Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/hst/wfc3/documents/ISRs/WFC3-2008-16.pdf Дата изменения: Thu Jul 17 00:42:50 2008 Дата индексирования: Sat Sep 6 20:43:33 2008 Кодировка: Поисковые слова: coma

ST-ECF Instrument Science Report WFC3-2008-16

The TV3 ground calibrations of the WFC3 NIR grisms
H. Kuntschner, H. Bushouse, J. R. Walsh, M. KЭmmel July 10, 2008

ABS

TRACT

Based on thermal vacuum tests (TV3; March/April 2008), the performance of the WFC3 near-IR G102 and G141 grisms together with the new IR detector (IR4; FPA165) has been assessed. In this ISR we report on the new throughput measurements for both grisms and show that the current trace and dispersion solutions are consistent with the ones established during TV2. Furthermore, a new flat-field cube is determined.

1. Introduction
The Wide Field Camera 3 (WFC3) recently underwent the third ground testing in thermal vacuum (TV) conditions. A new IR detector (IR4; FPA165) is now installed in the instrument, re-placing the detector used during the 2007 TV tests (IR1; FPA129). In this ISR we determine new throughput measurements for the NIR channel with two grisms, G102 and G141, for the shorter (800 - 1150nm) and longer NIR wavelengths (1100-1700nm), respectively. This instrument science report (ISR) is a follow up on the report by Kuntschner, Bushouse, Walsh & KЭmmel (WFC3-2008-15) of the TV2 calibrations. The calibration set-up and instrument configuration was similar to the one used in TV2. However, only a restricted set of calibrations and wavelength steps was carried out to establish primarily the new throughput measurements.
The Sp ace Telescope European Coordinating Facility. All Rights Reserved.


ST-ECF Instrument Science Report WFC3-2008-16

2. Calibration measurements
For each IR grism calibration measurements were carried out to allow determination of the spectral trace, the dispersion solution and the absolute instrument throughput as well as a wavelength dependent flat-field. A summary of all NIR grism calibration procedures is given in Table 1. In TV3 absolute throughput, trace and dispersion calibrations are measured only for a central position in the IR channel field of view. In TV2 the trace/dispersion solutions were carried out in a 5-point pattern covering the FoV (Kuntschner et al. 2008). Table 1. Summary of NIR grism calibrations during TV3 Test procedure IR14S01 IR14S02 IR14S03 IR14S04 IR16S10 IR16S60 Date 20082008200820082008200803 03 03 03 04 04 14/17/18 14/17 16 16 06 06 G G G G G G G rism 102 141 102 141 102 141 Purpose Flat field Flat field Absolute throughput Absolute throughput Trace/Dispersion (center) Trace/Dispersion (center)

The absolute throughput measurements were carried out with a calibrated monochromator source covering for the G102 grism the range 780nm - 1180nm in steps of 20nm; and for the G141 grism the range 1040nm ­ 1700nm in steps of 20nm. For the G102 the trace and dispersion calibrations were carried out starting with a set of three direct image - grism pairs using a white light stimulus. After that monochromator steps covering the range 760nm - 1180nm in steps of 20nm were performed. For the G141 grism, only one pair of direct image and grism image with white light stimulus was carried out at the beginning of the sequence. Furthermore a reduced number of monochromator wavelength settings in the range 1100nm ­ 1590nm in steps of 70nm was carried out. In all cases, the monochromator bandwidth was 10 nm. Additionally, the detector flat-field was determined with 10nm wide bandpasses in steps of 20nm, covering for the G102 grism the range 800nm - 1180nm; and for the G141 grism the range 1060nm ­ 1700nm. For each grism there are two sets of flatfield calibration available.

2


ST-ECF Instrument Science Report WFC3-2008-16

3. Analysis
In this section we describe the analysis of the TV3 calibration data yielding the confirmation of existing trace and wavelength calibrations and presenting updated throughput measurements and flat-field cubes.

3.1.

Trace calibration

The installation of the new IR detector will have a significant effect on the throughput measurements (see Section 3.3), however, it is expected that the main trace and dispersion characteristics, which are determined by the unchanged grisms, remain stable. Hence only one central pointing was observed for the trace and wavelength calibrations during TV3. The trace and wavelength calibration for science data will need to be carried out in orbit with astronomical standard source s (such as flux standard stars and PNe); here in this report we only check for consistency of the solutions derived from TV2 data with the TV3 observations. The analysis was carried out in the same manner as described for TV2 (see Kuntschner et al. 2008) and no differences were found in terms of trace quality and linearity of the traces. The trace definitions are of the form (Y-Yref)=DYDX_0 + DYDX_1*(X-Xref). In Table 2 we compare the values derived from TV3 data with the predictions based on the field dependent TV2 solutions. We find good agreement between the field dependent solution derived from TV2 calibrations and the central measurement taken during TV3. Specifically, the zero points of the linear trace agree to better than 0.1 pixel and the slope agrees to 0.2 pixel over 100 pixel in x. We conclude that the 1st order trace determined during TV3 for one central position is consistent with the field-dependent solution derived from TV2 calibrations. This confirms that the installation and alignment of the new IR detector within the WFC3 optical train has not resulted in any significant change with respect to the geometrical alignment with the grisms.

