THE UNITED KINGDOM INFRARED TELESCOPE
ANNUAL REPORT
1995 AND 1996

4. Report on Operations

4.4. Instruments

4.4.2. CGS4 (1-5m multiple resolution spectrometer with256256 array)

CGS4, UKIRT's most popular instrument and its most productive for the last two years in terms of number of publications, was upgraded to a 256256 array and ALICE2 electronics in the first few months of 1995. Other changes were made as well: the array translation drive was modified to improve long term reliability, the filter wheels were modified in order to allow the installation of a Wollaston prism for spectro-polarimetry, and a new slit wheel containing individually rotatable slits was installed in order to make changing wavelengths and slit angles more efficient and to allow a wider choice of slit widths. Unfortunately a mechanical problem with the new slit wheel affected the operation of CGS4 much of the time from shortly after its commissioning in spring 1995 until the end of 1996. Most types of observations remained feasible but data reduction was complicated and at some wavelengths observations with the echelle were no longer possible. Two attempts to repair the problem were made but the problem recurred shortly after each attempt. Subsequent analysis by Fraser Morrison at ROE and Ian Pain at JAC suggested that the problem is due to a small amount of differential thermal contraction of parts of the slit wheel made of different low-expansion metals. New parts, which are expected to eliminate this problem once and for all, are being fabricated at ROE; they will be installed in CGS4 early in 1997. ²

In 1996 the 150 l/mm grating was commissioned with the 256256 array. At R2000 (with the short focal length camera) this grating separates the OH sky lines from one another, so that about 50% of the J and H and short K windows are OH line-free. Some PATT runs obtained deep exposures on faint sources and obtained sensitivities within a factor of 1.5 of the predicted inter-OH value. (See the science highlights for examples of how this capability is being used.)

The Spectroscopic Multi-object Infrared Fibre System (SMIRFS), probably the first infrared multi-object fibre feed built for astronomy, was commissioned with CGS4 in 1995. Although the weather was poor during the commissioning time, sufficient observations were obtained to demonstrate the practicality of the project. As expected the throughput of CGS4+fibre is about half that of CGS4 alone, but with 14 fibres imaged onto the slit, significant gains are possible for suitable sources. The biggest gains are in the J and H windows because the background beyond 2.2m due to the warm fibres and their mounting block is significantly higher than for CGS4 alone. The Durham group now plans to construct a prototype fibre integral field unit (IFU), to image a roughly rectangular region of area 25 square arcseconds onto the slit. An IFU probably has more scientific uses in the infrared than a system of discretely positioned fibres.

Since spring 1995 the on-line CGS4 data reduction has used a GUI interface on a Sun workstation. This change has been well received by users. The Unix version of CGS4DR is now in use off-line as well, and data reduction in the VMS environment has been de-commissioned.