A Comparison of Continuous Clocking 1x3 Mode to Timed Exposure Mode E ciencies of ACIS

In this document we compare event detection e ciencies in the TE mode to the ones expected in the CC(1x3) mode in the ACIS-S detector array of the HETG spectrometer on board the Chandra X-ray Observatory. Data taken in TE mode have pulse heights determined from the charge in a 3x3 pixel event island. In CC(1x3) mode, events are read out on a row by row basis, and therefore the pulse height for an event is determined instead from a 1x3 pixel event island. While the QE of ACIS in TE mode has been well characterized, that of CC(1x3) mode remained quite unknown. Therefore, we investigate the changes in QE when using the CC(1x3) mode compared to the TE mode. We used data taken at XRCF in TE mode, and split each 3x3 event island into three 1x3 event islands.The main goal of the analysis is to develop guidelines for a proper order extraction for HETG CC(1x3) mode data. Our analysis shows a 5% drop in QE in CC(1x3) mode at the O K line at 525 eV for FI devices, and a 20% drop for the BI devices. The maximumdrop in QE depends on the size of the extraction bandwidth and is 55% around 3 keV for BI and around 50% at 8 keV for FI devices. The number of additional events created during the 3x3 to 1x3 island transformation can be up to 35% throughout the entire energy band. However, when the match of CC(1x3) mode QE is relaxed to 95% of the TE mode QE, the amount of contamination from additional events is greatly reduced.