Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/hst/wfc3/documents/ISRs/WFC3-2008-11.pdf Дата изменения: Sat May 31 01:34:41 2008 Дата индексирования: Sat Sep 6 20:49:35 2008 Кодировка:

Instrument Science Report WFC3 2008-11

WFC3 ETC Update
Thomas M. Brown May 29, 2008 ABSTRACT The WFC3 ETCs have been updated to reflect our preliminary findings from the most recent campaign of thermal vacuum testing. In general, the new ETC assumes both channels perform better than previously thought. Note that the ETCs are subject to further changes as we continue to analyze data from ground and flight tests.

Background
The Wide Field Camera 3 (WFC3) recently underwent ground testing in thermal vacuum (TV) conditions. A new IR detector (IR4; FPA165) is currently installed in the instrument, replacing the detector used during the 2007 TV tests (IR1; FPA129), which in turn replaced the detector used in the 2004 TV tests (IR2; FPA064). The flight UVIS detector (now called UVIS-1') was restored to the instrument for the 2008 TV test, replacing the spare UVIS detector that was installed during the 2007 TV tests (UVIS-2). In 2004, UVIS-1 was in the instrument, but that detector was reworked between the 2004 and 2008 TV tests, resulting in the new designation of UVIS-1'. Although the analysis of data from the 2008 TV campaign is still underway, we have preliminary results, and these have been incorporated into the WFC3 ETCs to guide observers in their Cycle 17 observation planning. The ETCs will be further updated if subsequent analysis of either ground or flight data implies significant changes from the current assumptions. This document merely summarizes the changes made to the ETC at a high level.

Changes
· Increase in IR throughput. We were already assuming the new FPA165 detector in the ETC (which has a higher QE than the previous FPA129 detector), but assumed the IR channel would show the same 2-20% throughput deficit (compared to the expectations from components) found in the 2007 TV tests. We now assume the 2008 TV test results, which found a throughput surplus of ~5% (compared to the expectations from components).

Operated by the Association of Universities for Research in Astronomy, Inc., for the National Aeronautics and Space Administration


Instrument Science Report WFC3 2008-11

· Significant increase in UV throughput, small decrease in optical throughput, little change in red throughput. Compared to the 2004 TV tests, the UVIS channel throughput shows little change in the 600-1000 nm range, a decrease of ~10% in the 250-600 nm range, and a decrease of up to 25% in the 200-250 nm range. These changes are consistent with changes seen in the calibration of the reference detectors used in the TV tests (i.e., they do not appear to be a real change in the WFC3 UVIS channel). Investigation of the reference detector calibration is underway, but for now we assume the new throughput values. However, we also found that the quantum yield correction in the UV is much smaller than expected, which more than compensates for the observed decrease in raw QE. The quantum yield effect causes UV photons to produce more than one election in the UV; for signal-to-noise purposes, the raw (observed) QE of the detector was previously normalized downward by a factor of /crit, where /crit is 339.6 nm. With the quantum yield correction removed from the ETC, the effective UV QE we assume is 5-40% higher in the 200-300 nm range, compared to the previous assumption. · Full well reduced. We have changed the IR channel full well from 75,000 e-/pix to 70,000 e-/ pix; this is the point where the count rate of a bright source becomes nonlinear at the ~1% level. The IR channel reaches zero bias voltage at a lower level of ~60,000 e-/pix, which can affect the apparent dark rate in a pixel, but this will be a small (<1%) effect for sources bright enough to reach this level within an IR exposure (maximum exposure time is 2800 sec). We have changed the UVIS channel full well from 75,000 e-/pix to 70,000 e-/pix. · No change in read noise. We still assume 14.6 e-/pix/multiread on the IR channel (effective read noise fitting up a 16-read ramp) and 3.0 e-/pix/read on the UVIS channel. · No change in dark current. We still assume 0.0221 e-/pix/sec on the IR channel and 0.0005 e-/ pix/sec on the UVIS channel. Note that the IR channel has an additional thermal background of 0.05 e-/pix/sec to 0.12 e-/sec/pix, depending upon the filter.

2