Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.stsci.edu/hst/acs/documents/isrs/isr0010.pdf Дата изменения: Mon Jan 8 19:24:35 2007 Дата индексирования: Sat Dec 22 23:21:13 2007 Кодировка: Поисковые слова: storm

Instrument Science Report ACS 00-10

FLATS: PRELIMINARY WFC DATA AND PLANS FOR FLIGHT FLATS
R. C. Bohlin, G. Hartig, and Z. Tsvetanov 2000 October

A

BSTRACT

The ACS WFC pixel-to-pixel P-flats with the build-3 detector repeat over a three month time span to a precision of better than 0.1% rms. The shape of the low frequency Lflat field varies with wavelength, where the ratio of the F435W/F814W L-flats differs from unity by ­10% in two patches near opposite corners of the field of view. At ~0.5%, the high frequency pixel-to-pixel structure in the WFC P-flats is about half that of the HRC build-2 and does not increase as much toward shorter wavelengths as the HRC build-2 with the Lesser coating. The amplitude of the fringes for continuum illumination is less than 1%, even in the narrow band ramp filters, while the fringing in monochromatic light is ­10% at 9300е. The goals of the laboratory WFC and HRC flat field program are to obtain the set of external flats that fully populates the pipeline database and to archive a baseline of high S/N internal flats for tracking post-launch changes.

1. INTRODUCTION
The WFC was externally illuminated by light from the Refractive Aberrated Simulator/Hubble Opto-Mechanical Simulator (RAS/HOMS), which accurately simulates the full field OTA illumination at 6328е. The refractive aberrations at other wavelengths should not affect the fidelity of flat fields. Several flat field exposures were obtained with the RAS/HOMS and the non-flight detector assembly B3 in 2000 June, as tabulated in Table 1. A translucent mylar diffuser screen was added to the RAS/HOMS at the pupil


Instrument Science Report ACS 00-10 plane. This source of external illumination should produce the same flat field as the sky through the OTA+ACS, except for the dominant OTA geometric distortion over the field, because only the post-pupil plane RAS optics are used in this configuration. This geometric distortion factor, G (Sparks et al. 2000) or M (Bohlin, et al. 1999, i.e. Paper I), must be incorporated before delivery of Mylar flats to the reference file database. A complete flat field for use in pipeline data reduction is the product of the low frequency variations, L, times the high frequency, pixel-to-pixel variations, P, where the geometric factor G is also included implicitly. Complete LP-flat fields are obtained merely by dividing the images of the flat field illumination by their average over some region of the detector, such as the central portion of the useful region. In the case of the preliminary HRC data in Paper I, flat field illumination was NOT obtained because of spurious reflections from the shiny edges of the non-flight detector mask; and only P-flats could be derived. For filters that fully illuminate both WFC chips, the lab illumination is proper; and the normalization for the full field flats is defined by the average of rows 1843:2039 near the top of the lower CCD chip 1, adjacent to the bottom of chip 2. This normalization process preserves any overall sensitivity differences between the two CCD chips across the ~40 pixel gap between the two independent pieces of the WFC detector. The relative response across the gap is wavelength dependent. Chip 2 is ~1.5% more sensitive in F435W, while chip 2 is ~8% less sensitive at 9300е. For display of the Figures for this presentation only, the chip 2 LP flats are renormalized to unity at the bottom of chip 2, rather than to chip 1.

2. FULL FIELD WFC L-FLATS
Flat field exposures have been obtained for the WFC during ground testing at Ball Aerospace and at GSFC. The ACS IDT has stored these images in a database and has assigned a unique entry number to each image. In general, pairs of images are co-added with cosmic ray rejection using acs_cr.pro, which also produces statistical uncertainty and data quality arrays. The results presented here are preliminary, because the WFC B3 detector assembly that recorded the images in Table 1 is scheduled to be replaced before flight. Figure 1 is an F435W LP-flat-field image, as illuminated by the RAS/HOMS with a reflective Spectralon diffusing screen near the RAS source plane and a continuum light source, while Figure 2 is the corresponding flat, as illuminated by the continuum internal lamp. The external RAS/HOMS flats with entry numbers before 16791 use this Spectralon screen and must be corrected for geometric distortion of the pixel size, as quantified in Figure 8 of Paper I, in order to compare with the other undistorted illumination modes. In addition, the Spectralon flats are corrected for a left-to-right non-uniform illumination of ~14%. The mylar diffuser at the RAS/HOMS pupil plane replaced the Spectralon setup for the external observation after entry 16791. Neither the mylar illumination nor the internal ACS lamp have the Spectralon geometric distortions that reach ~17% from corner-to-corner over the 3.4x3.4 arcmin field of the WFC. After both corrections to the Spectralon

