(ABRIDGED) We present in detail our new 3D numerical models for the accretion
of stellar winds on to Sgr A*. In our most sophisticated models, we put stars
on realistic orbits around Sgr A*, include `slow' winds (300 km/s), and account
for radiative cooling. We first model only one phase `fast' stellar winds (1000
km/s). For wind sources fixed in space, the accretion rate is Mdot ~ 1e-5
Msun/yr, fluctuates by < 10%, and is in a good agreement with previous models.
In contrast, Mdot decreases by an order of magnitude for stars following
circular orbits, and fluctuates by ~ 50%. Then we allow a fraction of stars to
produce slow winds. Much of these winds cool radiatively, forming cold clumps
immersed into the X-ray emitting gas. We test two orbital configurations for
the stars in this scenario, an isotropic distribution and two rotating discs
with perpendicular orientation. The morphology of cold gas is quite sensitive
to the orbits. In both cases, however, most of the accreted gas is hot, with an
almost constant Mdot ~ 3e-6 Msun/yr, consistent with Chandra observations. The
cold gas accretes in intermittent, short but powerful episodes which may give
rise to large amplitude variability in the luminosity of Sgr A* on time scales
of 10s to 100s of years. The circularisation radii for the flows are ~ 1e3 and
1e4 Rsch, for the one and two-phase wind simulations, respectively, never
forming the quasi-spherical accretion flows suggested in some previous work.
Our work suggests that, averaged over time scales of 100s to 1000s of years,
the radiative and mechanical luminosity of Sgr A* may be substantially higher
than it is in its current state. Further improvements of the wind accretion
modelling of Sgr A* will rely on improved observational constraints for the
wind properties and stellar orbits.Comment: 16 pages, 18 colour figures. Accepted by MNRAS. Full resolution paper
and movies available at http://www.mpa-garching.mpg.de/~jcuadra/Winds/ . (v2:
minor changes