2.7 Event Timing

Each event detected with the HRI is time tagged with a precision of 61 $\mu$s relative to the ROSAT spacecraft clock. The processing time for determining the position of an event is dependent upon the event location between the readout taps in the crossed grid charge detector and varies from 0.36 to 1.35 ms. For an event rate of 100 counts s^-1, the dead time losses will vary from 4 to 15 percent. The standard dead time correction formula can be used: n_t = n(1 + n$\rho$) where n_t is the true rate, n is the observed rate, and $\rho$ is the dead time per event. The dead time varies linearly with position, but since source photons will be recorded over a distribution of fine positions due to spacecraft motion and the point spread function of the telescope, there is no single value of dead time for a particular source observation. As a first approximation, it is possible to use a typical value of $\rho$ = 0.85ms.

A separate scalar on the HRI counts preprocessed events with a dead time of only 30 $\mu$s. The contents of this scalar are entered into the telemetry once each second as secondary science. Thus if the HRI is observing a bright source, the intensity as derived from the processed events in the image can be corrected for the dead time using the data rates determined from the secondary science scalar.