search

StingraySoftware / HENDRICS

Shell scripts for spectral-timing analysis of X-ray astronomical data.
http://hendrics.readthedocs.io
Other
22 stars 13 forks source link

CPDS Rebinning #50

Closed annacoerver closed 6 years ago

annacoerver commented 6 years ago

Hello,

I'm trying to rebin a CPDS (below), and am encountering issues with the geometric rebinning option. I've tried 1.03, 1.001, and 1.0001, and got the same plot all three times. Attached is the 1.001 rebinned spectra and the 1.0001 rebinned spectra . Constant binning seemed to be producing the expected results. My command line was as follows: HENrebin bary_r30_nustar_bin-6_E3-79_cpds.p -r 1.001 Is this user error? Any insight appreciated!

Original CPDS:

screen shot 2018-11-12 at 11 35 38 am

Rebinned CPDS (1.001 and 1.0001):

screen shot 2018-11-12 at 11 28 28 am

We also noticed that our PDS and CPDS cut off at 10^-3 at the low frequency every time. Do you know why that might be?

Thanks in advance! Anna

annacoerver commented 6 years ago

I just read the last issue related to HENrebin, and see that I should be using 0.001, not 1.001, in the command. (The HENDRICS documentation example uses 1.03)

The question about the PDS cutting off at 10^-3 still stands, if anyone has any insight!

Thanks, Anna

matteobachetti commented 6 years ago

@annacoerver sorry for not answering before. The lower frequency is driven by the length of the light curve segments that you calculate each PDS on. 0.002 Hz (the lower frequency I see in the plots) corresponds to ~500 s. Probably, you are using HENfspec with the default option, that cuts the light curve in segments of 512s, calculates the cpds on each segment, and averages them all.