2


Instrument Science Report ACS 00-10 flats, the response to external illumination with F435W declines by more than 10% from the lower-left to upper-right corner in Figure 1, while this corner-to-corner response for the internal lamp and F435W in Figure 2 is more uniform. In F814W, the internal flats become more than 10% brighter at the upper-right vs. the lower-left corner, while the external Spectralon L-flat is more uniform. Even though the WFC L-flats are a strong function of wavelength, the same doughnut pattern of change with wavelength is manifest for the internal and two different external illumination techniques. For example, Figures 3-4 show similar large-scale patterns that are contiguous across the two chips for the F814W/F625W ratios for internal and external illumination, respectively. Furthermore, the doughnut reappears in Figure 5, which is the smoothed ratio of monochromatic images spanning a similar wavelength interval of 8500 to 7000е. The monochromatic illumination is produced with the mylar diffuser in the RAS/HOMS, while the boxcar smoothing is done twice with a size of ~4% of the field width to remove the fringing pattern. In conclusion, the cause of the wavelength dependence of the shape of the L-flats must be assigned to the CCD detector assembly with the corollary that the external light sources are probably diffuse with no intrinsic pattern on a scale smaller than the size of images.

3. COMPARISON OF WFC AND HRC P-FLATS
In order to compare the WFC LP-flats with the build-2 HRC P-flats of Paper I, a 1024x1024 subimage is extracted from the center of the WFC chip2 where the WFC LPflats are most uniform. The center 100x100 pixels of this subimage show no systematic gradient, so that the rms dispersion can be compared with the counting statistics and with the statistical results for the HRC P-flats. Figure 6 displays the external-to-internal ratio for the 1024x1024 WFC subimages in F435W with a 0.98-1.02 stretch for comparison with the same HRC Figure 3 of paper I. Figure 7 is an individual external LP-subimage, while Figure 8 is the cross wavelength ratio F625W/F435W as in Figure 4 of Paper I. There are six salient differences between the WFC and the HRC: · There is a residual flocculent structure at the <1% level in the WFC ratio of Figure 6, which is present neither in the external/internal HRC ratio nor in the WFC external/ internal ratios at longer wavelengths. This artifact might possibly be attributed to the wavelength dependent focal length of the refractive optics in the RAS/HOMS, which could partially focus any variations in the surface of the Spectralon screen. For illumination with the mylar diffuser at the RAS/HOMS pupil, this artifact should disappear. However, testing this hypothesis should be a high priority during the first attempt to obtain LP-flats with the flight detectors. If any such residual structure persists, the mylar should be rotated to see if the cause can be attributed to the diffuser itself.

3


Instrument Science Report ACS 00-10 · At f/25, the shadows from the small dust grains on the detector window are weaker than for the f/68 HRC in Paper I. There are only two weak dust mote signatures near the bottom of Figure 6. As in the HRC case, the wider angle illumination with the internal lamp reduces the contrast in the shadows of the particulate contaminants. The external subimage for the F814W LP-flat Figure 7 shows these dust motes more clearly. Vertical lines caused by the step and repeat pattern from the CCD manufacturing process appear in Figure 7 at the fractional percent level every ~70 pixels in the x direction. No such artifacts are visible at the same 0.98-1.02 stretch of HRC images. The HRC `freckles', i.e. small spots with sizes of a few pixels and wavelength dependent depths of a few percent, are caused by irregularities in the HRC CCD backside treatment but are largely absent in the WFC with its different treatment. Compare Figure 7 with Figures 1-2 of Paper I. After correction for the counting statistics, the intrinsic rms structure of the HRC P-flats increases from 0.8% in F814W to 1.6% in F435W, while the corresponding WFC rms is in the range 0.5-0.7%. The ratio F625W/F435W in Figure 8 has an rms residual of 0.4% and lacks the snow storm of white dots that appear in the corresponding HRC Figure 4 of Paper I, where the residual is 1.0%. Despite the lack of freckles, the WFC F625W/F435W ratio shows a moderate profusion of middle size white spots that are more prolific in Figure 8 than in the corresponding Paper I Figure 4. These white spots have diameters up to ~15 pixels and correspond to black spots in Figure 7. Since these blemishes cancel in the external/ internal F435W ratio of Figure 6, these spots have the same characteristics as the HRC freckles, except for their larger size and lower abundance. If an HRC without a Lesser backside treatment is flown, then the HRC small scale artifacts should be more like the WFC.

