History: IfA's First Large CCD Mosaic Camera
by Louise Good
A team of only three people built the UH8K camera for less than $300,000 in only seven months At the time, it was the largest CCD camera in the world. Photo by Richard Wainscoat.
The previous issue of this newsletter (no. 40) stated that a camera designed and built at the IfA played a key role in the research that led to the 2011 Nobel Prize in physics. This camera, called the UH 8K CCD Mosaic Camera (UH8K for short) because it had 8192 x 8192 pixels (64 megapixels), contained the first of the new generation of large-format CCD mosaics.
The UH8K was designed and built to be used on the 3.6-meter Canada-France-Hawaii Telescope (CFHT) and occasionally at the UH 2.2-meter telescope, both on Mauna Kea. The camera was built at the IfA Detector Laboratory in Mānoa and saw first light at the CFHT in March 1995 and on the UH 2.2-meter telescope in August 1995.
The UH8K was built entirely by a team of only three people. Gerard Luppino, then a member of the IfA faculty, but now the head of GL Scientific, a Honolulu company that provides custom design and manufacturing services for scientific instruments, led the project as the principal investigator. He was also the designer, machinist, mechanical technician, and electronics technician. Postdoctoral fellow Mark Metzger wrote all the software. He left IfA to become a professor at Caltech and is now a senior research scientist at a hedge fund management company in New York. Visiting scientist Satoshi Miyazaki came from the National Astronomical Observatory of Japan. He subsequently built Suprime-Cam, the large camera on the T telescope, and is now leading the effort to build an even bigger one for Subaru, the Hyper-Suprime-Cam.
The UH8K camera was built on a shoestring budget of less than $300,000. The funding for the hardware came from the Director's Discretionary Fund and Luppino's grant from the National Science Foundation (NSF), with the money for Metzger's salary coming from an NSF grant, "Wide Field Imaging of Solar System Objects," to IfA astronomer Karen Meech.
They built the UH8K very fast. Luppino explained, "Most instruments take years to develop and produce. We knew we had an opportunity to beat everyone else developing similar large cameras if we could get onto the sky quickly, so we departed from the usual several-year schedule. From the time we had CCD wafers with confirmed, working devices to first light was seven months! In this period, we designed the CCD packages, diced the wafers ourselves (at Loral Aerospace's Newport Beach foundry), did all the device packaging and wire bonding (in my lab at IfA), designed the dewar (a thermoslike container used to keep the camera cold) and built it in the IfA shop, designed and built a large shutter and filter slide, procured the controllers, wired up the system, and wrote all the software to run the CCDs and acquire the images. The CCDs were poor compared with the devices we have now. But they were big, and there were eight of them. For many programs, area trumped perfect detectors."
Luppino's reason for building the UH8K was to use it to study weak lensing in clusters of galaxies, work he was doing with Nick Kaiser (now heading the Pan-STARRS project at IfA), Lev Kofman (left IfA for the University of Toronto in 1998; died in 2009), and Isabella Gioia (still a frequent visitor to IfA). "For this work, the UH8K was ideal. It was also well suited for the supernovae searches that led to the Nobel prize," said Luppino.
Meech's group used the camera for recovering comets. "The large area was important for comets whose orbital uncertainties were larger than the traditional CCD size. We also got some key observations of the really bright Comet Hale-Bopp, which had a huge tail," she said.
The CFH12K camera with Gerard Luppino.
By 2000, the UH8K was replaced on the CFHT by the larger CFH12K (100 megapixels), which used better detectors. Later, the detectors on the UH8K were also upgraded, and it was fitted with improved optics, developed by IfA's Klaus Hodapp, so it could be used on the UH 2.2-meter telescope. It was employed on that telescope until 2005, when it was accidentally damaged. Since the repair would be costly and another camera is available for use on the 2.2-meter, it is likely that the UH8K will now be retired.