The Advanced Camera for Surveys
  instrument > calibration > plans > rashoms feb2001 >
+ archive
+ Calibration
+ Activities
     · 2001
     · 2000
     · 1999
     · 1998
· Plans
+ Results
     · by item
     · by software
     · TV1
· Campaigns
· Facilities
· Database search
· Log files
· SMS list
· Software
+ detectors
+ filters
+ overview
+ publications
+ photo gallery
Team lava

Last updated
20 December 2001 08:43:08

Maintained by

Printable version

Coronagraph Performance


TITLE : Coronagraph Performance

CATEGORY : Calibration.


The main goal is to acquire a comprehensive dataset of HRC coronagraphic RAS/HOMS PSFs to assess the PSF suppression capabilities of the two spots (1.8 and 3.0 arcseconds in diameter) and of the Fastie finger. The PSF will be centered and moved under the spots and finger to characterize its off-center characteristics. We will also verify centration of the pupil mask using the M1 tip/tilt to steer the pupil image and examine the residual spider diffraction.


Execute this procedure once.


Flight build detectors HRC#1 (amp C) and WFC#4 (amps ABCD).


The optimal focus and tip-tilt position of the M1 corrector mirror should first be determined using the HRC focus and tip/tilt SMSs (see Item 1 : Image Quality vs Field Position).

A rudimentary version of this procedure was performed with the non-flight build detectors HRC#1 and WFC#3 in June 2000 with RAS/HOMS (see HRC Coronagraph). The agreement with the models was found to be generally good. For this test, the optimal focus and M1 mirror position used were : focus = 2291, inner cylinder = 56931, outer cylinder = 10308.


RAS/HOMS is in its normal configuration with the RAS source plate and pupil mask in position. The illumination is provided by the central fiber (A1) at HeNe 6328 Ang. A suitable complement of ND filters is available and ready to be positioned in front of the laser. The coronagraphic arm is in place. An operator is suited in the clean room to adjust the micrometer stage and move the ND filters. All images are acquired through the F625W filter.


IDL to view the images and J. Krist's software for data analysis and model comparison.

COMMAND MODE : Real time.


We will acquire a significantly more comprehensive and deeper dataset than was obtained in the June 2000 campaign. As a reference, we show in Fig. 1 an HRC flat field acquired in June 2000 at BATC with HRC Build 1 and with the illumination provided by RAS/HOMS (6328 Ang) through the F606W filter. The central 1.8 arcsec-diameter spot, the bottom 3.0 arcsec-diameter spot, and the 0.8 arcsecond-wide Fastie finger are clearly demarcated. No attempt had been made to align the end of the finger with the center of the bottom spot; it was simply assembled in the front of the detector, knowing that the detector would be replaced in the near future.

Cuts across the spots and the finger are shown in Figs. 2, 3, and 4. In coronagraphic mode, the HRC plate scale is 0.027 arcsec/pixel. The "true" width of the central spot is ~110 pixels or 3.0 arcseconds and the width of the bottom spot is ~180 pixels or 4.9 arcseconds. Similarly, the width of the Fastie finger is ~55 pixels or 1".5. The centre of the central spot is at (595,469) and the bottom spot is at (694,141). The Fastie finger makes an angle of 72.5 degrees with the bottom horizontal edge of the chip.

The program is divided into two parts. In the first part, the PSF-suppression capabilities of the occulting spots and Fastie finger will be studied for centered and offset positions of the PSFs. In the second part, the M1 corrector mirror will be tip-tilted symmetrically about its nominal position to evaluate the residual strength of the vertical and horizontal spiders.

Part I : PSF Suppression : Spots + Fastie finger

  1. The focus and tip-tilt of the M1 mirror are adjusted to their nominal positions, either in real time or with the SMSs. No changes are made to these quantities throughout this part.
  2. Several trial images and PSFs are obtained to determine the location of the spots, the correct exposure times, and the attenuation of the ND filters. Setup images will also be acquired without the coronagraphic arm to normalize to non-CORON conditions. The exposure times must be such that pixel saturation is never reached in the PSF, whether the ND filters are in the light path or not. In the Ball June campaign, an exposure time of 0.1 sec (gain of 1) was used with the ND 3+2 combination for an appropriate laser intensity.
  3. Take a bias frame.
  4. Position the PSF under the central 1.8 arcsec spot. The sweeping pattern consists of a crossed 5-point scan in both X and Y and a PSF outside of the spot for a total of 10 points. The central point of the cross is at the centre of the spot. The separation between the points will be 2 pixels, although a greater separation may eventually be specified for the four end points of the cross for better spatial coverage under the spot. Two images will be acquired at each point to improve the S/N ratio and reject cosmic ray hits. This then results in a total of 20 full frame HRC images.
  5. Take a bias frame.
  6. Repeat step 4 for the 3.0 arcsecond spot.
  7. Take a bias frame.
  8. Repeat step 4 for the Fastie finger but in this case, increase the separation between the points along the major axis of the finger, say 5 pixels.
  9. Take a final bias frame.

Part II : Verification of Pupil Mask Centration

  1. The positions of the inner and outer cylinders of the M1 corrector mirror are adjusted successively in opposite directions while the residual vertical (inner) and horizontal (outer) spider diffractions are measured. The diffraction errors can be inferred from the symmetry of the residual diffraction intensities. The cylinders are typically moved in a sequence of 20 steps. For the BATC June 2000 campaign, 14 full HRC frames of 20 sec each with the ND 3 filter were required for this part.


If time allows, repeat the entire procedure at a different wavelength.


The setup prior to the actual test will require ~15 full HRC frames. Part I will generate at least 60 full HRC frames as well as a few bias frames. Part II will produce ~15 frames. We then estimate a total of ~100 full HRC frames.


Assuming 2 minutes per HRC frame, the total estimated time for Part I is at least 2 hours, but more realistically twice this amount if overhead is included, such as positioning the fiber. In the June 2000 campaign, Part II took 1.2 hours. Including setup images and overhead, the total estimated time for this procedure is then ~8 hours.


The analysis will be performed by J. Krist using his own software routines.