Документ взят из кэша поисковой машины. Адрес оригинального документа : http://hea-www.harvard.edu/~garcia/ixo/high_rate_agn_v3.pdf Дата изменения: Tue Jan 31 22:38:35 2012 Дата индексирования: Tue Oct 2 02:01:09 2012 Кодировка: Поисковые слова: п п п п п п п п п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п

Smithsonian Astrophysical Observatory Memorandum To : From Date: Re: M. Cappi, Strong Gravity SDT/YB Team, XMS team : M. Garcia April 14, 2010 XMS high/mid/low resolution event rates for bright AGN mrg-2010-02v2

This memo arises out of the concern that the bright AGN that will be used to observed strong gravity effects ­ in particular those for which time-resolved measurements of test particle orbits will be made ­ may generate counting rates in the XMS that are sufficient to overwhelm it. If this occurs events will not be lost, but the energy resolution of the events will be degraded. Our current science requirements flow-down specifies the XMS as the primary instrument for these observations and we desire the full resolution, so this possible outcome is undesirable. The option of using a `diffuser' (see Willingale, 2008 BSDO_NFI memo) for these observations is also undesirable, as it lowers the overall effective area by 62% at low energies and up to 80% at 8 keV. The conclusion is that there is some degradation in the energy resolution, but the effect is moderate. Even in the worst case ~70% of the events are either high- or mid-resolution. There should be very little impact on the science, and is certainly a smaller effect than the diffuser would entail. However, this effect should be evaluated again if the XMS decay time constant is increased beyond the current baseline of 300usec. We used the event simulator `simx' (http://ixo.cfa.harvard.edu/wiki/IXO/IXOSimulators) to estimate the rates, and the list of AGN from Table 4-3 (reproduced below) of the ConX BEPAC response (dated Jan 22 2007). Fluxes over 2-10 keV were taken from this table, and spectral slopes and NH were taken from McKernan, Yaqoob and Reynolds (2007 MNRAS 379, 1359) or Grupe etal (2010 ApJS 187, 64). Mean value of flux in this table is 4.1x10-11 cgs, median is 3.5x10-11. A histogram of the fluxes is included in the plots below. The 2-10 keV fluxes in this table are from an ASCA survey of bright AGN that was completed several years ago. Counting rates were calculated for a representative set of these AGN, using both the slumped glass and silicon pore optics effective area files. For the SPO these areas came from the file ixo_matrices_20103103.zip which came from Nicola Rando on March 31, 2010. For the glass optics the areas are from ixo_ucal_5p0_081030.rsp. These are plotted below, labeled as the 2008 Glass (green) and 2010 SPO (red) curves (from Andy Ptak). The definition for high/med/low resolution events is shown on the last figure below. We note that the use of 1.5 msec and 6.0 msec as the high and mid res time intervals appears to be based on a one pixel time constant of 150 usec, and high and mid intervals of 10 and 40 time constants. If the relevant time constant is instead 300 usec the intervals are likely 3.0 msec and 12 msec and the high counting rate effects are more prominent. Both cases are considered below. We also did a few cases with intervals of 1.6ms and 2.9ms,


re: Caroline Kilbourne's email of April 18, 2010. The energy resolution for the high events is 2.5eV, mid is expected to be 3.5eV, and low >10eV. The below is from Randall's Feb 2010 draft (not yet finished nor released) memo on SIMX:

Event Grades in simx
The event simulator simx contains routines to calculate deadtime, along with special code to handle the more complex situations that arise with microcalorimeters. The process occurs in two steps. Step 1 considers the time-ordered event lists, stepping through each event, looking to see if the next event on that pixel is within rec, (listed in the simx mission definition as DeadTimePerEvent), and if so, the event at best Mid-res. If it is also within the MidResDeadTime, the event is Low-res. Otherwise, it's counted as provisionally High-res. Then the event list is stepped through again, looking at each event and when the previous event occurred. If this preivous event occurred within rtb (MidResDeadTime) seconds, then simx counts the event as Low-res. Otherwise, it's left as was. There is no specific marking for secondary events. The values in the parameter file for IXO are (units are in seconds): IXO:XMScore:DeadTimePerEvent IXO:XMScore:MidResDeadTime 6e-3 1.5e-3

F lu x (2-10keV) 1.5ms, 6.0ms IC4329A MCG6-30-15 MKN 509 MKN 355 3ms, 12ms IC4329A MCG6-30-15 MKN 509 MKN 355 1.6ms, 2.9ms MKN 509 MKN 355 7e-11 4e-11 6.6e-11 1e-11



NH High

SPO M id 13% 12 16 7 2 1 2 1 5 4 1 8 3 3 Low 11% 9 14 6 2 1 2 1 1 7 5 0 Total rate 340c/s 280 460 160 High 71% 74 61 82 5 5 4 6 3 7 3 9 M id 16% 14 20 10 2 2 2 1 6 6 2 1 4 7

Glass Low 13% 12 19 8 2 2 3 1 5 2 3 4 Total rate 430c/s 380 640 240

1.83 1.76 1.63 2 .3

3e21 6e20 1e20 4e20

76% 79 70 87 5 6 5 7 8 5 2 7

Same values as above

81 83

14 13

75 77

19 17