1998A&A...333..942O

We report on the Rossi X-ray Timing Explorer Proportional Counter Array (PCA) observation of the X-ray burster 1E 1724-3045 located in the globular cluster Terzan 2. The observation lasted for about 100 kiloseconds and spanned from November 4th to 8th, 1996. The PCA source count rate in the 2-20keV range was about 470cts/s. No large spectral variations were observed within our observation as inferred from color-color and hardness-intensity diagram analysis. The persistent X-ray emission shows a high level of noise variability, the so-called High Frequency Noise (HFN) with a fractional Root Mean Squared (RMS) of ∼25% in the 2x10^–3-40Hz range. The strong HFN together with the hardness of its X-ray spectrum suggest that 1E 1724-3045 is an ``Atoll'' source which was in its ``Island'' state during the observation. The Fourier Power Density Spectra (PDS) can be modeled in terms of the sum of two ``shot noise'' components for which the shots have a single-side exponential shape (i.e. instantaneous rise and exponential decay). The characteristic shot decay timescales inferred from the best fitting of the PDS are ∼680 and 16msec respectively. The two components contribute similarly to the total RMS (∼15%). The PDS contains also a third component: a broad and asymmetric peaked noise feature centered at 0.8Hz. This Quasi Periodic Oscillation-like (QPO) feature contributes at the level of ∼10% to the total RMS. Neither the shot timescales nor the QPO frequency vary with energy. On the other hand, the integrated RMS of all three components shows a positive correlation with energy up to at least 40keV. In addition, in the 2-20 keV energy band where the signal to noise ratio is the highest, we found evidence for a high frequency component which shows up in the PDS above 100Hz. In terms of the shot noise model, a Lorentzian fit of this last component implies a shot decay timescale of ∼1.4ms. We also show that 1E 1724-3045 has striking timing similarities with the black hole candidate GRO J0422+32 (Nova Persei 1992). Our observation demonstrates that a low frequency QPO simultaneously with a high level of RMS is not a timing signature unique to black holes. This extends the growing list of similarities between Atoll sources and black hole systems.