In 2000 August, we observed the black hole candidate 1E 1740.7-2942, the brightest persistent hard X-ray source within a few degrees of the Galactic Centre, for 10 ks with Chandra (ACIS-I). Attempting to compensate for pile-up effects, we found that the spectra were well-fitted by an absorbed power law, with photon indices F = 1.54(-0.37)(+0.42) (readout streak) and Gamma = 1.42(-0.14)(+0.14) (annulus), consistent with a black hole low/hard state. We have analysed a public observation performed by Chandra which utilized short frames in order to avoid severe pile-up effects. Subtracting the core point spread function from the whole image, we did not find evidence for any elongated feature perpendicular to the radio jet axis, as reported in a recent analysis of the same data. Moreover, comparing the radial profiles with those of an unscattered X-ray point source, we found an indication of an extended, previously undetected, X-ray scattering halo. The measured halo fractional intensity at 3 keV is between 30 and 40 per cent within 40 arcsec but drops below detectable levels at 5 keV Finally, by placing a limit on the X-ray flux from the radio-emitting lobe, which has been identified as the hotspot at the end of the northern jet of 1E 1740.7-2942, we are able to constrain the magnetic energy density in that region.