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 11:14:52

Maintained by

Printable version

Filter Wheel Position vs Ramp Wavelength


TITLE : Filter Wheel Position vs Ramp Wavelength

CATEGORY : Calibration


The main goal is to map the position offsets of filter wheel 2 (FW2) as a function of wavelength for each ramp segment. The optimal wheel position for five separate wavelengths spanning the entire wavelength range of each segment will be determined, allowing a comparison and refinement of the original component-level ramp calibration measured at GSFC. The FW2 offset for any wavelength can then be determined by interpolation through the five measured wavelengths.


Execute the entire procedure once.


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


The nominal offsets for the middle ramp segment needs to be determined with flat-field test exposures. In the document IN0077-112 : "Operations and Data Management Plan for the ACS", Nov. 1998, p. 54, the specified offsets are -77 steps for WFC1-MRAMP and +77 steps for WFC2-MRAMP for the central wavelength of the ramp. But our experience at BATC and GSFC indicates that these offsets are too large, resulting in vignetting of the middle segment. Instead, we find that offsets of +/- 62 steps are more appropriate.

The nominal location of the monochromatic patches for each aperture (HRC, WFC2-ORAMP, WFC1-MRAMP, WFC2-MRAMP, WFC1-IRAMP) must also be determined prior to executing the procedure.


RAS/HOMS is equipped with a Mylar flat field diffuser at the pupil mask location so that the monochromatic illumination from the QTH lamp correctly simulates the OTA beam angles over the entire field. The QTH lamp is coupled to the monochromator and the slit is adjusted to provide a 20 Ang bandpass at each tuned wavelength. A wider slit may be used at short wavelengths to increase the counts.


The IDL procedure (with RampCoeff.dat) can be used for a first estimate of FW2 offsets.

COMMAND MODE : Real time.


Monochromatic flat fields will be acquired for all the medium- and narrow-band ramp segments at five separate wavelengths spanning the range of each segment. All the ramp filters are in FW2. For the WFC, the resultant monochromatic patches are 90"x60" in size and are imaged on the WFC2/Amp D quadrant for the outer segment, the WFC1/Amp B or WFC2/Amp C quadrants for the middle segment (with the appropriate offset), and on the WFC1/Amp A quadrant for the inner segment. For the HRC, the monochromatic patch is 26"x20" and is imaged at the centre of the HRC FOV. Only the middle ramp segments can be imaged on the HRC. Details on the ramp filter geometry can be found in IN0077-112 : "Operations and Data Management Plan for the ACS", Nov. 1998, pp. 49-52. It is only necessary to read the WFC as quadrants (WFC1-IRAMP, WFC1-MRAMP, WFC2-MRAMP, WFC2-ORAMP) but since the full WFC FOV will be acquired on-orbit with the ramps, we also perform a full four-amp read-out for consistency.

The wavelengths for all the ramp segments are listed in Table 1 below (Z. Tsvetanov). Essentially, the central wavelength of each segment is bracketed by four regularly spaced wavelengths over the wavelength range of the segment. SMS spreadsheets are also given. The absolute FWoffset2 were calculated with the +/- 62 steps in (see below).

The offset of FW2, FW_off(lambda) vs X(lambda), which brings the monochromatic patch in the correct aperture is given on page 54 of IN0077-112 and the distance vs wavelength relation for each segment, X(lambda) vs Lambda (X in mm), is on page 53. The ramp filters were analyzed at GSFC and the transformation coefficients a_0, a_1, and a_2 were fitted by J. McCann (see the Filter Data Packages). These relations were implemented by Z. Tsvetanov in the IDL procedure and the coefficients in RampCoeff.dat.

The method is relatively straightforward :

Once the wavelengths and revised filter wheel offsets are known, new a_0, a_1, and a_2 coefficients can be determined by fitting the FW_off_new vs X(lambda) equation to the data points. The correct FW2 offsets can then be determined for any wavelength.




The WFC apertures will be read as subarrays/quadrants and the HRC as full frame. The minimum number of exposures (1 exposure/wavelength/aperture for a given segment) for successful completion of this calibration item is then 100 WFC quadrants and 25 HRC frames.


Assuming 5 minutes of overhead to tune each wavelength manually and 2 min for read-out and dump of each WFC quadrant and HRC frame, the minimum execution time is ~10.4 hours. Overhead for test exposures, refinement of the offset, and contingencies will add up significantly and could easily double the total time. A total of ~20 hours is probably realistic for this calibration item.


The new FW2 offsets will be fit to the wavelengths to update the coefficients a_0, a_1, and a_2 with any valid numerical package (IRAF, IDL, etc...).