·

·

·

·

In the WFC, particulate contaminants on the filters are not observed. The filters have been cleaned in the interim, and the faster WFC optics reduces the contrast of these shadows. Like the HRC, the short term WFC repeatability is excellent, e.g. over three months, the pixel-to-pixel repeatability of the WFC flats is better than 0.1%. The ratios July/September of 2000 for internal lamp exposures with F555W, F625W, and F814W have a rms pixel-to-pixel scatter of ~0.5%, which is the same as the counting statistics to an accuracy that limits other noise sources to <~0.05%. However, the internal F555W L-flat changed by as much as 0.5%, while the F814W L-flat changed by <0.2%. These changes cannot be attributed to the bias, which contributes <0.05% of the signal. Perhaps, the change in CCD temperature from -77C in July to -67C in September effects the short wavelength L-flats. Ramp L-flats are problematic, even for the region of monochromatic transmission at the observed wavelength. Consider uniform streaks of emission parallel to the length of the ramp filters along the direction where the central wavelength changes with position. For monochromatic emission the counts along the line rise and fall with the shape of the monochromatic transmission summarized in Bohlin & Tsvetanov (2000), while the signal for continuum emission changes monotonically across the entire field of view with the slope of the overall change in instrumental sensitivity with wavelength. In order to enable

4


Instrument Science Report ACS 00-10 continuum subtraction, the pipeline flatfields for ramps are the LP-flats derived from full field continuum exposures in broadband filters (Sparks et al. 2000). Additional study is required to quantify uncertainties in this L-flat correction scheme for the ramp filters. A question raised in Paper I is whether the broad band flats are appropriate for ramp filters with monochromatic illumination at wavelengths below 7000е, where there is no fringing. Monochromatic 6330е and continuum flats were obtained with the WFC FR656N and are divided by the F625W flat from the external Spectralon illumination. After correcting for counting statistics, residuals are <0.2% for both monochromatic and continuum light in the central 100x100 pixels at the 6330е position on FR656N. Any large scale artifacts in these corrected FR656N images are <<1%, so that broad band flats are appropriate for the WFC ramp P-flats below 7000е. For HRC with its wavelength dependent fine structure and slower optics, the broadband flats will leave larger residuals for monochromatic light and maybe even for continuum illumination.

4. FRINGING
In continuum light, fringing appears only in F892N, FR931N, and FR1016N at <~1% rms. F892N is used at only one filter wheel setting and should have a fixed fringe pattern in its flat. The worse ramp is FR931N with a coherent fringe amplitude of ­1% (rms=0.7%). After correction with a worst case F814W flat, the pixel-to-pixel scatter is reduced to rms=0.4%; but the fringe amplitude remains at 1%. In the baseline pipeline data reduction, there is only one flat field per filter, so that fringing in FR931N cannot be removed because of the multitude of allowed central wavelength settings. In order to reduce the uncertainty in relative point source photometry to <1% over the field of F931N, the results of the fringe modeling program must be applied. The baseline pipeline flat for a ramp filter will be the appropriate piece of the broadband flat at the nearest central wavelength. The three pieces of the flat for each of the five ramps will be assembled into one flat field image before delivery of the reference file. In monochromatic light, the fringing pattern seen around the edges of the ratio image in Figure 5 is ubiquitous. Figure 9 shows the worst case data for a 1024x1024 subimage region of chip 2, illuminated by 9300е light with a bandpass of 20е. Fringes with ~­10% amplitude are visible everywhere in the field, since the broadband F814W filter is in place. For the same 20е bandpass, the rms fringe amplitude ranges from 1% at 7500е to 6% at 9300е. Since the fringing is coherent over the size of the PSF, uncorrected photometry of monochromatic point sources would exhibit the full fringe amplitude range of ­10% variation across the field. No correction for fringing is planned for the pipeline data reduction. Figure 10 illustrates the fringe pattern for the 9300е light at the FR914M filter wheel setting, where FR914M is in the middle part of the field. The upper right-hand set of fringes are through the outer FR931N ramp filter, while the inner FR853N ramp segment at the lower left does not transmit at 9300е. Figure 11 demonstrates that the fringe pattern depends on the bandwidth of the illumination but is nearly independent of which filter is in

