Here we report on Swift and Suzaku observations near the end of an outburst from the black hole transient 4U 1630–47 and Chandra observations when the source was in quiescence .
4U 1630–47 made a transition from a soft state to the hard state \sim 50 d after the main outburst ended .
During this unusual delay , the flux continued to drop , and one Swift measurement found the source with a soft spectrum at a 2–10 keV luminosity of L = 1.07 \times 10 ^ { 35 } erg s ^ { -1 } for an estimated distance of 10 kpc .
While such transients usually make a transition to the hard state at L / L _ { Edd } = 0.3 –3 % , where L _ { Edd } is the Eddington luminosity , the 4U 1630–47 spectrum remained soft at L / L _ { Edd } = 0.008 M _ { 10 } ^ { -1 } % ( as measured in the 2–10 keV band ) , where M _ { 10 } is the mass of the black hole in units of 10 M _ { \mathord { \odot } } .
An estimate of the luminosity in the broader 0.5–200 keV bandpass gives L / L _ { Edd } = 0.03 M _ { 10 } ^ { -1 } % , which is still an order of magnitude lower than typical .
We also measured an exponential decay of the X-ray flux in the hard state with an e-folding time of 3.39 \pm 0.06 d , which is much less than previous measurements of 12–15 d during decays by 4U 1630–47 in the soft state .
With the \sim 100 ks Suzaku observation , we do not see evidence for a reflection component , and the 90 % confidence limits on the equivalent width of a narrow iron K \alpha emission line are < 40 eV for a narrow line and < 100 eV for a line of any width , which is consistent with a change of geometry ( either a truncated accretion disk or a change in the location of the hard X-ray source ) in the hard state .
Finally , we report a 0.5–8 keV luminosity upper limit of < 2 \times 10 ^ { 32 } erg s ^ { -1 } in quiescence , which is the lowest value measured for 4U 1630–47 to date .