KPNO continued its aggressive efforts to improve delivered image quality at both the Mayall and WIYN telescopes. In the area of instrumentation, the CCD Mosaic imager enjoyed heavy usage with scientific grade CCDs, significant improvements were made to the Phoenix high-resolution IR spectrograph, and the 4-color near-IR imager SQIID approached readiness for re-commissioning.
Image Quality Improvements
A key aspect of improving delivered image quality is controlling the thermal environment of the telescope and its components. Major improvements to the Mayall 4-meter began a decade ago with the installation of dome fans and the cooling of the horseshoe bearing oil. More recent changes have included an initial cooling system for the very thick primary mirror and the installation of the 22 dome vents, along with careful insulation of warm parts of the building.
The progress to date is reflected in the image quality statistics for the last three and a half years. While the median delivered image size remains at 1.1 arcseconds, the telescope delivers subarcsecond images nearly 40% of the time. The most encouraging sign of progress is that we achieve image quality of 0.85" or better a full 25% of the time.
Efforts over the last year have focused on the thermal and mechanical control of the 4-m primary mirror. Improvements of more than 0.1" in the median are anticipated from correcting low-order bending modes and in tighter thermal management of the surface of the primary mirror. The 4-meter Active Primary Support system (4MAPS) was installed during summer shutdown 1998. Activity during FY 1999 centered on characterizing the performance of the system and optimizing the speed and operability of the control system. With some further minor mechanical modifications in July, the system is now in routine use in its ``emulation mode," in which it mimics the performance of the previous passive air system. Engineering nights this fall will enable the production of look-up table based force maps, which will then add the low-order corrections to the basic support pattern. The goal is to be running with the look-up tables at all three foci by the start of the next calendar year. In addition, the tilt mechanism on the f/8 secondary was fitted with precision encoders. These will allow implementation of a tilt correction with telescope position to compensate for the minor decollimation that induces measurable coma. The coming year will see the installation of a wavefront camera at the RC focus addressable by the pickoff mirror. The camera will permit the determination of zero-point offsets for the lookup tables on a nightly basis while a spectrograph or other major instrument is installed.
Major modifications were made to the primary mirror area. The coudé optics train was replaced with air ductwork, in a fashion similar to the design employed by CTIO. A large exhaust fan in the former area of the #5 mirror now extracts air from over the primary to aid rapid front surface equilibration. In addition, the old iris mirror cover was replaced by one with gaps between the elements, enhancing air flow into the mirror well at most telescope positions. The project will be completed in the coming fiscal year with the installation of a chiller/dryer unit and air plenum at the primary. At that point, we will have the capacity for enhanced front surface cooling and nearly uniform air extraction.
The WIYN telescope also received considerable attention in the area of delivered image quality. Although the median image quality reported from the 10-second exposures for seeing tests is 0.8" or less, longer guided exposures tended not to reproduce that level of quality. One source of image blur was found to be a suite of resonances in the telescope top end, centered at about 23 Hz. Those resonances had to be eliminated, both for the improvement of telescope delivered image quality and to allow the possibility of rapid image motion compensation. KPNO engineers devised a damping scheme that has proven to be effective, with the result that guided exposures more nearly represent the distribution of seeing from the seeing test exposures.
Attention to operational detail has also paid off in better images. Wavefront images were showing that some low-order deformations of the WIYN primary were not being properly corrected. After painstaking work, the problem was traced to small temperature non-uniformities induced by variances in the calibration of thermal sensors in the mirror thermal control system. Careful cross-calibration has brought those non-uniformities to well within 0.1°C, with the expected improvement in primary mirror performance.
A major improvement in image quality is the design goal of the WIYN Tip/Tilt imager. The project passed its Preliminary Design Review in March 1999, with a strong endorsement from the external review committee and the WIYN Science Advisory Committee. The rapid steering mirror will address a 4 arcminute field of view, with correction bandwidths of up to 25 Hz. The system will have a detector scale of 0.12"/pixel and will be optimized for V through I bands. It will be packaged within the existing instrument adaptor, allowing rapid switching between the conventional imager and the tip/tilt module. The expectation is that images of 0.33 arcsecond FWHM will be achieved, with up to a 50% increase in Strehl ratio in some bands. The instrument is now in detailed design phase, with an expected deployment in 2001.
The efforts of Chuck Claver, Charles Corson, Phil Massey, and the strong technical support are paying off, with every major telescope delivering subarcsecond images routinely. The 4-meter delivered 0.5" images at the RC focus on recommissioning at the end of summer shutdown; the 0.9-m image diameters were within 1.06 pixels. KPNO telescopes are being tuned to deliver image quality of the higher standard now being demanded.
The CCD Mosaic imager is heavily scheduled for science with its thinned, science-grade devices. Mosaic was used for 132 nights in Semester 1999A, and is scheduled for 121 nights in 1999B. The power of this instrument illustrates the value of the wide-field capabilities of both the Mayall and CTIO Blanco telescopes.
The Phoenix spectrograph was taken out of service to correct a number of problems in the initial configuration. All the instrumental problems have now been solved. The most apparent was a strong astigmatism in the collimator. The condition was corrected by installing three mounting pads made from a piece of aluminum foil. The width of the spectrum on the detector is now determined by the seeing rather than by unwanted optical distortion. Spectral resolution of 70,000 can be achieved with a 2-pixel slit. The spectrograph demonstrated 10% throughput in the H-band, which is very high for an echelle design, including the slit losses in a night of 0.7" seeing. The speed of mechanisms was also greatly increased, to improve efficiency in target acquisition and configuration changes. Phoenix will be offered on the 4-meter for Semesters 1999B and 2000A. It will then be modified for shared use on Gemini South and the SOAR telescope.
Work has proceeded this year on upgrading SQIID (Simultaneous Quad Infrared Imaging Device). Through fixed dichroic beam splitters and filters, the instrument records the image of a single field simultaneously in four bands, J, H, K, and L'. NOAO's co-sponsorship of the ALADDIN development project for 1024 x 1024 InSb arrays has led to a number of partially working devices with excellent single quadrants. Four such devices will be deployed in SQIID. The field of view will be almost 6 arcminutes on a side on the 2.1-m and will give finer sampling over 3 arcminutes on the 4-meter. To handle the equivalent of a 4-quadrant device, the knowledge gained from developing controllers for Gemini will be applied to a KPNO-based data system. The instrument is scheduled for testing this coming November, and may be offered to users on a shared-risk basis during Semester 2000A.