Timing Analysis with XMM-Newton EPIC data

The following thread shows you in principle how you can extract interesting timing information out of XMM-Newton EPIC data. You will find the period of a variable source and create a folded light curve of the object

The starting point is a calibrated, concatenated event list (run the pipeline on the ODF or take the EVENT file from the pipeline products).
Note, that the specific parameters are related to the example object in the workshop.
We will run SASand XRONOS.

In pink is marked what you have to type at the command line or some useful scripts.
 

##########################script to extract a timing event file form ODF################################
#!/bin/csh -f
source $HOME/setsas.csh
or ...
setenv SAS_DIR /sas/Linux/RedHat9/sas6_1_0
setenv SAS_PATH $SAS_DIR
source $SAS_DIR/sas-setup.csh
...
setenv SAS_ODF /SAS_Workshop/Timing/PSRB1509
setenv SAS_CCFPATH /ccf/pub
setenv SAS_CCF ccf.cif
 

echo running cifbuild
cifbuild --withccfpath=no

echo running odfingest
odfingest --odfdir=$SAS_ODF
set sumfile=`ls -1 *SUM.SAS`
setenv SAS_ODF $sumfile
epproc timing=yes

#################################################################################################

1. Set the environmental variables
source $HOME/setsas.csh
...
setenv SAS_ODF /SAS_Workshop/Timing/PSRB1509
setenv SAS_CCFPATH /ccf/pub
setenv SAS_CCF ccf.cif
set sumfile=`ls -1 *SUM.SAS`
setenv SAS_ODF $sumfile

2. Do the barycentric correction with barycen

identify your event file and make a copy of it
ls *TimingEvts*

cp *TimingEvts* events_bary.ds

work now on the events_bary.ds

start SAS
sas

select the barycen tool
select your events_bary.ds and run barycen. (The TIME column will be overwritten by the 7.0.0 corrected times.)

3. Extract the object region with xmmselect
extract the region where your source is located
xmmselect table=events_bary.ds
select the region where the source is located (see Spectral analysis of PN point-like sources)
and save this event list as source.ds
 
 
 

4. Find the period of the object with the powspec tool
(http://heasarc.gsfc.nasa.gov/docs/xanadu/xronos/xronos.html)

powspec
Ser. 1 filename +options (or @file of filenames +options):
source.ds
Name of the window file ('-' for default window)
-
 Expected Start ... 11794.26074101622  (days)       6:15:28: 24  (h:m:s:ms)
 Expected Stop .... 11794.37184408115  (days)       8:55:27:329  (h:m:s:ms)
 

 Arrival Time Input File(s)
 

 Default Newbin Time is:   1.1718760    (s) (to have 1 Intv. of    8192 Newbins) Type INDEF to accept the default value

Newbin Time or negative rebinning:
0.02 ( you have to give here a binning that is smaller than the period of the object ( in the example 150 ms))

Newbin Time ......    0.20000000E-01  (s)
 Maximum Newbin No.            479966

 Default Newbins per Interval are:        8192
 (giving      59 Intervals of         8192 Newbins each)
 Type INDEF to accept the default value

Number of Newbins/Interval
479966
 Maximum of       1 Intvs. with       524288 Newbins of      0.200000E-01 (s)
Default intervals per frame are:         1
Type INDEF to accept the default value
Number of Intervals/Frame
1
Results from up to       1 Intvs. will be averaged in a Frame
Rebin results? (>1 const rebin, <-1 geom. rebin, 0 none)
0
Name of output file[default]
(enter)
Do you want to plot your results?[yes]
(enter)
Enter PGPLOT device[/XW]
(enter)

and WAIT until power spectrum appears.

to zoom to the highest peak use

r x 5 10   ( if the highest peak is located between 5 and 10 Hz)

find in that way the global maximum of the power distribution.
This value in Hz gives you a start value for further analysis. Change unit to [s]  via P=1/f  (one of the formulas easy to keep in mind...)

exit
 

5. Find the exact period of the object with the efsearch tool
(http://heasarc.gsfc.nasa.gov/docs/xanadu/xronos/xronos.html)

The following lines are a small csh script, that executes the period search via folding versus homogeneity for this special object. You can also type the commands one after the other at the command line, but because you might have to run this more often and later with a high accuracy (what takes some time) it is nice do get this done by the script. The script produces also a .ps file. For this you have to create a plot file (see below p_search.pco)

##########################################script####p_search.csh##########################################
#!/bin/csh -f

efsearch cfile1="source.ds" \
window='-' \
sepoch=11794.31629254868\
dper=0.151114 \
nper=10 \
nphase= 10 \
nbint=INDEF \
dres=0.1E-5 \
plot=yes plotdev="/ps" outfile="-" \
plotfile="p_search.pco"

#######################################################################################################
explanation of keywords can be found at: http://heasarc.gsfc.nasa.gov/docs/xanadu/xronos/xrohlp.html

###############################plotfile###########p_search.pco############################################
hardcopy period.ps/ps
exit
######################################################################################################
 
 
 

6. Produce the folded light curve of the object with the efold tool
(http://heasarc.gsfc.nasa.gov/docs/xanadu/xronos/xronos.html)

Also here we provide you with a example scripts, that will work for the data in the workshop.

##########################################script###fold.csh##############################################
#!/bin/csh -f

efold \
nser=1 \
cfile1="source.ds" \
window='-' \
sepoch=11794.31629254868 \
dper=0.151113327 \
nphase= 100 \
nbint=INDEF \
nintfm=1 \
plot=yes plotdev="/ps" outfile="-" \
plotfile="fold.pco"

######################################################################################################
explanation of keywords can be found at: http://heasarc.gsfc.nasa.gov/docs/xanadu/xronos/xrohlp.html

###############################plotfile###########fold.pco############################################
hardcopy light_curve.ps/ps
exit
######################################################################################################