200
14 Spectrography
the CCD will be on the order of 80% efficient. Hence, in a typical eight-element
configuration, total throughput will be on the order of 0.98
8
× 0.75 × 0.8 = 0.51.
So we lose about half our light within the optics. It is important, therefore, when
undertaking photometry that the exposure time be long enough to get a good signalto-noise ratio and that the slit be opened to the optimal aperture.
As with imaging, there is always the option of binning to improve the signalto-noise ratio. As individual columns of pixels all represent the same wavelength,
assuming that the image is not rotated, we can, and certainly should, vertically bin our
spectra to a single row. If we have higher spectral resolution than we need, we also
have the option to bin linearly after binning the columns. Binning is good practice
if you are not particularly interested in the line width. It will also enhance the bright
lines when the spectra are plotted and reduce confusion within the plot.
Just as it is possible to undertake field photometry, it is possible to undertake
field spectrography, but not with a standard long-slit spectrograph as shown above.
Multislit photometry is possible using a modified long-slit spectrograph. However,
there are problems with this approach, and it can be inefficient.
Fibre-fed multi-object spectrographs (MOS) use a mounting plate with holes
drilled where the image of the stars would fall. Integral field units (IFU) divide the
field into a grid with a small lens covering each cell. The lenses focus light into fibreoptic cables glued to the back of the lens. The fibres feed into the spectrograph and
form multiple spectra. IFUs are powerful tools for determining large-scale movement
such as the rotation of galaxies.
It is possible to undertake slitless spectrography using an inline transmission
grating such as a Paton Hawksley Star Analyser. In-line transmission gratings can
be fitted to the nose of a camera or webcam or inserted into a filter wheel. They
produce low-resolution spectra of all the objects in the field, which may result in
confused and overlapping spectra. However, they are inexpensive, easy to use, and
are becoming increasingly popular with amateur astronomers.
14.3 Taking and Reducing Spectra
To some degree, the taking and reducing of spectra is very similar to that of normal
imaging. Although the software used for the reduction is different, in general, the
camera control software is the same.
As with imaging, you will need to take dark, flat, and bias calibration frames and
apply them either with the camera control software or a spectral reduction package
such as RSpec, VSpec, IRIS, or IRAF. Remember that the spectrograph is a CCD,
and it needs to be cooled and managed in the same way as a standard imaging CCD.
For each spectrum, both reference and calibration spectra are required; this is in
addition to bias, dark, and flat frames.
14 Spectrography
the CCD will be on the order of 80% efficient. Hence, in a typical eight-element
configuration, total throughput will be on the order of 0.98
8
× 0.75 × 0.8 = 0.51.
So we lose about half our light within the optics. It is important, therefore, when
undertaking photometry that the exposure time be long enough to get a good signalto-noise ratio and that the slit be opened to the optimal aperture.
As with imaging, there is always the option of binning to improve the signalto-noise ratio. As individual columns of pixels all represent the same wavelength,
assuming that the image is not rotated, we can, and certainly should, vertically bin our
spectra to a single row. If we have higher spectral resolution than we need, we also
have the option to bin linearly after binning the columns. Binning is good practice
if you are not particularly interested in the line width. It will also enhance the bright
lines when the spectra are plotted and reduce confusion within the plot.
Just as it is possible to undertake field photometry, it is possible to undertake
field spectrography, but not with a standard long-slit spectrograph as shown above.
Multislit photometry is possible using a modified long-slit spectrograph. However,
there are problems with this approach, and it can be inefficient.
Fibre-fed multi-object spectrographs (MOS) use a mounting plate with holes
drilled where the image of the stars would fall. Integral field units (IFU) divide the
field into a grid with a small lens covering each cell. The lenses focus light into fibreoptic cables glued to the back of the lens. The fibres feed into the spectrograph and
form multiple spectra. IFUs are powerful tools for determining large-scale movement
such as the rotation of galaxies.
It is possible to undertake slitless spectrography using an inline transmission
grating such as a Paton Hawksley Star Analyser. In-line transmission gratings can
be fitted to the nose of a camera or webcam or inserted into a filter wheel. They
produce low-resolution spectra of all the objects in the field, which may result in
confused and overlapping spectra. However, they are inexpensive, easy to use, and
are becoming increasingly popular with amateur astronomers.
14.3 Taking and Reducing Spectra
To some degree, the taking and reducing of spectra is very similar to that of normal
imaging. Although the software used for the reduction is different, in general, the
camera control software is the same.
As with imaging, you will need to take dark, flat, and bias calibration frames and
apply them either with the camera control software or a spectral reduction package
such as RSpec, VSpec, IRIS, or IRAF. Remember that the spectrograph is a CCD,
and it needs to be cooled and managed in the same way as a standard imaging CCD.
For each spectrum, both reference and calibration spectra are required; this is in
addition to bias, dark, and flat frames.