May 31 - June 2 Technical Observing Plan


  • Good Seeing Plan
    • 5/30: 1st half SX AdSec Astig LUT, AGw#1 Transformation (DLM, KB) (+AO rested shell test JH)
    • 5/31: 1st half Nonsiderial (ME), then AGw#1 (DLM, KB)
    • 6/1: AO close loop with CCD 47 (DLM, KB, JCG), then ActO (DLM, KB)
  • Poor Seeing Plan
    • Binocular Range‐Balanced collimation LUT as bad seeing backup.
  • 6/2 ‐ 6/5: AO commissioning (IQ, Arcetri)

Summary of 31 May 2012

Seeing was variable from 1.5" to 2" with a couple half hour stints of ~1.0". Not good enough for AGw#1 transformation data collection

Tasks performed
  • Collected two Adsec Astigmatism Hysteresis curves
    • One with Adsec LUT turned off to measure the hysteresis curve to see if it has change since it was measured last fall
    • One with the new AdSec LUT turned on to measure how well this new linear LUT compensates for the elevation dependent astigmatism. As we were collecting the data it looked like the LUT was down the middle of the hysteresis curve. That is, when we were decreasing elevation between presets, the Act Opt added negative (-)Z6. As the elevation increased with new presets, positive (+)Z6 was needed. It's not clear yet, but it may be best to clear C00 (or if possible just the Z6 component of the primary forces) for each preset so we are starting with the Adsec LUT each time we collimate, instead of have the astigmatism from the last preset still on the primary
  • Tested Source Extractor Active Optics Algorithm
    • Works well for stars down to 16th magnitude
    • New background subtraction method caused spurious Z11 for dim (<15th mag) stars. My have to trash it
    • Old 4-corner background subtraction works well for source extractor for conditions tested
  • Tried to collimate with shell rested
    • Could improve rested guide image with several collimation iterations, but could not get SH spot in all subaps (about half or a bit more) so the core of the guide image got more concentrated, but still had a large halo.
    • Tried to limit the number of SH spots found (finding spurious spot in empty subaps), but need to work on algorithm a bit more
    • Can try to send Z12-21 along with current zernikes to see if image improves.
  • Olga tested the Binocular Range Balance routines for Gregorian focal stations. We had LBC authorized on the right so the RangeBal failed. May try to get RFBG authorized on the right tonight or tomorrow night for another try at testing

Preparation :

DONE Install Source Extractor Algorithm in IDL(Doug)
DONE Create New Stone Fields (Tina)
DONE Check IDL collimation Routines, specifically /IDL flag (Doug)
ALERT! Add flag to send all Zernikes from 4-22 for Rested Shell collimation (Doug)
DONE Contact Jesper (Doug)
DONE Test transform_process on old IRTC data (Doug)
ALERT! Add Binocular Commands to IDL (Doug)

Observing Plan

Confirmation of SX Adaptive Secondary Astigmatism Look-up Table

Tina finished the data mining for the Total Astigmatism due to Adaptive Secondaries last Friday and Runa confirmed that the Astigmatism applied from the LookUp Table are in rms Wavefront. The plots attached are from these data and show that the current Astig LUT applies about 1/2 the needed amount. Runa and I installed a new linear LUT (see plot) on Adsec SX and spent a few minutes and a few presets on Friday night to confirm that the guide star stay mostly round (2:3 ratio at most) for presets from 75 to 53 elevation.

If this new LUT is better by a factor of ~2, then an iteration or two (ie a minute or two) could be saved on some presets during science with LUCI and MODS. I think it with be worth ~4 hours of technical time to confirm the improved collimation during science time.

The three hours would be to:

DONE AdSec LUT OFF, Collimation OFF
(30 minutes) preset to 10 objects from 30 to 85 elevation, with collimation and LUT turned off, to understand how much the ellipticity of the guide star changes. The high, low, high, medium, low, medium, etc to understand the magnitude of the hysteresis. Basically remeasure the Astig hysteresis curve.
DONE AdSec LUT ON, Collimation ON
(2 hours) preset to 10 objects from 30 to 85 elevation with LUT and collimation (active optics) on to measure the total astigmatism due to the Adsec. The forces on the primary will be cleared with a C00 during each preset so we are sure the Active Optics + LUT is the total astigmatism.

BTW, these tests can be performed in poor seeing, 1.0-1.5". Of course good seeing would make the results more definite, but is not needed.

We will collect five sets of collimation data to confirm three things:
  • Compare Center-of-Weight with Source Extractor
  • Compare collimation with and without Clear C00 at each Preset
  • Compare Old and New AdSec Astig LUT
The purpose is to have consistent sets of data where only one parameter has changed. Each data set will consist of 14 stars at the same, various elevations of

To do this we will collect five sets of data, with the following parameters, in order of priority:
ALERT! New LUT, Collimate with GCS/COW, Clear C00
ALERT! New LUT, Collimate with IDL/SEXT, Clear C00
ALERT! New LUT, Collimate with IDL/SEXT, without Clear C00
ALERT! Old LUT, Collimate with GCS/COW, Clear C00
ALERT! Old LUT, Collimate with IDL/SEXT, Clear C00

# of iterations 1 2 3 4 5
without clear C00 1/14 (7%) 4/14 (29%) 10/14 (71%) 13/14 (93%) 13/14 (93%)
with clear C00 0/13 (0%) 2/13 (15%) 9/13 (69%) 9/13 (69%) 11/13 (85%)

# of iterations 1 2 3 4 5
without clear C00 1/14 (7%) 7/14 (50%) 12/14 (86%) 13/14 (93%) 13/14 (93%)
with clear C00 3/13 (23%) 10/13 (77%) 13/13 (100%) 13/13 (100%) 13/13 (100%)

# of iterations 1 2 3 4 5
without clear C00 2/16 (13%) 7/16 (44%) 11/16 (69%) 13/16 (81%) 15/16 (94%)
with clear C00 0/10 (0%) 3/10 (30%) 8/10 (80%) 9/10 (90%) 9/10 (90%)

-- DougMiller - 29 May 2012
Topic revision: r6 - 23 Jun 2012, KonstantinaBoutsia
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