Although there is no special observing mode for variability measurements, the standard photometric modes (PHT03 and PHT22) can be used in possibly in combination with special mission planning features, in order to cover a wide range of temporal resolution. Thus, variability time intervals between 1/16 sec (PHT-P) or 1/8 sec (PHT-C), respectively, and 6 months (mission duration 18 months) can be monitored.
The longest contiguous exposure time for an individual photometric AOT module is 2048 sec. If longer monitoring times shall be achieved, a concatenation of the same observing mode can be performed. The source monitoring will then be interrupted by the internal calibration performed at the end of each individual photometric module and the instrument set-up at the beginning of the next module. For variability periods longer than several hours, no continuous measurement can be performed, and only parts of the object's light curve can be monitored. If certain phases shall be matched, the mission planning features ``fixed time'' and ``periodic observations'' (cf. ISO Observer's Manual) should be used. For variability periods considerably shorter than the maximum AOT exposure time, the required temporal resolution can be obtained by averaging only the corresponding number of integration ramps. The number of ramps can be reduced to only one integration ramp per temporal resolution element leading to the above mentioned numbers of 1/16 or 1/8 sec.
For an estimate of the achievable accuracy the following rule of thumb plus
additional caveats can be given: The mean source flux 
must be
monitored with
sufficient S/N in order to recognize true source flux changes
against the
noise level. If we request that the relative source flux change 
shall exceed a 5 
detection limit, then
can be measured, where S/N is the signal-to-noise ratio for the mean
source flux achieved during the time resolution element. The formulae given in
sections 4.2 and 6.4.3, respectively, should be
used, with the requested time
resolution as integration time. For instance, if the source flux can be measured with
S/N = 50, the above formula tells that a variability amplitude of 10% can be
detected with 
. Furthermore, the detector signals must be stable,
yielding a minimum monitoring time of 32 sec for repeated measurements (in order
to achieve a sufficient relative accuracy between the measurements (cf.
Sect. 2.8), and a recommended minimum monitoring time of 128 sec for a single
measurement monitoring short-term variations.
The normal data reduction will be sufficient for those measurements aiming to detect long-term source variations measured at several phases. For short-term variations in the time scale of only one or a few reset intervals, a dedicated data reduction scheme different from the normal pipeline processing will have to be applied. Correlation functions will help to enhance weak periodic source variations, and investigation of short-term irregular intensity variations (flickering) may be hampered by occasional cosmic ray hits of the detectors. For more detailed information about an optimised observing strategy or instrument set-up for variability measurements, please contact the ISOPHOT PI or the ISOPHOT Instrument Dedicated Team.