The detector efficiency is divided into the real efficiency and the dead time which is due to the delay following a photon detection during which another event can not be recorded.
The observed count rate is connected with the real count rate C as
C_o = C*(1-D)
The presence of the BGO shielding (see Section ), calibration source, and the Compton coding induces an efficiency loss due to good events which by chance fall in their respective time windows.
For IBIS the term ``efficiency'' is reserved for the real physical detection efficiency. The effects induced by Veto, calibration source, and Compton tagging are considered as additional dead times to be added to instrumental dead time.
Numerically, dead time is calculated as the product of the count rate and the time window (e.g. the dead time due to the time of the photon detection is equal to the product of the module count rate and the coding time. The dead time due to the veto is equal to the product of the veto count rate and the veto time window, etc.) The resulting dead time is taken as a sum of all the dead times connected with different effects. Simple summing of dead times can be done since the estimations show that the coincidence probability between different types of events is low. The values of the count rates is taken from the HK Data Structures IBIS-DPE.-HRW and IBIS-DPE.-CNV and the corresponding time windows are either measured (instrumental) or read into HK or IC (all others: VETO, Compton).