next up previous
Next: Camera Alignment Up: No Title Previous: Camera

Setting up the focus of the spectrograph

This section deals with the initial setup up of the camera, not to routine focussing after the initial setup, and involves checking the tilt of the CCD inside the camera. This entails adjusting three micrometers (A,B and C) at the back of the camera. Refer to ?? for details of routine focussing during use.

The Tek CCD has four readout amplifiers - one on each corner. The presentation of the image is determined by the amplifier being used for readout. In reflection mode, the current mapping from slit and grating to the chip is:-

picture25

At least three fibre bundles, one at each end of the slit and one near the centre, need to be illuminated with a comparison source - typically an argon lamp. Install the R1200R ISIS grating into the grating cell. It is okay to leave the grating cell at the change position (-25000 mdeg) to check the focus.

Set the slit focus to 0 (this should be the autocollimation position), slit translation to -3500 (this gets all of the start fibre on the chip), put in the right hartmann shutter, and take an exposure (1 second, say). Measure the y-centres of well exposed arc lines at the top, middle and bottom of the left hand, centre and right hand edge of the CCD - a total of nine points.

Open the right Hartmann shutter, and close the left one. Repeat the exposure and measure the y-centres of the same arc lines, and determine the difference (right-closed) - (left-closed).

picture34

Note that the CCD surface is not flat - it is cut as a section from a spherical surface, so it is not possible to position the CCD to give (near) zero Hartmann shifts over the complete imaging area. The compromise was to adjust the micrometers to give a positive hartmann shift at the centre of the CCD, and a negative (balanced) shift around the edge.

There is no serious degradation in performance if the shifts are within 0.3 of these value, either in tilt or piston, as the resolution of the spectrograph is limited by the fibre size (at least for the 2.5 arcsecond fibres). Having said this, it is wise to keep the shift in the centre close to +0.5. This should normally be accommodated using the slitfocus.

If it should prove necessary to adjust the tilt of the CCD, then note the following:-

Magnitude of the adjustments:-

Note that normally after an adjustment in the dispersion direction, a small piston will be necessary, as the pivot is constrained by the copper cold finger, displacing the pivot point away from the chip centre.

Rotation is achieved by rotating the complete CCD ring, using a (small) rotation knob just below the pumping port on the camera. There is no accurate calibration!

To check rotation, use a continuum source instead of the calibration lamp. Find the line centre at the top and the bottom for a fibre at the centre of the imaging area of the CCD (i.e. column  512 + underscan region).

Very approximately, half a turn anticlockwise on the rotation knob changes the difference (top - bottom) by +1 pixel. This is very approximate - the link between the rotation knob and the CCD ring is a stranded cable, so is not at all consistent.


next up previous
Next: Camera Alignment Up: No Title Previous: Camera