Next: Seeing at the 4-m
Previous: CCD MOSAIC Imager Status
Table of Contents - Search this issue - NOAO Newsletter Home Page

NOAO Newsletter - Kitt Peak National Observatory - June 1997 - Number 50


New 4-m Corrector/ADC Commissioned and in Use with PFCCD

We expect that the standard PFCCD system will remain in demand at the 4-m until blue-sensitive CCDs with good cosmetic qualities become available for the new Mosaic camera. However, PFCCD users will immediately begin to benefit directly from the Mosaic project thanks to the commissioning of the new prime-focus 4-m corrector and atmospheric dispersion compensator (ADC). Details of the new corrector/ADC can be found in the June 1996 NOAO Newsletter; here we briefly summarize some of the changes that PFCCD users can expect.

The PFCCD corrector is a four-element design that replaces the small doublet corrector that has been in use with the PFCCD. Users of the old system have long been aware of the degradation of image quality as one approaches the edges of the chip; the new corrector produces excellent images over the entire CCD. During commissioning tests, we have obtained sub-arcsecond images over the entire field, and the point-spread function variations seem smaller than with the doublet.

In addition, the new corrector has a built-in ADC which further improves the image quality, particularly at U and B, for observations obtained away from the zenith. As the accompanying figure shows, without ADC there is a natural spread in a U or B image of about 0.5" at an airmass of 1.5, due simply to differential refraction over the widths of the bandpasses. The ADCs will remove this extra spread. The ADCs are run by the telescope operator, who must select the approximate bandpass ("U" through "I") to get the right amount of correction; the two prisms are then put in "track mode" and their positions corrected once a minute. If an observer chooses not to use them for some reason, they can be locked in a null position. Observers can also expect improved scattered light performance because of careful consideration given to coatings, metal blackenings, and spacing of elements.

image
Caption:Plot of the image spread in altitude resulting from differential atmospheric refraction at the blue and red ends of the filter bandpass for the U, B, V, R, and I filters, as a function of the airmass.

We found that twilight sky exposures obtained in bright twilight flattened dark-sky exposures to a fraction of a percent both in broadband and narrowband exposures, with dome flats adequate only at the 1-2% level. This is similar to what was the case with the old corrector. As described in the March 1997 Newsletter, there is a faint ghost of the telescope pupil visible in exposures. This reflection is typical of such correctors, but we have managed to greatly reduce its effect by using a special coating (Sol-gel) on the critical optical surface. In practice, the level of the ghost image in raw frames is < 3% for most interference filters and at "U"; for other broad-band filters the intensity is far below this. These ghost images flat-field out well (< 0.1-0.3%) using twilight flats, although we expect that observers may be disconcerted when examining raw frames with strong windowing around the sky level. Exposures through a narrow interference filter at 9500Å shows a considerably larger effect (10%), although we expect this to also flat-field out as well. For applications requiring better than 0.1-0.3% flat-fielding, we recommend (as always) that flat fields be constructed from dark sky exposures.

The f-ratio with the new corrector is somewhat slower (f/3.1) than with the old corrector, providing improved sampling. The image scale is now 0.420"/pixel, and the field covered is 14.3' × 14.3'. Although this is nominally smaller than the 16' × 16' obtained with the old corrector, the useful field is probably about the same or somewhat larger, as superb image quality can be obtained over the entire field.

We were also able to obtain an improved measure for the count rates obtained with the PFCCD system. For a Johnson-Kron-Cousins U=B=V=R=I = 20th magnitude star, we expect the following count rates in e-/s:

                     U     B      V     R      I
                    45    270    330   340    175

The "Direct Imaging Manual for Kitt Peak" (available either by anonymous ftp or via the documentation page on the Kitt Peak home page) has been updated to reflect all of these changes.

We are grateful to the many individuals who made the new corrector and ADC possible, including David Vaughnn, Ming Liang, Dick Sumner, Rich Reed, Gary Muller, David Mills, Shelby Gott, Bob Marshall, Dave Dryden and Bill Schoening.

Phil Massey, Taft Armandroff, George Jacoby


Next: Seeing at the 4-m
Previous: CCD MOSAIC Imager Status
Table of Contents - Search this issue - NOAO Newsletter Home Page

NOAO is operated by the Association of Universities for Research in Astronomy (AURA), Inc. under cooperative agreement with the National Science Foundation