TITLE: GCN CIRCULAR NUMBER: 13692 SUBJECT: GRB 120817B: BAT location in ground analysis and refined analysis DATE: 12/08/20 14:35:48 GMT FROM: Jay R. Cummings at NASA/GSFC/Swift J. R. Cummings (GSFC/UMBC), S. D. Barthelmy (GSFC), W. H. Baumgartner (GSFC/UMBC), E. E. Fenimore (LANL), N. Gehrels (GSFC), H. A. Krimm (GSFC/USRA), C. B. Markwardt (GSFC), D. M. Palmer (LANL), T. Sakamoto (GSFC/UMBC), G. Sato (ISAS), M. Stamatikos (OSU), J. Tueller (GSFC), T. N. Ukwatta (MSU) (i.e. the Swift-BAT team): At 04:02:29 on August 17, 2012, Swift-BAT triggered (BAT trigger # 531259) on the IPN burst GRB 120817B (Golenetskii et al., GCN Circ # 13670). No source was found onboard. In ground analysis of the failed-trigger event sample, a significant source was found at RA, Dec 8.310, -26.428, which is RA (J2000) 00h 33m 14.4s Dec (J2000) -26d 25m 42s The estimated 90% confidence position error is 2.8 arcmin radius. This position was 5.5% coded in the BAT FOV. It is 5 arcmin from the center of the IPN triangulation, within the IPN error box. As seen in BAT, the burst was a single spike 0.10 +/- 0.02 seconds long. The time-averaged spectrum from T+0.2 to T+0.3 sec is best fit by a simple power-law model. The best-fit photon index is 0.60 +/- 0.3. The total fluence in the 15-150 keV band was (1.7 +/- 0.4) x 10^-7 ergs/cm^2. The errors quoted are at the 68% confidence level.