COS ETC 21.1 Cycle 21 Warnings
Exposure Time and S/N calculated by the ETC for the G130M/1055 and G130M/1096 Settings are not correct
The focus of the G130M/1055 and G130M/1096 settings were recently adjusted to increase their resolution. As a consequence, the spectral height and size of the resolution element (resel) have changed. Because of these two changes, exposure times may be underestimated by as much as a factor of 4, and S/N may be overestimated by as much as a factor of 2. The ETC released in December 2012 (21.1) assumes a resel of 30 pixels and an extraction height of 47 pixels, but the true resels are smaller and depend on wavelength; the true extraction heights are larger. The ETC will be updated for phase II of Cycle 21, in the meantime the correction factors given in the Known Issues page of the ETC should be used. For more detailed information on how to determine precise corrections, users can refer to ETC Corrections for resel size and extraction height.
Changes in the FUV sensitivity and new throughput curves<1200 Å
A new version of the ETC (20.1.1) was released on January 26, 2012. This version contains COS-only related updates and affects proposers using COS for Cycle 20 proposals:
Throughputs of all FUV gratings have been updated to take into account that the sensitivity of the FUV channel is currently decreasing at substantially higher rates than before. The new throughputs are projected to January 1, 2013.
These new projections may increase the required time to reach a certain S/N by as much as 50%. Users are strongly recommended to repeat any calculations performed in preparation for the Cycle 20 call for proposals with the new version of the ETC.
The throughput of the G285M/NUV grating was also updated at this time. While there has been no change in the slope of the sensitivity decline, which is ~11% per year, this throughput had not been updated in the last two years, leading to exposure times that would be underestimated by ~25% by the middle of Cycle 20.
Users proposing for Cycle 20 with the G285M grating should also update their calculations by using the new version of the ETC.
A software bug in the ETC algorithm that simulates dispersed light target acquisitions (for both FUV and NUV) r esulted in overestimated exposure times for a certain S/N, by as much as a factor of 2. Users proposing for Cycle 20 are strongly encouraged to redo their calculations with the new version of the ETC.
Resolution of New COS/FUV G130M Modes<1200 Å
The two new COS/FUV G130M modes (cenwave=1055 and cenwave=1096) have resolution of R ~ 3000 at 950 A, decreasing to R ~ 1000 at 1150 A and dispersion of ~ 0.01 A/pixel, similar to the dispersion of the other, higher resolution, modes. This implies that at 950 A one resolution element is around 30 pixels. Since the ETC uses a resolution element that is 5 times smaller (6 pix), it will overestimate the time required to achieve a certain S/N per 30 pixel resolution element, while underestimating the S/N per 30 pixel resolution element for a given time. To calculate the correct exposure time per real 30 pix resolution element, divide the exposure time returned by the ETC by 5 (= 30/6). To calculate the correct S/N per real 30 pix resolution element, multiply the S/N returned by the ETC by 2.2 (= 5^0.5).
The two new COS/FUV G130M modes (cenwave=1055 and cenwave=1096) have resolution of R ~ 3000 at 950 A, decreasing to R ~ 1000 at 1150 A and dispersion of ~ 0.01 A/pixel, similar to the dispersion of the other, higher resolution, modes. This implies that at 950 A one resolution element is around 30 pixels. Since the ETC uses a resolution element that is 5 times smaller (6 pix), it will overestimate the time required to achieve a certain S/N per 30 pixel resolution element, while underestimating the S/N per 30 pixel resolution element for a given time. To calculate the correct exposure time per real 30 pix resolution element, divide the exposure time returned by the ETC by 5 (= 30/6). To calculate the correct S/N per real 30 pix resolution element, multiply the S/N returned by the ETC by 2.2 (= 5^0.5).
COS BOA Calculations for Wavelengths <1200 Å
The transmission of the BOA aperture at the short wavelengths seen by the G140L 1105 and 1280 cenwaves, while not fully characterized, is expected to be close to zero (due to MgF2 cutoff for wavelengths shorter than 1200 A). The values reported by the Spectroscopic COS ETC, for calculations using the BOA at wavelengths shorter than 1200 A, are not accurate and should not be used for planning any COS observations.
COS Web Form Bug
Version 18 of the ETC, which was installed in the Fall of 2009, contained a bug in the web forms for COS imaging ETC that affected calculations performed using a point source with a size other than the default. A fix to this problem was installed on 2/17/2010. Users who think their previous ETC calculations may have been affected by this problem should repeat their calculations.
The problem corrected by this new version of the ETC occurred when the "Point Source" option is selected in section 2 of the ETC input form and "circular region" is selected via the pull down menu option, the calculation is performed using a circular region of radius 0.4 arcsec, regardless of the value selected in the pull down menu.
ETC Calculations Using COS Spectroscopic TA with G285M/2739
Using this mode in the ETC will cause a JavaNullPointerException error. Although not available in the ETC, there is a workaround for this problem. If you need to use this mode, please contact the helpdesk help@stsci.edu.
Warning regarding G140L cenwaves in ETC vs APT
Cycle 18 users are advised that the "1250" G140L cenwave in ETC 18.0.5 is identical and refers to the same mode as the cenwave called "1280" in APT and the Cycle 18 COS Instrument Handbook. GOs preparing Cycle 18 phase II should do ETC calculations with G140L/1250, but use G140L/1280 setting in their APT phase II. Cycle 17 users should NOT change and re-submit their phase IIs in relation to this change in cenwave name. All Cycle 17 phase IIs should use the G140L/1230 nomenclature instead.