The COS FUV detector is a windowless cross delay line (XDL) device with two segments (FUVA and FUVB). Each segment has a dark current that
is correlated to the Solar Cycle. The dark rate for both FUVA and FUVB is monitored closely. The scatter in the dark rate has been decreasing since
the solar maximum in 2014/2015, as shown in:
http://www.stsci.edu/~COS/fuv_darks/dark_vs_time_FUVA.png
http://www.stsci.edu/~COS/fuv_darks/dark_vs_time_FUVB.png
The ETC value for the COS/FUV detector dark rate is 4.0x10-6 counts/s/pixel for both segments. This was based on a conservative estimate of the FUVA and FUVB dark rate
at the time of the last ETC update in the summer of 2015. However, the dark rate on both detectors has been evolving with time and is now more than a factor of two lower, with
a 95% confidence upper limit of 2.8x10-6 (FUVA) and 2.6x10-6 (FUVB) counts/s/pixel. As of April 2016 we recommend using a value of 2.0x10-6 counts/s/pixel
for FUVA and FUVB. In the majority of COS programs, the source count rate will be much larger than the dark current, and the S/N will not be strongly affected by these effects.
However, proposers considering COS/FUV observations of faint, background-limited targets may want to adjust their ETC calculations to use the lower background rate.
Furthermore, users planning background-limited G140L observations of faint sources can also reduce the size of the COS FUV extraction box from 57 pixels (the default assumed by the ETC) to 40 pixels, when calculating the background levels. Analysis of the G140L cross-dispersion profiles has shown that 40 pixels is adequate to extract the source.
For faint COS targets, the combination of the smaller extraction box and reduced dark rate can have substantial impact on the exposure time needed to reach a given S/N ratio. An updated exposure time tnew can be calculated from the current exposure time told reported by the ETC using this equation:
- tnew/told = (S_ETC+0.35D_ETC)/(S_ETC+D_ETC),
- S/Nnew/S/Nold = √(S_ETC+D_ETC)/(S_ETC+0.35D_ETC).
For convenience we provide some examples of the correction factors. For cases where the source rate per pixel equals 1 times the background rate per pixel, tnew/told=0.68 and in a given exposure time, S/Nnew/S/Nold=1.22.
S_ETC/D_ETC | tnew/told | S/Nnew/S/Nold |
---|---|---|
1 | 0.68 | 1.22 |
2 | 0.78 | 1.12 |
3 | 0.84 | 1.09 |
Proposers are warned that STScI cannot guarantee that the dark rate will be lower than the current ETC value when their observations execute due to a large orbital and time variability. If the impact to the number of requested orbits is minimal, PIs are advised to use the conservative dark rate currently implemented in the ETC. Assuming values lower than 4.0x10-6 cnts/s/pixel for FUVA and FUVB when calculating exposure times for Phase I proposals will be at the observer's own risk.
The Cycle 24 Phase II ETC will be updated with segment-dependent dark rates that are appropriate at the time, using the most recent monitoring data.