Next: Alignment of the Spectrum on the Chip
Previous: Focusing the Spectrograph and Telescope
When higher dispersion gratings are used with the Gold Camera, users should be aware of a potentially troublesome "feature" of the system which becomes evident during the focusing procedure. The problem exhibits itself much as astigmatism does at a telescope focal plane. There are actually two positions of best collimator focus depending upon whether you are measuring in the spectral direction or in the spatial (along the slit) direction. When the spectrograph collimator is focused one must be sure to find the best compromise focus (spectral and spatial) or smallest "circle of confusion".
While not yet fully understood, the problem is connected with the extent of ellipticity of the collimated beam striking the grating. At high tilt angles, such as that required by high dispersion gratings, the beam produced by the collimator is more highly elliptical than that produced by gratings at smaller tilts. The spectrograph camera does not handle highly elliptical beams well and produces astigmatic-like images. The gratings most affected are those used at tilts of less than 14.00 units (see Table 4). Those particularly affected are numbers 36, 47, 56(orders 2 and 3) and number 35(2nd order). Note that tilt readout values run inversely with the angle, i.e. higher tilt angles occur at lower numbers.
To work around this problem, one needs to focus the collimator in the spectral direction and in the spatial direction. For spatial focusing, one uses the quartz lamp and a special set of holes on the decker plate. The FWHM of the height of the quartz spectrum is minimized using the collimator. The spectral focus is done in the traditional manner using spectrograph slit and the HeNeAr calibration lines. The compromise focus is reached by considering the collimator position for both methods.
For the highest dispersion configuration there is also a set of special oval masks that can be installed over the grating to apadize the beam into a more circular form. The spectral focusing is done with the mask in place, but most observers opt to remove it for the real observations.