|ROSAT Home Page||ROSAT
Off-axis rays strike the mirror surface at a shallower angle such that the projected geometric area decreases. Furthermore (but to a lesser extent), the average reflectivity of the gold surface is reduced. In addition, off-axis rays suffer a higher degree of obstruction due to the radial struts which stabilize the XMA (see Sect. 2.1.1), and for very large off-axis angles due to the field-stops, which are introduced to reduce stray light. Together, these effects lead to an energy-dependent decrease in the effective area of the XRT as a function of (increasing) off-axis angle (collectively referred to as vignetting).
There is an additional effect due to the magnetic deflector. This was designed such that it does not additionally block X-rays on-axis. However, for an offset angle , the radial support struts of the deflector are no longer in a projected shadow of the XMA support struts, and therefore an additional loss is encountered - see also [Aschenbach1990b].
To verify the theoretical vignetting function using in-flight data, the radial profiles of the detector maps created from all-sky survey data [Snowden et al.1994] were studied.
After masking out the window-support structure ring and ribs and binning the counts in radial annuli, the count-rate ratios then give the vignetting at very high S/N. There is good agreement with the preflight vignetting function (which was parameterized as a double power-law function) out to . Above 40', there are small discrepancies where the model diverges from the data. The survey data initially dip slightly under the model but then rise up and go above the model, crossing it at about 54'. The size of this discrepancy and the radius where is starts appears to be a function of energy. This effect could be due to systematic errors in the PSPC event linearization, in particular due to the ``window correction'' (see Sect. 3.5.3).