XMM-Newton Users Handbook

3.4.4.8.1 Pile-up prediction

For the normal spectroscopy modes listed in Table 8, it is estimated that in rough terms, any individual CCD with a total count rate in all orders of 12 counts s$^{-1}$ in RGS1 (6 counts s$^{-1}$ in RGS2) may suffer from pile-up of about 2%. It should also be borne in mind that lower limits apply for emission-line-rich sources in which some pixels are much more brightly illuminated than others.

Table 11 provides estimates of the fluxes in the individual RGS1 and RGS2 CCDs for which pile-up of the order of 2% is expected in point sources with a smooth continuum.

Table 11: RGS 2% pile-up flux limits per CCD$^1$ for point sources in Spectroscopy mode and using 8 CCDs$^2$

Instrument	CCD	Flux	Instrument	CCD	Flux
		erg cm$^{-2}$ s$^{-1}$			erg cm$^{-2}$ s$^{-1}$
RGS1	1	7.1 $\times 10^{-10}$	RGS2	1	3.1 $\times 10^{-10}$
RGS1	2	5.0 $\times 10^{-10}$	RGS2	2	2.1 $\times 10^{-10}$
RGS1	3	3.8 $\times 10^{-10}$	RGS2	3	1.6 $\times 10^{-10}$
RGS1	4	3.3 $\times 10^{-10}$	RGS2	4	-
RGS1	5	2.9 $\times 10^{-10}$	RGS2	5	1.5 $\times 10^{-10}$
RGS1	6	3.2 $\times 10^{-10}$	RGS2	6	1.4 $\times 10^{-10}$
RGS1	7	-	RGS2	7	1.9 $\times 10^{-10}$
RGS1	8	7.1 $\times 10^{-10}$	RGS2	8	3.7 $\times 10^{-10}$
RGS1	9	20.0 $\times 10^{-10}$	RGS2	9	14.1 $\times 10^{-10}$
Double-node readout			Single-node readout

$^1$ An estimate of which energies are sampled for a given CCD is provided in Table 9.
$^2$ Limits for normal Spectroscopy or HCR modes reading 8 CCDs per exposure.
If n CCDs are read, the limits change by a factor of $\approx {\rm n}/8$.
In Small Window mode the limits are reduced by a factor $\sim 4 \times {\rm n}/8$, see § 3.4.5.1.

Pile-up on bright, narrow, emission lines should not be a major point of concern. The brightest emission lines in Capella (Fig. 88) suffer from pile-up of about 5% in RGS2 (with $\sim$ 0.04 counts per pixel per frame) and half that value in RGS1. Only two other active stars (HR 1099 and AB Dor) have occasionally reached count rates of this order during flares.