3


ST-ECF Instrument Science Report WFC3-2008-16 Table 2: Comparison of 1st order linear trace solutions between TV2 and TV3 for one central position Xref [pix] G102: TV3 TV2 prediction G141: TV3 TV2 prediction 348.74 552.57 0.85 0.95 0.009 0.008 349.52 462.23 -1.27 -1.25 0.013 0.011 Yref [pix] DYDX_0 DYDX_1

3.2.

Wavelength solutions
from the TV2 analysis were used to extract the he standard aXe task AXECORE. The FITS files analyzed by custom built IDL scripts where the monochromator spot was measured by fitting a

The trace definitions derived monochromator spectra with t produced by aXe were then location of the peak of each Gaussian.

The wavelength solutions were found to be well approximated by linear fits to wavelength versus pixel offset. In Table 3 we compare the values derived from TV3 data with the predictions based on the field dependent TV2 solutions. We find good agreement between the field dependent solution derived from TV2 calibrations and the central measurement taken during TV3. Specifically, the wavelength zero points agree to within 0.25 pixel (or 6 е) and the dispersion is consistent to within 0.1 е. We conclude that the 1st order wavelength solution determined during TV3 for one central position is consistent with the field-dependent solution derived from TV2 calibrations.

4


ST-ECF Instrument Science Report WFC3-2008-16 Table 3: Comparison of 1st order wavelength solutions between TV2 and TV3 for one central position Xref [pix] G102: TV3 TV2 prediction G141: TV3 TV2 prediction 348.74 552.57 8950.4 8954.0 46.70 46.80 349.52 462.23 6371.9 6377.6 24.52 24.56 Yref [pix] DLDP_A_0 [е] DLDP_A_1 [е/pix]

3.3.

Absolute throughput calibrations

For each of the calibrated monochromator settings (see also Section 2) and all visible orders, the detected counts were measured with the IRAF task imexam. The aperture radius was adjusted with 3 iterations, generally resulting in radii of about 17 pixel. The ratio of detected flux (using an effective gain of 2.26 e/DN1) versus the incoming flux as recorded in the image headers (keyword OSFLUX, given as flux per second) gives then the instrument efficiency. Figures 4 and 5 present the efficiency curves for WFC3 with the G102 and G141 grisms, respectively. The plots show the efficiency of WFC3 as measured in TV3 but do not include any contributions from the OTA throughput. For G102, the peak efficiency of ~50% is reached between 1020 and 1120nm in the +1st order. Efficiency above 10% is achieved over a broad wavelength range from 800 to 1140nm. Due to the excellent design of the grism, the throughput in the +2nd and zeroth order is much lower with a maximum of 7% and 2%, respectively. The mean, relative increase in efficiency of the first order between TV3 and TV2 measurements is ~60%. The relative efficiency increase ranges from 67% in the blue to 52% in the red for G102, which is due to the increased QE of the new IR detector.
1

The effective gain is defined here as the product of the traditional gain value of 2.60 e/DN and the Intra-Capacitance (IPC) correction of 0.87. 5


ST-ECF Instrument Science Report WFC3-2008-16 For G141, the peak efficiency of ~55% is reached between 1400 and 1600nm in the +1st order. Efficiency above 10% is achieved over a broad wavelength range from 1080 to 1680nm, thus giving continuous wavelength coverage from 800 to 1680nm with the two grisms together. Due to the excellent design of the G141 grism, the throughput in the +2nd, 3rd and zeroth order is much low with a maximum of 10, 1 and 2%, respectively. The mean, relative increase in efficiency of the first order between TV3 and TV2 measurements is ~45%. The relative efficiency increase ranges from 70% in the blue to 21% in the red for G141. Tables 6 and 7 present the efficiency measurements in tabular form for the G102 and G141 grisms, respectively.

Figure 1: Instrument efficiency of various orders for the WFC3 G102 grism as function of wavelength. For comparison the grey line shows the efficiency of the first order derived during TV2 with the previous NIR detector.

6


ST-ECF Instrument Science Report WFC3-2008-16

Figure 2: Instrument efficiency of various orders for the WFC3 G141 grism as function of wavelength. The wavelength coverage is interrupted for 0th order on account of insufficient signal. For comparison the grey line shows the efficiency of the first order derived during TV2 with the previous NIR detector.

