XMM-Newton Users Handbook


next up previous contents
Next: 4 Observing with XMM-Newton Up: 3.7.1 A comparison of XMM-Newton vs. Chandra Previous: 3.7.1.2 Non-dispersive spectroscopy: an example


3.7.1.3 XMM-Newton EPIC vs. Chandra ACIS-I pile-up comparison

It has been shown above (§ 3.3.9) how pile-up affects the accuracy of spectral fits and the shape of the X-ray PSF. Figs. 115 and 116 show the fraction of piled-up events for different numbers of counts per CCD frame (in full window imaging mode). One can see in Fig. 116 that pile-up effects for given source fluxes are more severe for ACIS-I by more than an order of magnitude compared to EPIC MOS and even by two orders of magnitude compared to EPIC pn.

Figure 115: Comparison of Chandra ACIS-I vs. XMM-Newton EPIC (pn and MOS) pile-up for different total frame count rates. The frame times are 3.3, 2.8 and 0.07 seconds for ACIS-I, MOS and pn, respectively.
\begin{figure}\begin{center}
\leavevmode
\epsfig{width=0.78\hsize, file=figs/Pile-up_2.ps}
\end{center}
\end{figure}

Figure 116: Comparison of Chandra ACIS-I vs. XMM-Newton EPIC (pn and MOS) pile-up for different incident source fluxes, after conversion of counts per frame to flux units, adopting an $\alpha = -1.7$ power law spectrum with an absorbing hydrogen column density of $3\times 10^{20}$ cm$^{-2}$.
\begin{figure}\begin{center}
\leavevmode
\epsfig{width=0.78\hsize, file=figs/Pile-up_5.ps}
\end{center}
\end{figure}


next up previous contents
Next: 4 Observing with XMM-Newton Up: 3.7.1 A comparison of XMM-Newton vs. Chandra Previous: 3.7.1.2 Non-dispersive spectroscopy: an example
European Space Agency - XMM-Newton Science Operations Centre