Previous: Rotation alignment
Previous Page: Rotation alignment
Next Page: Focus
A check to ensure that the CCD is parallel to the focal plane of the spectrograph follows. By using the routine TILT which is fairly self-explanatory inside the SETUP mode (see section 4.4) one can check that the focus is the same at six different windows. The positions of these windows should include bright arc-lines. In the red, with a high dispersion grating, CuNe lamp is the best one.
DEKKER 0 1LAMP ON SLIT 185 other settings as before
Following the Glance run check the existence of two intense arc-lines near the top and bottom of the chip. See figure 4.9. Then
SETUP (Loads software for TILT and FOCUS) TILT and answer prompts. exposure time > 10 secs. (when prompted).
This routine then does two exposures opening one Hartmann shutter at a time (first LEFT then RIGHT) and finally prints out the shift at 6 positions as above. Note that there is a variation due to the effects of pixel-sampling, i.e. if TILT is repeated shifts may vary by .05 - 0.1.
Tilt correction is achieved by adjusting cryostat screws with care. Approximately 1/2 turn clockwise of screws A, B or C will increase the tilt vectors by 1 line shift. Screw B should tilt the cryostat (and CCD) about its long axis, aligned with the dispersion direction. It is not possible to adjust tilt without modifying the focus, but hopefully this adjustment should rarely be needed. A change in focus (or tilt) of about 0.1 or 0.2 (Hartmann shift) across the CCD does not seem to be significant; and indeed very time-consuming to perfect.
Referring to Figure 4.8 again, slacken the three locking nuts before attempting to adjust the cryostat position. Movement is effected by the capstan screws. Also remember that the micrometers are for reading only, and cannot move the cryostat i.e. back off the micrometer, adjust the capstan and then gently screw in the micrometer and take a reading. For spectroscopic work it is the top-bottom (dispersion) tilt that is most important and for this adjust capstan B. On the 500 mm camera it is found that turning B anticlockwise by 1 turns, increases top-bottom by 1 pixel.
The best way of being confident about the ``tilt'' Hartmann shifts provided is to do two tilt set-ups without adjustment between them. The results should be similar in both cases. As with ROTATION, this work should only be done by a DT or a SA and is only necessary when a camera change has occurred.