Previous: Overview
Up: Overview
Next: Performance
Previous Page: Overview
Next Page: Performance

Design

The layout of the IDS spectrograph is shown in Figure 3.1. The following components can be discerned:

• Slit and dekker unit. The same slit and dekker unit is shared by the IDS and FOS-1. The maximum slit length is 44.3 mm (4 arcmin); the slit width is variable from 40 m (0.216 arcsec) to 1.745 mm (9.43 arcsec) in steps of 5 m (0.027 arcsec). The entire slit assembly can be removed, if required. Dekker plates are interchangeable in sets; each individual set consisting of 8 apertures and a clear position. Several standard dekkers, incorporating single slots, pairs of slots, a ``comb'', and a coronographic set-up, are always available.

  A multislit unit is also available. This consists of ten parallel slitlets, each 16s long with a 7s gap between each pair and with a fixed width of 270 m (1.46s at the Cassegrain focus). Each slit can be moved perpendicular to the slit orientation by up to 1.6 in either direction. With FOS-1 a useful field of 4 by 3 can be covered, whereas the IDS has a smaller field of 4 by 1 . A program is available for finding the optimum arrangement and typically one needs 20 or 30 objects in the field to make use of all 10 slits. Astrometry to within 0.3s is needed to use the unit effectively. There is only one module that has to be preset using a Coradograph xy table in the INT building. Clusters of objects spread over less than 1 are less likely to benefit as they are too small. To install the multislit assembly, the conventional Cass cluster slit assembly is replaced by a similar one which holds the multislit module or a long-slit equivalent. This instrument change requires lowering the IDS and is a daytime operation. The long slit module allows observations to continue while the multislit module is being preset with the Coradograph for a new field. The device should not be regarded as a general purpose multiobject coupler but rather as a special purpose addition to the INT Cassegrain cluster instrumentation. See Using the multislit unit at the INT Cassegrain focus (R. Ellis et al. 1986) for more information.

• Filter slides. In addition to the above-slit filter positions in the A&G unit, there are 3 filter positions available below the slit for both colour and ND filters. One of these positions is always clear. Five colour filters are currently available: UG11, BG18, BG24 and BG38, measuring 19 mm 60 mm 2 mm and made from Schott glass. The wavelength dependence of the transmission of these filters is shown in figures 2.5 and 2.6. The copper sulphate filter cuts out the part of the spectrum from 600 - 650 nm redwards and is usually employed to block the red leak of the UG11 filter or to remove the FOS-1 first order. The neutral density filters below the slit provide a choice of ND = 0.5, 1.0 and 1.5. Note that the use of filters below the slit makes it necessary to refocus the spectrograph.

• Collimator. The collimator is of 1275 mm focal length, giving a collimated beam with a diameter of 85 mm. There is a choice of three different collimators, with different coatings for maximum reflectivity in the wavelength range of interest. The efficiency of the collimators as a function of wavelength is shown in Figure 2.3. The collimators can be exchanged during the night by the support staff. The collimator is mounted on a remotely-driven sliding carriage, making it possible to focus the spectrograph under computer control.

• Cameras. Two cameras are available with the Intermediate Dispersion Spectrograph. Both are of the folded short-Schmidt type, described in detail by Wynne (M. N. R. A. S., 180, 485, 1977). The focal lengths of the two cameras are 235 mm and 500 mm.

  Along the direction of the slit (i.e. perpendicular to the dispersion direction), the focal lengths of 235 and 500 mm imply slit to detector reduction factors of 5.43 and 2.55 respectively, and hence scales at the detector of 29.4 and 13.8 arcsec mm.

  Along the dispersion direction, the slit to detector reduction factor is more complicated, since it depends on the grating angle. Tables 3.1 and 3.2 give the slit width in arcsec which projects to 30 m at the detector, when each grating is used at its central wavelength.

  Tables 3.1 and 3.2 also list the dispersion provided by each grating and camera combination. It can be seen that a wide range of dispersions is available, from 7 Å mm to 270 Å mm.

  

  

  

  

• Gratings. The gratings have a ruled area of 102 mm 128 mm. A wide selection is available, and some of the important characteristics of each grating are summarised in Tables 3.1 and 3.2. The efficiency of the gratings as a function of wavelength is shown in Figure 2.4. Grating changes can be made during the night by support staff.

• Cross-disperser. A transmission grism, R60B with 60 lines mm and blazed at 4800 Å, can be inserted into the collimated beam in the 235 mm camera to provide a cross-dispersion option. This is used in conjunction with an R150 echelle grating, which is blazed at 2.7 m. The FOS-1 dekker mask is used to limit the slitlength to 25 arcsec in order to avoid order confusion. The wavelength range for the CCD and dispersion provided by this grating in each order, for a grating angle of 60 , are summarised in Table 3.3. There are no gaps in wavelength coverage with the IPCS. The CCD is not normally considered the detector best suited to use with the cross-disperser, because part of the wavelength coverage is lost between the orders due to the physical size of the chip. However, it is usually possible to adjust the grating angle in the range 59.5 - 60.5 so as to ensure that all interesting spectral features are placed in the chip. Besides the advantages typical of CCDs, the combination of GEC chip and cross-disperser, seems to have a higher sensitivity than the combination of IPCS and cross-disperser. 13 counts/sec/Å were obtained for a m=11.9 star at an airmass of 1.32 with the CCD. 1 count/sec/Å therefore, can be expected for a m=14.8 at an airmass of 1.0. This count rate corresponds to a m=13.9 with the IPCS, but of course the IPCS's quantum efficiency is much lower at 5500 Å than that of the CCD. The central wavelength and wavelength coverage in Å can be obtained for each order as and

  

• Detectors. The IDS can be used with two detectors: the IPCS or a coated GEC P8603 CCD. Changing the detector on a particular camera can only be carried out during the day. However, it is possible to have a CCD set up on one camera and the IPCS on the other, and operate the two systems consecutively during the same night. Switching from one camera to the other only requires the grating to be rotated by 180 around the optical axis (or exchanged), and the control software to be reinitialized. Note that if you only specify a detector to be on one camera in your proposal, you cannot guarantee what will be on the other (e.g. it might be in use on another telescope). If you need both cameras, you have to specify that well before your run.