observation ended. I'm aborting the LUCI script.

0606: Taking a through mask image with BrGamma to check the alignment. 606 is source and 607 is sky. With N1.8. Sky was at a -5, -5 relative detxy offset.

608, 609: sky, obj with N3.75 (to move to object, I gave a +5, +5 relative detxy offset). On 609, there is a wisp from one of the stars and not the other. I don't know what it is - I thought I had waited until the offset completed to start the exposure. Aside from the wisp, the centroid seems well-defined and the star positions appear as they did initially, but because of this feature, I would trust the set of N1.8 images more than the N3.75 ones for a check on the source position at the start & at the end.

AZ queue

10:47 Reconfiguring to binoLUCI

The seeing has been variable but trending down and it is a tough call. I think I'll go to the telluric for the high priority program and, as it is not at that high elevation, seeing may be worse than what I had seen and the decision may then be clear-cut.

10:52 Slewing to HIP56736 for EC52327. No star found on the right side, so David is checking pointing.

10:59 Resending the preset. Far from collim on DX, but on SX it is collimated and FWHM ~1.1-1.2" on average.

11:05 Now collimated on the right side and taking the sky and object images. I had not checked the pupil & FS on right side. We'll go to zen & rotator 210 to check.

On DX, moved to N1.8 and N30FS via IMGUI:

DX: 10 Jhi+J, 986.5, 1069 37.5, -45
11 PVlens+J PViewer in

P: 964, 1110

CS: 931, 1111

12 PVlens + K PViewer in CS: 952, 1102

The pupil is not perfectly aligned but it is close. And the FS on the N3.75 images looked OK.

Below is image 12 (right). I think it would not be a good use of nighttime to tweak it and I'll move on - though we don't have time for the high priority target and the seeing was variable anyway. I'll spend a bit longer adjusting this and move to the backup.

dX, dY for the pupil: -12, -08 (but later on I realized that I had I entered -80 in PFSalign and that is why 13 was so much worse)

dX, dY for the FS 37.5, -45

Based on these, the following commanded motions were sent to the MCU: FM1m1 -17730 (moved), FM1m2 1415 (moved), FM4m1 -12009 (moved), FM4m2 -7677 (moved).

13 PVlens + K - much worse

Backing out adjustments: FM1m1 +17730; FM1m2 -1415; FM4m1 +12009; FM4m2 +7677

14 & 15 confirm that things have gone back to where they were.

11:50 I was choosing a target from the AZ backup, but the winds came up and David said we had to close.

---+++ Closed Dome work

While we are closed, I'll check the pupil & FS on DX. The rotator is at 210.

16 N1.8, N30FS clear+K - rotators moving    
17 PVlens + K PViewer in    
18-28 are "junk" - normal mode, ndit>1 and not enough counts at K with no sky.      
29 N1.8 N30FS Jhi+J    
30, 31 PVlens + J    
32*

PVlens+J ndit=12

PVin

CS: 917, 1081

P: 942, 1081

dX,dY = -25, 0
33* Jhi+J PVout 986, 1071 dX, dY = 38, -47

With these inputs, PFSalign -> FM1m1 -4635; FM1m2 -145; FM4m1 -8454; FM4m2 -7409

34 Jhi+J PVout    
35 PVlens + J PVin

CS: 931, 1098

P:931, 1100

 
36 PVlens + J PVin   more lights
37 Jhi+J    
Now, move camera back to N3.75 via IMGUI.
38 Jhi+J 1093, 924 so the FS shifted in that camera motion, just as it did for L1 during the night

Zoom in on image 35 showing plug (not the plug on the rigid, but something else that appears to be centered within the rim) and M3. The green circle is centered at 931, 1100

Calibrations: WASP12

Lights off in the chamber, telescope & zenith and rotators in their park positions. I did not take the darks yet as these are usually taken before using LUCI, when there no persistence.

luci1.20201227.0NNN.fits   DIT x NDIT Comments

612-616

626-630

lamp off 8 x 2  

617-621,

631-645

halo3 on 8 x 2

4000-16000 cnts.

Repeated lamp-on flats.

622 lamp off 20 x 1  
623 arc 20 x 1  
624 arc 20 x 1  
625 arc 20 x 1  

-- OlgaKuhn - 27 Dec 2020

This topic: Commissioning > Main > WebHome > Com20201227
Topic revision: 27 Dec 2020, OlgaKuhn
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