2.1 Introduction

The ISOCAM instrument on board the Infrared Space Observatory ISO was designed to map selected areas of the sky in the spectral region from 2.5 to 18$\mu$m at various spatial and spectral resolutions.

Figure 2.1: The layout of ISOCAM.

With ISOCAM, spectral features at wavelengths inaccessible from the ground were studied. The morphology of objects at a scale of a few arcseconds was investigated through broad and narrow band filters. Spectral imaging was also possible with Circular Variable Filters (CVF) at a spectral resolution of up to $\Delta \lambda / \lambda = 40$.

Figure 2.2: One of the first ISOCAM raster observations was pointed at the M51 galaxy. The image is the sum of two LW3 (14.3$\mu$m) filter raster measurements (Sauvage et al. 1996, [51])

The spatial resolution was determined by the diffraction limit at longer wavelengths, and was limited mainly by undersampling of the PSF at shorter wavelengths. For a 60 cm diameter telescope the diameter in arcseconds of the Airy disc (FWHM) is given by $\lambda/D = \lambda~[\mu{\rm m}]/3$, or $2.4 \cdot \lambda/D = 0.8 \cdot \lambda~[\mu{\rm m}]$ for the first dark ring. Regarding the optical performance of ISO, images of point sources were made, clearly showing up to the fifth Airy diffraction ring. A detailed description of the telescope and the pointing system is given in the ISO Handbook Volume I, [40].

ISOCAM provided imaging capabilities across a field of view of up to 3$^{\prime}$ diameter. It consisted of two optical channels, used one at a time, each of which containing a $32\times 32$ pixels infrared detector array. The short wavelength channel (SW) operated between 2.5 and 5.2$\mu$m; the long wavelength channel (LW) between 4 and 18$\mu$m. Each channel included lenses covering a range of magnifications (yielding fields of view having 1.5, 3, 6 and 12 $^{\prime \prime }$ per pixel), fixed filters and CVFs. During the ISO pre-flight thermal vacuum tests it was discovered that the apparent output voltage of one of the LW detector columns (`column 24') was always at zero Volts. In space it was found that, because of the repositioning jitter in the camera wheels, the edges of the array (essentially the outermost columns), did not always receive sufficient light when using the 6 $^{\prime \prime }$ per pixel lens.