At the end of August, 1997, new focal reducing optics have been installed and tested at MRO. The old focal reducing optics were not particulary appropriate for the current camera. They were not designed for a field of view (FOV) as large as the current camera affords. As such, there was a considerable amount of vignetting in the old optics. In addition to the vignetting, it has been known for some time, that the point spread function (PSF) in the old optics was variable as a function of position in the field of view. Stellar images at the edge of the field using the old optics had an elliptical shape with a ratio of major to minor axes of approximately 1.25. This was a serious impediment to doing crowded field photometry over the entire image.
Ed Mannery was kind enough to donate some of his personal time
in order to design the new MRO focal reducing optics. The physical mounts for
the lenses were built here at UW in the physics machine shop.
Figure 1 shows Ed's design of
the new focal reducing optics which consists of 5 optical elements. These
elements are mounted within the filter slide mechanism. The filters fit
between optical elements 1 and 2. The optics were built by Coastal Optical
Systems, Inc. All five optical elements
were coated by QSP Optical Technology, Inc. with very broad band
anti-reflection coatings. These coatings insure that single interface
reflection losses are less than 2% between 310 and 1000 nm. The average
relection losses are less than 1.5% in this interval.
At the end of August, 1997, these new focal reducing optics
were installed and tested at MRO.
Figure 2 shows an image of the open cluster NGC129 taken through
the old focal reducing optics. The circular truncation of the image field of
view was caused by the vignetting of the optics. The two vertical streaks are
caused by bad columns on the CCD. The variable PSF in the
image is present, but not obvious owing to the stretch used in the image
display. The wings
of the PSF are more symmetric than its core, but the core usually dominates
the total energy enclosed in each stellar profile.
Figure 3 shows an image of the
same cluster taken through the new focal reducing optics. The vignetting is
completely gone. The stellar images are circular throughout the entire FOV.
This image was taken with the pixels binned 2 x 2 on the chip.
Measurements have been made on the new focal image scale. For
1x1 binning we now have 0.605 arcseconds/pixel. This is equivalent to an
f-number of 6.71 at the detector. Our full field of view is now
10.32 x 10.32 arcminutes. The MRO manual will soon be updated to reflect
these facts.
The optics now allow good crowded field photometry to be done across the entire
image of the detector. Bivariate Gaussian fits to stellar profiles across
the FOV show that there is no detectable variation in the image
PSF as a function of position in the FOV. The following table shows the
results of analyzing the stellar profile of 40 stars across the FOV of a
random star field in the constellation Cassiopeia. The measured Full Width at
Half Max (FWHM) of stellar profiles is given as a function of their distance
from the center of the FOV.
Measurements of the relative flux of a cluster of
stars made at different
positions in the FOV have been taken in order to estimate the vignetting
that exists with the new optics. These measurements have
shown that the vignetting across the entire field of view is
is less than 2%. These estimates are somewhat limited by
stability in the photometric conditions at the time of the exposures.
One measurement indicates that the flux of stars placed near the right hand
edge of the detector (ie. within about 60 arcseconds of the eastern edge of
the detector) may be as much as 4% below the flux of these same stars
measured near the center of the FOV. However, similar measurements at
the other three quadrants of the FOV show 1.7% variations in
throughput with the higher flux appearing near the edge of the FOV.
Ed Mannery has made a ray-trace analysis of the throughputs
expected of the optics based on the absoprtions expected from the glasses
and the reflections from the coatings at the glass interfaces. This analysis
shows that the throughput near the edge of the detector is expected to be
1.8% HIGHER than the throughput near the center of the FOV. This is perfectly
consistent with the stellar measurements reported above for 3 of the 4
quadrants measured. Our conclusion is that it appears that the new optics
are working exactly as expected as far as vignetting is concerned.
Users should be aware of the existence of a 1.8% systematic throughput
variation as a function of the distance from the center of the FOV.
This work was enabled in large part by contributions to the
observatory by amateur astronomers from the Seattle area. We wish to express
our thanks to their generous efforts to help make MRO an active scientific
facility. In particular, we wish to thank Jim Naiden, George Best, Larry Shea,
Gerald West, James Pryal, Karl Heinz Bohm, Keith Allred, and John Mitchell.
Without their support, this improvment to the telescope would not
have been possible.
Radius from center (in arcmin) FWHM (in pixels)
Number of stars in sample
R < 4.67 3.60 +- 0.04 13
4.67< R < 7.00 3.58 +- 0.07 22
R > 7.0 3.60 +- 0.26 5
A Word of Thanks
