Aug 10-12, 2012 TMAS First Installation on PALM-3000

Palomar Mountain, CA

Last weekend, Sergi, Jennifer M. and I took TMAS up to Palomar for its first interface with PALM-3000, in advance of our late September 'TMAS first light'. TMAS installed fine using the 2 forward SWIFT pucks and two TMAS-specifically attached rear pucks, though we found that AO output beam as science focus was 165mm from the bench (not 170mm), with the beam 'climbing' even closer to the bench (instruments hanging off the bench). We used our own TMAS FM5 (T-FM5) to steer our pupil though our relatively tight ADC clearance and were able to dial out any TMAS vignetting.

We hand tuned 8 Zernikes, and evaluated the PSF at both lambda = 900nm (see below) and lambda = 440nm. We obtained MGS data for Rick by moving our science camera back and forth on its translation stage (without MGS, we expect to be limited to ~ 100 nm RMS, which is 10% Strehl at 440nm). In the long run, it's my hope that the visible light camera can provide better MGS calibration of the high-order DM shape than can be made with PHARO.

We ran into trouble with the lodm_flatmap_wl, I think because the routine that creates these files relies on stable AO behavior. We ran DM_Reg after hand-tuning and before running the flat map routine, but our result was a pretty lousy flatmap. Eventually, we restored an older file, which was close. We've saved several files in p3k/tables directory having TMAS in the filename (and a date tag), including both flat maps and centroid offset files.

We also explored the image quality across the ~21" x 25" FoV. As expected from the design, there is some degradation in the corners of this FoV (but since it's less than the atmospheric isoplanatism will be at these wavelengths, it's acceptable. We also confirmed the stimulus beam walks off the stim fold mirror, vignetting, before reaching the corners of our FoV (FWIW, the walk-off did not appear to be centered on the ''regular' x,y position of the stimulus stage.

We imaged the TMAS pupil (using quite clean (e.g. not dirty)) TMAS optics and found plenty of amplitude variations. We also grabbed data a few millimeters away from the 'best focus' pupil image, showing the rapid emergence of HODM-induced high spatial frequency diffraction effects. I'm including below an image taken 4mm away from 'best pupil' (note, if I recall, our pupil is about 20 mm diameter (TBC)).

We confirmed that the ADC can indeed disperse light, but didn't have time to calibrate the motor rotation angle vs. dispersion (nor did we minimize the deviation under rotation). These will be topics for lab work in the coming month (as well as incorporating the Robo-AO atm dispersion predict model.)

For most of our work, we used the 50/50 beamsplitter (which we replaced with 960nm upon leaving). Compared to an 'beamsplitter-less' look at the PSF, there is little high-order change, even at 440nm wavelength, suggesting the 50/50 has excellent transmissive wavefront quality.

On the control side, we proved and made improvements to the Andor camera control code. A nice test was to take 200 image cubes of 200 frames each and having them shifted-and-added automatically (albeit without speed optimization). These fast cameras produce a lot of data, well, fast and I'm very encouraged by the capability of the GPU-based quasi-real-time processor. We still suffer from relatively low open-shutter efficiency due to bottlenecks both writing the data to our (cheap, non-optimal) disk storage and getting the data out of the Andor NEO camera (where we await an upgrade from Andor that's slipping month-for-month.)

Integration of the motor control, installation of the filter wheel, and fabrication of both a dynamic-range helping "OD mask" for the input focal plane are among our near-term priorities. For the 2nd TMAS run (Oct 27), we're also considering extension to either/both a coronagraphic test and a single-mode fiber feed test that could inform a collaborative test with Leidi.

As usual, Mike, Steve, Carolyn, and Jean were great and we're looking forward to returning TMAS to the mountain again soon.

-- Rich (for Sergi and Jennifer)


This topic: Palomar/TMAS > WebHome > TmasFirstInstallNotes
Topic revision: r1 - 2012-08-16 - RichardDekany
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback