The Gumball upgrade project is a project to improve the capabilities and throughput of the Cassegrain calibration source known as Gumball. The Gumball was designed to be used with MOS/SIS, but with the coming of OASIS and the change from SIS to OSIS, the abilities required of the calibration source have changed.
The main goals of the project were to replace all of the Fresnel lenses to improve IR throughput, improve overall throughput, improve spectral coverage with the addition of a Fabry-Perot calibration source, improve the control electronics and software, and eliminate any dependence on additional optics in the instrument using Gumball. All of these goals, except the last have been, or will be, met with this project. The final goal, instrument independence, was found to be optically impractical during the design of the optics. This means that there will still be a calcor lens in MOS/SIS and lenses will be added to AOB and OASIS.
The Fresnel lenses, one in the Gumball and the other in MOS/SIS (the calcor lens) have been replaced with fused silica lenses as has the central mirror lens. The diffuser in the Gumball has been replaced with a highly reflective and spectrally uniform diffuser, and reflectors have been placed behind the pencil spectral lamps. These changes were tested with MOS during the November 1997 run. The throughput was seen to improve by about a factor of 10. This proved to be too much light for MOS, so the reflectors were removed and neutral density filters were introduced to bring the light levels down. This should greatly help OASIS, however, which had problems with extremely long calibration exposure times during the first two runs. A fused silica lens has been placed into AOB, but not tested, and a similar lens will be placed in OASIS for use with f/8 modes.
All of the optics for the Fabry-Perot source have now been received, and most of the mounting and electronic hardware is finished. The source will be assembled and fully tested when the final parts are received. This should take place by the end of December. The removal of the helium lamp to make room for the Fabry-Perot feeds has prompted the addition of two Hg(Ar) pencil lamps. These lamps will be placed on the same optical bench as the argon pencil lamps.
The electronics box has been completely redesigned and rebuilt using a ProLog computer for control. The new box senses whether a lamp is actually on or not and will give this information to the Pegasus. The new controller is also capable of allowing variable exposure times for different lamps in a multi-lamp calibration exposure. This is done by turning on all of the lamps then turning off the individual lamps during the exposure after the