Variable X-ray source near Cen X-3

Abstract

THE satellite Ariel V, launched on October 15, 1974, as part of the US-UK collaborative space programme, contains a rotation modulation collimator experiment designed to measure the positions of X-ray sources, which was built by the Mullard Space Science Laboratory. The experiment responds to X rays in the quantum energy range 3.1–9.3 keV, the area of the proportional counters being 256 cm² which, allowing for grid transmission, counter efficiency and so on, leads to an effective area of 102 cm² at 6.0 keV. The pitch-to-separation ratio of the collimator grids is 112, corresponding to a full width at half maximum (FWHM) of the image of a point source of 15′ (ref. 1). The spin axis was manoeuvred to point towards a number of sources in succession in the Centaurus region, including the important binary source Cen X-3, from December 17, 1974 to January 31, 1975, to carry out position and spectrum determinations. Soon after arriving at the first pointing direction in the sequence, a correlation map produced by overlaying data from orbits obtained on December 17–19 showed the presence of a new source not listed in the Uhuru (3U) catalogue, which we have designated A1118-61 from the position given below. The field of view of the experiment is 17° FWHM, so that the light curve could be followed over an interval of 46 d. The light curve is given in Fig. 1 in units of counting rate in our detector, which can be converted approximately to Uhuru counts by multiplication by 8. In the initial and final periods of observation, the source was rather weak, and could be detected at a satisfactory significance only by overlaying orbits, with a consequent reduction in time resolution. The latter improved to one orbit (101 min) when the source was near its maximum brightness. We have, for clarity, averaged the intensity over 12 h in this period, but significant intensity variations, exceeding those attributable to counting statistics, were observed in the one-orbit averages. Where an upper limit only is given, it corresponds to three standard deviations of the background level.