Comment Re:Cooling question (Score 2, Informative) 239
How does the cameras sensors not melt and achieve good accuracy by staring into the sun ? Surely they must be cooled off, but how ?
Oddly enough, the CCD's generally are not cooled at all. The amount of light falling on the detector is actually not that great. Remember that each pixel corresponds to less than 0.1 x 0.1 arcsec, which means it covers about one-billionth of the solar surface (and hence the flux is one billion times less than integrated sunlight). Then you start taking very small slices in wavelength (0.01 nm or less, compared to the 100's of nanometers over the sun's emission peak in the visible wavelengths). Toss in a polarizer too (though they didn't use one in these observations), and next thing you know, you are running out of photons! That's why we need a big (by solar telescope standards) 4-meter telescope like the ATST ( http://atst.nso.edu).
The exposure times in observations like these are also very short, on the order of 20 milliseconds or less, so there is no time for the dark current to build up during a long exposure (this is why nighttime CCD's, with exposures of minutes or hours, are often cooled). For some applications, even simple video rate CCD's can be used (the problem often being the small number of pixels).
As you might surmise, even if the detector isn't getting hit with that many photons, a lot of extra light is going through the telescope. Getting rid of waste heat IS a problem, and, as is the case with the Swedish Tower, often the main body of the telescope (entrance aperture -> main mirror -> instrument feed) is kept in a vacuum to reduce currents from heated air in the optical path. However, the Swedish Tower appears to be at the limit for the size of the entrance window (must be of optical quality and with minimal stress) that can be used (the entrance window is the size of the main lens on the Yerkes telescope - the world's biggest refractor). That is why bigger telescopes like the GREGOR (1.5 m) and ATST (4 m) will be open, like nighttime telescopes, and will have to use different methods of thermal control. We can't go bigger than four meters now because of the limits of our thermal control capabilities.
Oddly enough, the CCD's generally are not cooled at all. The amount of light falling on the detector is actually not that great. Remember that each pixel corresponds to less than 0.1 x 0.1 arcsec, which means it covers about one-billionth of the solar surface (and hence the flux is one billion times less than integrated sunlight). Then you start taking very small slices in wavelength (0.01 nm or less, compared to the 100's of nanometers over the sun's emission peak in the visible wavelengths). Toss in a polarizer too (though they didn't use one in these observations), and next thing you know, you are running out of photons! That's why we need a big (by solar telescope standards) 4-meter telescope like the ATST ( http://atst.nso.edu).
The exposure times in observations like these are also very short, on the order of 20 milliseconds or less, so there is no time for the dark current to build up during a long exposure (this is why nighttime CCD's, with exposures of minutes or hours, are often cooled). For some applications, even simple video rate CCD's can be used (the problem often being the small number of pixels).
As you might surmise, even if the detector isn't getting hit with that many photons, a lot of extra light is going through the telescope. Getting rid of waste heat IS a problem, and, as is the case with the Swedish Tower, often the main body of the telescope (entrance aperture -> main mirror -> instrument feed) is kept in a vacuum to reduce currents from heated air in the optical path. However, the Swedish Tower appears to be at the limit for the size of the entrance window (must be of optical quality and with minimal stress) that can be used (the entrance window is the size of the main lens on the Yerkes telescope - the world's biggest refractor). That is why bigger telescopes like the GREGOR (1.5 m) and ATST (4 m) will be open, like nighttime telescopes, and will have to use different methods of thermal control. We can't go bigger than four meters now because of the limits of our thermal control capabilities.
