GRB 090904***A***, Swift-BAT refined analysis S. D. Barthelmy (GSFC), W. H. Baumgartner (GSFC/UMBC), J. R. Cummings (GSFC/UMBC), N. Gehrels (GSFC), H. A. Krimm (GSFC/USRA), A. Y. Lien (GSFC/UMBC), C. B. Markwardt (GSFC), D. M. Palmer (LANL), T. Sakamoto (AGU), M. Stamatikos (OSU), T. N. Ukwatta (LANL) (i.e. the Swift-BAT team): Using the data set from T-239 to T+723 sec from the recent telemetry downlink, we report further analysis of BAT GRB 090904***A*** (trigger #361831) (NAME, et al., GCN Circ. xxxx). The BAT ground-calculated position is RA, Dec = 264.195, -25.215 deg which is RA(J2000) = 17h 36m 46.8s Dec(J2000) = -25d 12' 55.2" with an uncertainty of 1.2 arcmin, (radius, sys+stat, 90% containment). The partial coding was 8%. [Description about the mask-weighted light curve] T90 (15-350 keV) is 64.0 +- 16.0 sec (estimated error including systematics). The time-averaged spectrum from T-15.2 to T+64.8 sec is best fit by a simple power-law model. The power law index of the time-averaged spectrum is 1.56 +- 0.08. The fluence in the 15-150 keV band is 1.3 +- 0.1 x 10^-05 erg/cm2. The 1-sec peak photon flux measured from T+8.33 sec in the 15-150 keV band is 2.7 +- 0.8 ph/cm2/sec. All the quoted errors are at the 90% confidence level. The results of the batgrbproduct analysis are available at http://gcn.gsfc.nasa.gov/notices_s/361831/BA/