TITLE: GCN GRB OBSERVATION REPORT NUMBER: 2568 SUBJECT: INTEGRAL-SPI/ACS GRB data: timing correction DATE: 04/04/14 11:04:09 GMT FROM: Arne Rau at MPI A. Rau, A. von Kienlin, G. Lichti on behalf of the INTEGRAL SPI-ACS GRB team (Garching); K. Hurley on behalf of the IPN team; M. Beck on behalf of the INTEGRAL Science Data Center (ISDC) Geneva; report: The Anti-Coincidence Shield (ACS) of the Spectrometer onboard INTEGRAL (SPI) has provided more than one hundred GRB triggers for the 3rd Interplanetary Network (IPN) over the last 16 months. By studying IPN triangulations involving the ACS and one or more distant spacecraft, such as Ulysses and Mars Odyssey, it was determined that the absolute timing uncertainty of the ACS lightcurves was of the order of 100 ms at most. These triangulations involved bursts with known positions, such as SGRs and GRBs with long-wavelength counterparts. Accordingly, IPN triangulations up to now involving the ACS and distant spacecraft have included a conservative estimate of the systematic uncertainty of this order. Recently, for cases involving precisely known IPN positions, an additional triangulation of some of these bursts using Konus-Wind and/or Helicon-Coronas-F data in conjunction with ACS INTEGRAL revealed a systematic inaccuracy in the ACS timing such that the ACS was lagging behind UTC by 134 +- 12 ms (S. Golenetskii, E. Mazets, V. Pal'shin and D. Frederiks, private communication). Further investigations by the INTEGRAL SPI/ACS Team confirmed this timing error by comparing the rise times of (most likely) particle-induced short events in the ACS and corresponding saturation signatures in the Ge-detectors of SPI. Using this method, the error was found to be 125 +/- 10 ms. The cause of this error is still unresolved. Starting with bursts which occurred after April 2 2004 11:00 UTC, a correction of 125 ms has been added to the publicly available ACS light curve files (http://isdcarc.unige.ch/arc/FTP/ibas/spiacs/). Bursts which occurred before this date remain uncorrected at this time, but it is emphasized that a timing correction of ~125 ms should be added to all data files since the start of the mission. Application of this correction to IPN bursts with known positions yields better agreement with the triangulated positions. Accordingly, the value of the ACS systematic uncertainty will now be reduced. This message is citable.