Summary of meeting at the UKATC on 12/07/2000

Attendees

• Vik Dhillon.
• Tully Peacocke.

Optics

• Tully presented his initial ideas for the optical design of ULTRACAM. The collimator for the WHT Cassegrain design needs elements approximately 15cm in diameter to cope with the large linear diameter of a 6 arcminute field. Such large optics are expensive, especially the Calcium Flouride element required for operation in the u-band. As the linear diameter depends on focal length, and most telescopes have Cassegrain focii of approximately f/10, the optics required for 2m telescopes are much smaller (approximately half the diameter) and hence more affordable. For this reason, Tully's initial designs were for Aristarchus, as this is the telescope where ULTRACAM will probably see most use.
• Tully's design for a 6 arcminute field on Aristarchus will work on the 2.5 arcminute field of the WHT Nasmyth focus at GHRIL, but the chip will be vignetted beyond this when used at WHT Cassegrain.
• Tully's Aristarchus design consists of a two-piece collimator which passes 320-980 nm. The front part of the collimator is interchangeable and cheap, and is the only optical element in ULTRACAM which needs to be changed when changing telescopes. The back part is expensive and fixed, and contains the CaF₂ element. The cameras are almost identical apart from their coatings and apart from the u-band camera, which has a CaF₂ element in it. The cameras are telecentric (i.e. they give a constant pixel scale regardless of the telescope) and the pupil is exterior to the cameras.
• Tully presented some spot diagrams for the Aristarchus design, showing that, at worst, the FWHM of the profile would be just over 1 pixel, i.e. approximately 0.3 pixels. This is under half the median seeing value of the La Palma site and should hence be fine. Tully will confirm the FWHM value, which is rather hard to determine directly from the spot diagrams, at a later date.
• Tully also presented an alternative design where the u-band is removed and ULTRACAM instead has a g-band arm, an r-band arm and an i/z-band arm. This improves the throughput and the optical quality, at the expense of the u-band. VSD emphasized the importance of the u-band to the science ULTRACAM hopes to perform and the idea was dropped.
• VSD re-iterated the importance of field of view to the ULTRACAM project. Without at least 4 arcminutes, and preferably 5, it becomes very difficult to find suitable comparison stars for differential photometry. VSD emphasized that field of view is more important than both throughput (as long as it is as high as is affordable) and image quality (as long as it is better than the median seeing) for the ULTRACAM project. Given this, we must aim to design ULTRACAM for operation on the longest focal length telescope we can afford, which given the fact that most Cassegrain focii are around f/10, means the largest aperture telescope (i.e. the WHT). Tully agreed with these points, but emphasized that the issue here is cost. He agreed to look at the WHT Cassegrain design again and come up with a costing for the optics.
• VSD emphasized that the £25k currently allocated to purchase the optical components could be added to from other areas of the grant (particularly the computing budget) so perhaps the 5 arcminute field on the WHT might be affordable. Furthermore, VSD stated that it should be possible to borrow filters and computers for the commissioning run, which could then be purchased via separate grant applications to the Royal Society, British Council or the Nuffield Foundation. Our overall policy is to ensure that the instrument is as flexible as possible, which means designing the optics for the most demanding telescope we can afford (i.e. a 5 arcminute field at the WHT Cassegrain). This will ensure that ULTRACAM works equally well on virtually all 4-m class telescopes (e.g. the AAT, Calar Alto, SOAR) as well as all 2-m class telescopes.
• VSD raised the issue of ghosts, formed by reflections within the dichroics. Tully felt that these could be adequately controlled using suitable coatings on the back faces of the dichroics. He agreed to look into it using a software package designed specifically to study ghosts in optical systems.
• VSD also raised the issue of the chip orientations. The dichroic reflections cause the u-band and g-band CCDs to be mirror images of the r/i/z-band arm. The u-band and g-band arms can be reconciled by 180 degree rotations, but the r-band arm will be a mirror image of these two. This means that any stars on the r-band CCD will be in different locations on the u-band and g-band CCDs, which means that one of DJI's requirements, that the windows are all on the same location on each chip, is violated. Tully pointed out, however, that they had experienced this kind of problem with NAOMI and it was a simply matter to fix by rewiring the r-band CCD so that the mirror image is compensated for using a different readout sequence.

Contract

• Design: Up to three different optical designs, including recommendations for coatings, will be fully costed prior to the Preliminary Design Review. These are:
  1. A design for a 5 arcminute field of view at WHT Cassegrain.
  2. If the above is not affordable, a design for as large a field of view as possible at WHT Cassegrain.
  3. If it proves impossible to afford a field of view larger than approximately 2.5 arcminutes at WHT Cassegrain, a design which delivers a 2.5 arcminute field at WHT Nasmyth and a 5 arcminute field on Aristarchus.
  The pixel scale in all of the above designs will be approximately 0.3 arcseconds per pixel. The image quality will be such that the stellar profile is dominated by the effects of seeing during median (0.7 arcsecond) conditions on La Palma. No throughput requirements will be specified, although it is understood that the maximum throughput which can be afforded and which maintains the required field of view will be strived for.
• Ordering: The UKATC will provide sufficient information to allow the customer to place a contract with an optical manufacturer for the components.
• Mounting: The CaF₂ elements will be mounted by the manufacturer, using a mounting barrel designed and manufactured by Sheffield, based on advice provided by the UKATC.
• Acceptance: The UKATC will perform acceptance tests of the optics using the interferograms supplied by the manufacturer.
• Alignment and testing: The UKATC will provide advice and, if necessary, staff effort during the final alignment and testing phase of the entire optical system at Sheffield or the UKATC (wherever proves easiest).
• Staff effort: The total staff effort required to deliver the above is estimated to be 30 days, to which will be added 5 days for contingency. This figure of 35 days is hence the maximum the customer will be charged for the work described above.
• Future work: The initial optical design will be for use at WHT Cassegrain. Future use at other telescopes will require different collimator designs. The UKATC agrees to undertake this work as and when it arrives, and will charge for it separately.

vsd, 19-july-2000