5


Instrument Science Report ACS 00-10 the beam. The heavy solid line is a cut through the center of chip 2 in the FR931N region of Figure 10. The light solid line is the same 20е bandpass light through the F814W filter and matches to better than 1%, although the fringe amplitude may be reduced systematically by a fraction of a percent. The dotted line is for a 40е bandpass, which does systematically reduce the fringe amplitude by ~1% with respect to the fringes for the 20е band width. Therefore, the fringe pattern depends only on the CCD and the spectral distribution of the illumination at the CCD. Since the fringe amplitude changes rapidly with wavelength, obtaining a dense enough grid of monochromatic flats to correct monochromatic images at any wavelength is impractical. For example, dividing a monochromatic image at 8520е by an 8500е image does not diminish the fringe pattern. The only practical way of correcting the fringing is through modeling. The ECF is committed to modeling the fringing pattern in order to correct the HRC and WFC grism spectra. The results of that effort should be applicable to fringe removal in the imaging modes.

5. RECOMMENDED LAB PROGRAM
This section updates and expands the requirements in Paper I for lab CCD flats. See Paper I for a discussion of on-orbit calibration plans. While internal illumination provides pixel-to-pixel corrections, complete flat fields must be defined by uniform diffuse illumination with a proper f/24 OTA beam. Uniform illumination is difficult to achieve on orbit (Cox et al. 1987), although narrow band WFPC2 flats are made by labor intensive processing of streak flat observations of the sunlit earth (Biretta 1995). A full set of on-orbit sky flats from normal science exposures may require years to accumulate sufficient counting statistics; but checks of the more popular filters should be available near the end of the first year of science operations. Therefore, our basic philosophy is to produce a complete a set of flat fields on the ground, while using the internal lamps to monitor changes on-orbit. The RAS/HOMS provides a convenient flat field source longward of its refractive optics cutoff at ~3500е. For the four HRC short wavelength filters below 3500е and for a check on the longer wavelength RAS/HOMS flats, a second ground calibration setup that accurately simulates the full-field OTA beam into ACS is desirable. Alternatively, the assumption that the change with wavelength of the flat field is adequately measured by the internal lamps could be invoked to manufacture these four short wavelength continuum flats from the ratio of high S/N internal flats. Approximate LP-flats will be obtained at GSFC with a deuterium lamp illuminating a diffusing screen in front of the ACS. The G800L grism on the WFC covers the 5500-10,000е region with a peak sensitivity at 6500-7000е. P-flats for the broadband filters F625W or F814W are adequate for flat fielding grism images. Shortward of the onset of fringing at ~7000е, the rms scatter of ~0.7% should be reduced below ~0.2% by application of an F625W flat, because the worst case residual should be typical of the 0.14% residual rms scatter in the F555W/F625W

