UT 20150527 LUCI2 AO Nighttime Commissioning

Observer LUCI-Team: JHeidt, APramskiy, DThompson
Observer AO: N/A
Telescope Operator: SAllenson
LBTO Support: TSargent
Watchdogs: Al Conrad (LBTO), RGredel (Luci-team)

Plan for the night

- This and the following 3N+ will be used to verify LUCI2 readiness for science operations in seeing-limited mode

- Detailed plan depends on observing conditions (forecast says partly cloudy tonight)

DThompson will be the master doing the observations

Summary

- Quite some tests in ima- and spec-mode

- Quite some bugs found, some are removed, more to be done

- Several improvements are still possible

Details

All times are UT
Files are luci2.20150527.NNNN.fits

02:24 Sunset, mostly clear, a couple of cirrus around, Steve opens, low wind, T = 10deg, Hum = 30%, low wind

Pseudomono mode for SX M2 work in parallel

02:44 Steve checks pointing and collimates on BS9148

Initialize all units...OK
"Import Program" added to the "queue actions" dropdown. It is MUCH faster,
but the "queue actions" button should be "import program"

Seeing is really good. Going to a NB imaging program...

03:10 Bomans Mrk36 script (imaging H2)
immediate error from telescope, sync offset received, no preset
Authorize NONE on left
Preset worked. LUCI TelServ not set up with Pseudomono in mind

03:15 Re-sent Bomans Mrk 36 script, no guide star found...check pointing

03:22 Re-sent Bomans Mrk 36 script and started, images 1-100,
lots of suggestions discussed with Alexander

06:56 BS9164 HK + zJspec
three errors on FW2 moving to zJspec, init between cleared error but
immediately got another. Then there was an error on readout (countdown
timer went to -120s). FW2 was out of position. I had to move 50 steps
on the MCU GUI and initialize. Seems to work now.

Readout error again...countdown timer went way below zero.
Alex shutdown GEIRS and restarted it.

The last two exposures did not start cleanly. Instrument reconfigured OK
and the offset was executed, but the readout did not start and at least
once the readout did start but this was not correctly indicated on the
item in the queue. Alexander restarted the scheduler.

BS9166 Al executed this script (he also took notes, see bottom)

BS9170 Al executed this script (he also took notes, see bottom)

09:04 Norris script NGC6341_ima.xml
Script set up with very short exposures...4s so the WFS cannot keep
the telescope collimated. Insert pauses every ~5 exposures or so and
allow GCS to collimate the telescope at least one cycle. Also, only
the first preset had a rotator angle specified, the second one did not and this 2nd preset failed (bug in Scriptor).

09:49 NGC6543_H2_test.xml 4x30.0s 2 dithers, to check exposure times
Instrument changed from H2 to Ks between the exposures?
  1. 157 supposed to be H2
  2. 159 BrG
  3. 161 FeII
  4. 165 H2 (one sky by hand, re-do script)
    I cannot guarantee that I did not push the "submit" button on the
    instrument GUI in the middle and accidentally moved in the Ks filter.
    Both 156 and 157 are labeled "Ks" but 156 is definitely an H2 image
    and 157 looks like Ks (or at least a broadband image).

10:30 Cloudy.
10:49 Still cloudy, getting thicker. Closing since we are now past twilight.
11:13 Still cloudy, only 1 star visible on all-sky cam.

11:44 When preparing LUCI2 for sleep, the UI hangs (while trying to put in blind/blind filters)
xkill, and restart UI, OK

############################################

  • Notes for improvement:

Need full startup script (opens geirs, UI). Alexander thought it already exists.

Readout status panel info needs to be changed
- include countdown timer here if at all possible.

READ2 gui needs a "refresh" button to set the changeable
values to reflect the current configuration of the instrument.
Change one thing and hit submit can give you a whole different configuration.

Correct way to continue a script after it has been paused is to click the "commit" button.

"Subtract" on RTD of two images is taking >60s. May have been temporary, subsequent ones worked OK.

There should be a clear way to "Abort script" cleanly...on the Observer gui.

  • And here are Al's notes:

- The load script under queue actions should remember the directory you're in

- The queue action pull-down should be a button

- The instrument GUI \xA0has two columns: selections for desired on the
left and status of actual on the right.\xA0 There is a problem that
many undesired moves take place when you want to use this panel
to only move one element.\xA0 There are many possible solutions.
(e.g. a button that says "update left hand side to current state")

- More label changes
a) unpause => remove pause
b) synch => binoc synch
c) asynch => remove binoc synch

- Fewer digits of precision in ctime (may be removed or moved as part of a general re-work of this panel.

- In data save menu, improve clarity (e.g., only 3 choices: intg, cube, or norm; is one suggestion).

- Clarify labelling of context menu (includes the one above)

- Item list, clarify (is covered in Dave's hand-written "red" notes)

- In status panel, clarify (is covered in Dave's hand-written "red" notes)

- Clarify the status panel readout area (include count down timer)
(details are in covered in Dave's hand-written "red" notes)

- Need an abort button (seems to be no way now to abort)

-- %USERSIG{JochenHeidt - 2015-05-27}%
Topic revision: r5 - 28 May 2015, JochenHeidt
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