7


ST-ECF Instrument Science Report WFC3-2008-16

Table 3: Instrument efficiency for WFC3 G102 (based on TV3)
Wav elengt h (nm) 7800 8000 8200 8400 8600 8800 9000 9200 9400 9600 9800 10000 10200 10400 10600 10800 11000 11200 11400 11600 11800 0th 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. ord er 0076 0189 0248 0241 0238 0188 0218 0189 0122 0130 0092 1st 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. order 0537 1539 2029 2738 3078 3405 3801 4123 4341 4506 4709 4625 4941 5145 5022 4976 5067 5008 4318 0598 0011 2nd order 0.0291 0.0641 0.0707 0.0677 0.0579 0.0568 0.0437 0.0400 0.0286 0.0231 0.0199 -

8


ST-ECF Instrument Science Report WFC3-2008-16

Table 4: Instrument efficiency for WFC3 G141 (based on TV3)
Wav elengt h (nm) 10400 10600 10800 11000 11200 11400 11800 12200 12600 13000 13400 13800 14200 14600 15000 15400 15800 16200 16400 16600 16800 17000 0th 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. order 0007 0027 0099 0211 0225 0222 0173 0165 0119 0056 0031 0053 0152 0114 0096 0168 0116 0021 1st 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. order 0073 0312 1352 2872 3844 4082 4454 4871 4962 5335 5470 5487 5569 5556 5598 5502 5525 5255 5118 4853 2102 0617 2nd 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. order 0041 0144 0504 0881 0958 0844 0625 0458 0297 0237 0121 3rd 0. 0. 0. 0. 0. 0. ord er 0003 0012 0039 0132 0085 0101 -

0. 0. 0. 0. 0. 0. 0.

3.4.

Flat-field determinations
as he of to of

An identical procedure for producing monochromatic flat-fields was followed for TV2 (see Kuntschner et al. 2008). Using CASTLE and the monochromator t G102 and G141 grisms were illuminated with light over the wavelength range each grism: for G102 8000 to 11800е in steps of 200е; for G141 10600 17000е, also in steps of 200е. The monochromator slit was set to produce light wavelength width 100е.

Figure 3 shows the G102 image at 10000 е. The illumination pattern was removed by fitting a surface using the IRAF task imsurfit as for TV2. Figure 4 shows the resulting normalized flat-field for G102 at 10000е. There is a circular dark region (~40 pixel diameter) at the bottom of the detector, which corresponds to a region of dead pixels. A further notable feature of reduced detector sensitivity is placed towards the lower right of the detector. Only the G102 flat is shown here; the G141 9


ST-ECF Instrument Science Report WFC3-2008-16 flat-field images are very similar in appearance. The sets of flat-fields were fitted pixel-by-pixel with wavelength as for TV2, and the residual images on the fit showed mean values of 0.8 and 0.6% for the G102 and G141 grisms, respectively. Figure 5 shows examples for a few selected pixels. Two pixels lying at the mean flat-field response of 1.0 are show in blue and green, as representative of the majority of the pixels. The red points show the wavelength dependence for a 'high' pixel and the magenta points for a 'low' pixel. The open circles show the points generated from the polynomial fit.

Figure 3: Example of the monochromatic image at 10000е taken with the G102 grism. The mean level is 12330 ADU and the greyscale range ±20% about the mean (black - white).

10


ST-ECF Instrument Science Report WFC3-2008-16

Figure 4: The resulting normalised flat-field image for the G102 grism at 10000е, derived from the image shown in Figure 3. The greyscale is ±20% about the mean.

11


ST-ECF Instrument Science Report WFC3-2008-16

Figure 5: The selected pixels. plot shows the coordinates of (magenta) and is stored in the

variation of the flat-field as a function of wavelength for a few Each point refers to the mean value over nine pixels. The upper behaviour for the G102 grism and the lower plot for G141. The the pixels are indicated. The open circles (for the low pixel high pixel (red) values) are derived from the polynomial fit, which flat-field cube.

12


ST-ECF Instrument Science Report WFC3-2008-16

4. Conclusions
This ISR presented the WFC3 IR grism calibrations carried out during thermal vacuum 3 (TV3; March/April 2008) with the new IR detector (IR4; FPA165). The trace and dispersion solutions derived at one central field position are found to be consistent with the field-dependent solutions derived during TV2. Updated efficiency curves are presented for the G102 and G141 grism modes. Peak efficiency of WFC3 with the G102 and G141 grisms reaches now ~50% and ~55%, respectively. The resulting sensitivity files are used in a simulation package (aXeSIM; KЭmmel, Kuntschner, Walsh, 2007) for HST/WFC3 grism observations and are available from ST-ECF Web pages (http://www.stecf.org). Furthermore, we derive up-dated flat-field cubes that provide pixel-to-pixel information as function of wavelength to an accuracy of about 1%.

References
KЭmmel, Kuntschner, Walsh, 2007, ST-ECF Newsletter, 43, 8 Kuntschner, Bushouse, Walsh, KЭmmel, 2008, ST-ECF Instrument Science Report, WFC3-2008-15

13