6


Instrument Science Report ACS 00-10 ratio of external flats. Longward of ~7500е, uncertainties in the fringing correction will likely dominate uncertainties in grism spectra. Because of the shorter wavelength coverage, the HRC PR200L may require a data cube of P-flats. Likewise, the wavelength dependence of the L-flat on WFC greatly complicates the low spatial frequency L-flat correction for G800L. If the wavelength dependence of the L-flat is the same for the grism as for the other filters, then the typical L-flat errors at 5500е and 8140е are <5% and <10%, respectively, when an F625W LP-flat is used as the G800L flat. To test this hypothesis and verify the uncertainties and to test the possibility of a data cube L-flat correction scheme for the grism, a monochromatic flat at 6250е should be obtained on G800L, in addition to the same illumination specified below for F625W. All fringe flats, which require monochromatic illumination, must be obtained in the lab before launch. Since the intrinsic rms pixel-to-pixel scatter is as low as 0.5% in the WFC, flat fields must have at least 40,000 electrons/px to avoid degrading the intrinsic S/N. To make flat fielding a worthwhile enterprise by significantly improving the noise to say 0.25% rms in a WFC image with no Poisson noise, the flat field exposures must have at least 160,000 counts/px for each filter. If the preliminary reproducibility result of ~0.1% is maintained, then exposures with a million counts are useful; and external CCD flats with a million counts should be obtained, as time permits. HRC images and WFC subimages can be obtained in less than the two hours for 12 full-field WFC readouts required at the full well of 80,000 electrons. For polarization measurements of small objects at ~1% precision, 0.1% stability of the flats would be especially useful. 5.1 Broadband Filters There are 25 HRC and 20 WFC broadband filters, including six polarizers on both cameras, a grism on the WFC, and a grism and a prism on the HRC. In a few supported combinations, the polarizers will be crossed with filters in another wheel. Since the internal lamps will be used to monitor changes in the flats, a high S/N CCD baseline at the million count level should be obtained on the ground. A set of three baseline flats that cover the wavelength range of each CCD at the best guess orbital temperature setpoint is suggested. This data set should be obtained at a few epochs before launch to monitor stability and to measure any decline in brightness of the internal lamps as a function of wavelength. If significant wavelength dependent changes occur, then a finer grid in wavelength would be required to transfer the measured changes in the internal flats to the full set of external flats. 5.2 Ramp Filters 5.2.1 Continuum Light and Short Wavelengths without Fringing The 12 narrow and 3 medium bandpass ramp filters provide a filter set with bandpasses of 2% or 9% and cover the wavelength range from 3710-10710е. Three separate ramp fil-

7


Instrument Science Report ACS 00-10 ter segments share the WFC field of view at each of five filter wheel positions. Only the five middle segments may be used on the HRC. The central wavelengths of the ramp filters change continuously with position along each segment. The filters have already been characterized in terms of transmission in each bandpass and to the 10^-5 level out of band (Bohlin & Tsvetanov 2000). The ramp P-flats for monochromatic light below 7000е and for continuum light are sufficiently well defined by continuum light through the broader band filters that fall closest in wavelength. This broadband set is comprised of F330W, F435W, F475W, F555W, F606W, F625W, F775W, F814W, and F850LP. The ramp filter FR388N at 3796A is the farthest (1.13x) from the bracketing central wavelengths of 3354 (F330W) and 4297е (F435W). Since the flat fields change more rapidly with wavelength at the shorter wavelengths, the rms residuals will probably be the largest at 3796е. Bandpass and sensitivity information is available in the STScI Exposure Time Calculator package for ramp filters (Van Orsow et al. 2000). To quantify the errors in the ramp flats, high S/N monochromatic and continuum flats are needed at a few test wavelengths. The acid-test wavelength should be 3796е. As discussed in Section 3 above, WFC monochromatic 6330е flats adequately match F625W flats; but residuals in the HRC could be worse due to the wavelength dependence of the Pflats and to dust particles. An ideal set of monochromatic test flats would consist of the five wavelengths at the geometric midpoints between the central wavelengths of the broadband filters F330W, F435W, F475W, F555W, F606W, and F625W, where fringing is not present. With exposures through the appropriate ramp and adjacent broadband filter, this test would quantify the differences in the dust motes and in LP-flats for monochromatic and broadband illumination. In addition, central wavelength monochromatic flats at 3300, 4350, 4750, 5550, 6060, and 6250е are required to measure the pixel-to-pixel structure and to fully quantify possible gains from interpolation schemes that might be required to remove any wavelength dependent fine structure. 5.2.2 Long Wavelengths with Fringing in Monochromatic Light At the longer wavelengths, the fringing pattern from interference of monochromatic light in the silicon layer of the CCDs is the dominant flat field effect. The best removal technique should accrue as a `fringe' benefit from the ECF modeling of the fringe patterns in grism spectra. Wolfram Freudling has provided the following wish list of 20е bandpass monochromatic images to be taken through the nearest broadband filter for both WFC and HRC: 7000, 7500, 7820, 7860, 7900, 7910, 7920, 7930, 7940, 7980, 8250, 8520, 8560, 8600, 8610, 8620, 8630, 8640, 8680, 9000, 9150, 9320, 9360, 9400, 9440, 9480, and 10000е.

