spectra

The Rigel telescope also features a spectrometer in addition to its main camera. The spectrometer for the Rigel Telescope is an Ocean Optics USB device that is connected to the telescope through a 300 mm fiber. The telescope features a 45 degree mirror in front of the camera that "cuts" off a bit of light and channels it into the fiber which is then fed through a 25 mm slit and into the spectrometer. The problem with this picture is a geometric one. Picture this: a beam of light 300 mm in diameter is fed through a 25 mm slit. Much of the original signal is lost in this process as the only light that actually enters the spectrometer is that which goes through the slit. To solve this problem a collimating device will be constructed this summer with two lenses that will concentrate the light from a 300 mm beam down to a 50 mm beam. This will mean that a much smaller ratio of the signal will be excluded from entering the slit and being processed, but as of right now all data that has been taken (including the data below) is suffering from a dramatic loss of signal due to the geometry of this problem.
Once this collimating device is implemented, there is still a matter of making sure that we can actually land a star perfectly on the tiny 300 mm fiber. The procedure for accomplishing this task was to determine to an extremely high degree of accuracy the offset from a position on the camera to the position of the fiber off of the 45 degree mirror which is to the side of the camera. This offset was determined to be 20 pixels South of the center of the image in the y direction and 729 pixels West of the center of the image in the x direction. The procedure for taking the spectra of a star will be to first image the star and determine it's precise position on the image and then make the telescope move the tiny offset that would be required to get the star's light to reflect off of the 45 degree mirror and directly into the 300 mm fiber.
Finally once the above problems were tackled there was still an issue with the collected data itself. Generally there is a lot of thermal extraneous noise included with the spectra. The most common way to get rid of this noise is to first take a spectra of a dark area (so that you will only get background noise) and then subtract the background thermal noise from the collected data to get the actual spectrum of the star. However, a slightly less straight forward approach, and certainly one that is more instructive to students of astronomy is to use a Fourier transformation to determine at which frequencies the noise is most dominant. Once this is done it is possible to set the signals corresponding to those frequencies equal to zero and replot the spectrum. The result is a nice clean spectrum from which it is possible to discern a blackbody curve and several absorption lines, namely the Balmer lines.