The Ohio State University College of Mathematical & Physical Sciences Department of Astronomy

Optics

Paul Byard reported on his ADC design efforts. His working design was for a full-field ADC located 0.5-meters in front of the focal (slit) plane, with a baseline design modeled on the ADC used in the Blanco 4-meter telescope at CTIO. This optic will be 350mm in diameter and 280mm thick (!), requiring 10-degree prisms. The total mass of glass is 75kg. Internal absorption gives a thruput of 74% at 370nm is 74%, dropping to 10% at 334nm (the corrector scales linearly in size, but absorption scales exponentially, of course).

There are other impacts, including a change in the telescope back focus amounting to about 1.5-microns at the edge of the secondary. The full consequences are TBD. Bruce suggested Paul consider an ADC located 1-meter from the focal plane, this would reduce the thickness by 50%, reducing the absorption. One might consider a combination of fused silica and calcium fluoride for a smaller ADC field, but larger pieces, esp. of CaF2, are probably prohibitively expensive or simply unattainable.

Note that while the instrumental baseline does not call for an ADC, the amount of differential atmospheric dispersion between 350 and 700nm is of order 2 arcseconds. An ADC can eliminate most of this, but at some cost as this exercise shows. More food for thought...

In other items, Paul has also been looking at the design of a direct viewing prism that would provide an R=100 @ 600nm low-res mode for MODs. We also need to explore how simlarly designed prisms might server as cross-dispersers for R=8000 or R=16000 modes.

Mechanical Systems

Tom O'Brien is on vacation this week.

Acquisition & Guiding

There was considerable discussion of possible A&G issues. Most of the discussion was about the process for "field acquisition and slitmask alignment" in multi-object mode, although other modes (single slit) would use largely the same system. Two basic modes were discussed:

Direct Imaging

Switch into direct imaging mode and view the focal plane with the science detector, then view the sky through the slitmask and align on the targets directly. This places requirements (TBD) on the amount of time to switch between configurations, the stability of the configuration, etc. It uses existing instrument functions. We do something like this with OSIRIS for single-slit spectroscopy, and it is a common field acquisition & mask alignment scheme used by other instruments (e.g., LRIS, DEIMOS, MOS/SIS, etc.).

Rear-Slit Viewer

Have a mobile CCD viewer that runs around behind the slit and looks out. This allows the instrument to stay in spectroscopic mode. Align by looking through holds cut in the mask at particular alignment stars, minimum number and disposition TBD. Would require an X-Y stage, pickoff optics, CCD detector and readout electronics, viewer camera focus motion, etc. There are no current examples of this type of field acquisition & mask alignment scheme in use or planned to our knowledge (although if anyone knows of such, please direct us).

Action items arising from this discussion:

1. How long does it (or should it) take to change the instrument configuration? This determines the impact on observing efficiency with using the direct imaging mode for target acquisition and mask alignment.

2. What are the requirements for target acquisition and alignment for all modes, not just multi-object mode? Is some kind of slit viewing capability for long-slit modes, for example, useful or desired by the potential user community?

The issue of flat-field and wavelength calibration has finally raised its ugly head. Two questions to be addressed in future meetings:

1. What do we need?
2. How do we do it?