ISOCAM is used in its so-called parallel mode while observations are performed using other ISO instruments. When an ISOCAM observation is scheduled, it arrives on the source in parallel configuration and it is put back in parallel mode at the end of the AOT, before leaving the source. There are two standard parallel mode configurations, but ISOCAM always goes through a standby mode at the beginning of the observation which is equivalent to the most sensitive of the two, using the LW2 filter and the 6" pixel field-of-view (pfov) lens. This configuration will saturate for point sources brighter than about 0.5 Jy in the IRAS 12 micron band, assuming a stellar spectrum. The saturation limit in the LW2 filter in this configuration is 0.6 Jy.
Therefore, if the target source is bright enough, ISOCAM may saturate when the source is acquired, even though the observation itself is safe as far as saturation risks. ISOCAM saturation is highly undesirable as it introduces transients which can affect the instrument for a long period, up to 1 hour or more. In the most extreme cases, a strong saturation event can leave artifacts which contaminate subsequent observations (especially of faint sources) until the end of the revolution. In order to allow observations of relatively bright sources to coexist with the parallel mode, a change has been made to the details of the AOT-to-OCT logic, and to PGA, to allow the observer to control the instrument set-up before and after the actual observation. This is done by adding dummy observations with the desired observing configurations to the beginning and to the end of a concatenated chain of observations. Special values of the DRK parameter are used, in all three AOTs, as flags to suppress the automatic transition to parallel/standby configuration between the real and dummy observations. Details of how to implement these changes are given in Chapter 7. Of course, this procedure has a price. The two dummy observations add 4 minutes to each concatenated chain of observations using them. Care should be taken in the assessment of the saturation risk. It is not straightforward to transpose a 12 microns IRAS flux, to an LW2, 6" lens flux. Actual spatial extension of the source and its spectral shape should be taken into account. It may be possible to avoid dummy pointings in raster observations if the conditions at the first and/or last raster point permit safe start-up or resumption of parallel mode.