Previous: Rotation
Up: Setting up the CCD Detectors
Next: Collimator Focus
Previous Page: Rotation
Next Page: Collimator Focus

Tilt and coarse Focus

The tilt of the cryostat can be adjusted in two dimensions to ensure that the CCD is coplanar with the focal plane of the spectrograph. The cryostat is held to the mounting ring by three clamps on the mounting ring which each hold a capstan attached to the cryostat flange against the mounting ring. The height of the cryostat can be adjusted by rotating the capstans, and can be measured with three micrometers, mounted on the cryostat flange alongside each capstan, which can be brought against a metal stud in the surface of the mounting ring. The capstans can be locked by tightening the thread that they run through with an Allen screw in the cryostat flange.

The tilt of the CCD can be checked with the FOCUS procedure on the DMS. With a slit width of around 150 microns and a long slit dekker position set the exposure time, filters and central wavelength to give a uniform distribution of line from one of the comparison lamps; make sure that there are reasonably strong lines near each end of the spectrum. The exposure time should be set to give a reasonably strong signal, but the lines to be used must not exceed 32767 adu at any point along their length. Set up a CCD window about 600 pixels wide with the ICL procedure WINDOW; close the left Hartmann shutter by typing:--

BHART 1 or RHART 1

at the ICL interface and take an exposure with GLANCE When it has read out, at the DMS control window type:--

FOCUS

Select three strong lines evenly spaced from the one end of the spectrum to the other by placing the cursor on each in turn and pressing the space bar on the DMS keyboard. Each line profile will be shown on the DMS display, and the line can be rejected if it appears too weak, or blended with or close to another line. The best results will come from strong isolated lines. When three lines have been selected type:--

FOCUS-LEFT

at the DMS control window. The programme will then list the centroid and an estimate of the full width half maximum at three positions along each line. The arrangement of the listed positions on the alphanumeric screen maps to the DMS display, so the first position listed is in the top left hand corner of the image as displayed on the DMS display.

Now open the left Hartmann shutter and close the right by typing:--

RHART 2 or BHART 2

at the ICL interface and take another exposure. At the DMS control window type:--

FOCUS-RIGHT

This routine will list the positions and the full width half maxima of the lines together with the shifts since the first exposure. These Hartmann shifts indicate how far from the focal plane the CCD is at each point, and the aim of the tilt adjustment is to make all of the shifts the same. If the shifts are more than 3 pixels then the spectrograph is probably too far out of focus for this procedure to succeed, and it will be necessary to adjust the collimator position with BCOLL or RCOLL, or else to move all three capstans so that the spectrograph is closer to focus.

The relative Hartmann shift from one end of the spectrum to the other, or from one end of the slit to the other, should be no more than about 0.2 pixels. If they are more than this then the capstans will need to be adjusted. There are three capstans, labelled A, B and C. The dispersion is horizontal when the telescope is parked at the zenith; moving capstan B affects the tilt along the slit, whereas moving capstan A or C affects both the tilt along the spectrum and that along the slit. For this reason it is easier to adjust the tilt along the spectrum first, and then the tilt along the slit.

To move a capstan the micrometer should be read, then backed off. The small Allen screw should be loosened until the capstan is free to rotate. It may be possible to rotate the capstan without releasing the clamp, this is probably the best thing to do if the amount of capstan rotation is less than one complete turn. If it is larger the clamp should be released; the capstan turned by an appropriate amount with one hand, and the clamp applied again; all the while the other hand should be used to hold the cryostat firmly in place. After the clamp has been tightened the micrometer can be read again. The micrometers should never be used to take the weight of the cryostat, and should only be brought into contact with the studs when the capstans are firmly clamped.

One complete turn of a capstan will result in a change in the micrometer reading of 0.5 millimetres. A one pixel differential shift along the spectrum will require a movement of capstan A or C of 4 complete turns, or 2 turns of each in opposite directions, which will keep the focus in the centre of the chip the same. A one pixel differential shift along the slit (600 CCD pixels) will require 7 complete turns of capstan B to correct it.

The sign of the differential Hartmann shift measured by this procedure will depend upon the way that the CCD is mounted in the cryostat, and which on-chip amplifier is being used. The telescope manager or technical staff may know the direction, but if they are not available it can be established by trial and error.

The Hartmann test should be repeated after every move of the capstans; it should not be necessary to run the FOCUS routine each time because the lines should not shift very far. If FOCUS-LEFT gives widely different centroids or widths for what should be the same line then it may be necessary to run FOCUS again. When the differential shifts are satisfactory the micrometer readings should be noted in the ISIS log, together with the date and details of the cryostat used. The micrometers should then be backed off, and the firmness of the clamps checked. It is not normally necessary to tighten the small Allen screws, unless the capstan is particularly loose



Previous: Rotation
Up: Setting up the CCD Detectors
Next: Collimator Focus
Previous Page: Rotation
Next Page: Collimator Focus


Fri Jan 7 15:34:48 GMT 1994