The detection and characterization of X-ray periodic signals play a role of paramount importance
in the process of identifying new compact objects or new classes of them, and studying the mecha-
nisms powering the observed emission. New emission mechanisms have been discovered in this way.
The greatest part of the periodic signals arise from the rotation of a compact star or from the orbital
motion in a binary system. The most important examples in which the modulation is due to the compact
star spin are:

(a) accreting magnetic neutron stars (NSs) in X-ray binary systems;
(b) spinning down isolated NSs, the emission of which may be powered by the dissipation of rotational,
thermal, or even magnetic energy (as in the cases of classical radio pulsars, X-ray dim isolated NSs,
and magnetars, respectively);
(c) magnetic white dwarf (WD) systems, such as polars and intermediate polars (IPs).
Orbital modulations of the X-ray flux are observed in many classes of NS and black hole (BH) X-ray
binaries and cataclysmic variables (CVs).


Artist’s impression of a pulsar. Ⓒ Guillermo Rodriguez


The discovery of a periodic modulation generally happens through the timing analysis of the X-ray
light curves of the sources which lie in the field of view, FoV, of a detector. However, for most
observations taken with imaging instruments, only the target of the pointing is analysed. With the
high energy missions currently operational (such as Chandra, XMM–Newton, and Swift), which have wide
energy ranges, high sensitivities and, often, high time resolution, the average number of serendipitous
X-ray sources detected in a typical Chandra observation is ∼40 though, during observations where all
the CCDs (each having a FoV of 8.3′ x 8.3′ ) are operating and/or pointed to crowded fields, up to about
100 sources can be detected. It is therefore evident that up to 99% of the information may remain
unexplored. Since the number of high quality time series stored in the present-day X-ray archives has
now reached the millions, these data clearly hold a huge potential for new discoveries and make data
mining an urgent task to achieve them. Various projects aimed at the search for X-ray pulsators were
carried out in the past. During the 1990s, our team carried out systematic timing analyses of the then
booming number of bright serendipitous sources—about 50 000 objects detected with ROSAT and EXOSAT
(Israel et al. 1998). The effort was based on discrete fast Fourier transform (FFT) analysis and resulted,
among other findings, in the detection of pulsations from sources which have become prototypes of new
classes, such as the anomalous X-ray pulsars (4U 0142+614; Israel et al. 1994) and the double-degenerate
ultra-short period X-ray binaries hosting two WDs (HM Cnc; Israel et al. 2002). We also discovered an X-ray
pulsator in a binary systems in a previously unobserved evolutionary phase, a post-common envelope stage:
HD 497985 (RX J0648.0–4418). Which is a 13-s-period source which hosts either one of the most massive
and fastest spinning WDs observed so far, or an unusual accreting NS spinning up at a low and extremely
steady rate (Israel et al. 1997, p. 411; Mereghetti et al. 2009, 2016).

The ChAndra Timing Survey at Brera And Roma astronomical observatories (CATS @ BAR)
project, is the largest ever systematic search for coherent signals in the classic X-ray band. Among the
aims of CATS @ BAR there are: (i) to identify intrinsically faint X-ray pulsators at a flux level which
has remained unexplored so far; (ii) to extend the validity of already known emission mechanisms
towards lower X-ray fluxes; (iii) to look for new classes of compact objects and new astrophysical
objects showing coherent signals. We have already found about 40 new X-ray pulsators, with periods
in the 8–64 000 s range. For a relatively large fraction of them, we rely upon more than one archival
pointing, allowing us to check for the reliability of the signal detection. A by-product of our systematic
search analysis is a detailed map of the spurious (instrumental) coherent signals which affect the ACIS.
In our live catalogue we report the current results of our search for new X-ray pulsators in all the archival
data of the Chandra Advanced CCD Imaging Spectrometer  Imaging (I) or Spectroscopic (S) array in
imaging mode. We have fo far analysed about 10 700 datasets, corresponding to the first ∼15 years of
Chandra operations, and applied a FFT analysis coupled with an ad-hoc signal detection algorithm to about
190 000 of about 430 000 light curves extracted.

For more details see our  published work on these discoveries. See as well previous discoveries within the
CATS @ BAR Project:

and check out our on-line catalogue of new X-ray pulsators for the latest discovered X-ray pulsators.