8


Table 1. ACS Prelaunch Data Catalog of Full WFC Data
ENTRY
16066 16067 16068 16069 16070 16071 16072 16073 16074 16075 16076 16077 16079 16080 16081 16082 16083 16084 16085 16086 16087 16088 16089 16090 16092 16093 16094 16095 16096 16097 16098 16099 16100 16101

FILENAME
CSIJ00167185419_1 CSIJ00167190437_1 CSIJ00167191455_1 CSIJ00167192513_1 CSIJ00167193530_1 CSIJ00167194547_1 CSIJ00167195607_1 CSIJ00167200624_1 CSIJ00167201650_1 CSIJ00167202706_1 CSIJ00167203736_1 CSIJ00167204816_1 CSIJ00167212031_1 CSIJ00167213101_1 CSIJ00167214120_1 CSIJ00167215136_1 CSIJ00167220228_1 CSIJ00167221248_1 CSIJ00167222515_1 CSIJ00167223724_1 CSIJ00167224827_1 CSIJ00167225927_1 CSIJ00167231209_1 CSIJ00167232451_1 CSIJ00167235707_1 CSIJ00168000913_1 CSIJ00168002119_* CSIJ00168003325_* CSIJ00168004531_* CSIJ00168005737_* CSIJ00168010830_1 CSIJ00168011912_1 CSIJ00168012954_* CSIJ00168014036_*

OBSTYPE
INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL

DATEOBS
15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 15/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00

EXPTIME (s)
3.4 3.4 3.3 3.3 2.4 2.4 3.0 3.0 1.5 1.5 4.4 4.4 15.4 15.4 2.0 2.0 5.6 5.6 115.0 115.0 46.0 46.0 127.0 127.0 112.0 112.0 112.0 112.0 112.0 112.0 27.4 27.4 27.4 27.4

DETECTOR
WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC

FILTER1
F555W F555W F775W F775W F625W F625W F850LP F850LP F606W F606W F475W F475W CLEAR1L CLEAR1L CLEAR1L CLEAR1L F550M F550M F502N F502N F658N F658N CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L

FW1 OFF
0 3 0 3 0 3 0 3 0 3 0 3 0 0 0 0 0 3 0 3 0 3 0 0 0 0 0 0 0 0 0 0 0 0

FILTER2
CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L F435W F435W F814W F814W CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L F660N F660N FR388N FR388N FR388N FR388N FR388N FR388N FR505N FR505N FR505N FR505N

FW2 OFF
0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 3 0 0 0 0 0 0 0 3 0 3 62 65 -65 -62 0 3 62 65

CCD TEMP (C)
-76.8 -76.8 -76.7 -76.7 -76.7 -76.8 -76.9 -76.9 -76.7 -76.7 -76.8 -76.8 -76.7 -76.8 -76.8 -76.8 -76.8 -76.7 -76.7 -76.7 -76.9 -76.9 -76.9 -76.7 -76.8 -76.8 -76.8 -76.8 -76.8 -76.8 -76.8 -76.9 -76.8 -76.8

Instrument Science Report ACS 00-10

9


ENTRY
16102 16103 16105 16106 16107 16108 16109 16110 16111 16112 16113 16114 16115 16116 16123 16124 16125 16126 16127 16128 16129 16130 16469 16470 16491 16492 16521 16522 16523 16524 16525 16526 16527 16528 16530 16531

FILENAME
CSIJ00168015118_1 CSIJ00168020221_1 CSIJ00168024145_1 CSIJ00168025229_1 CSIJ00168030313_* CSIJ00168031357_* CSIJ00168032441_1 CSIJ00168033525_1 CSIJ00168034605_1 CSIJ00168035627_1 CSIJ00168040649_1 CSIJ00168041711_1 CSIJ00168042733_1 CSIJ00168043816_1 CSIJ00168161723_1 CSIJ00168162745_1 CSIJ00168163807_1 CSIJ00168164829_1 CSIJ00168165851_1 CSIJ00168170913_1 CSIJ00168171957_1 CSIJ00168173012_1 CSIJ00171192950_1 CSIJ00171193359_1 CSIJ00171232649_1 CSIJ00171233810_1 CSIJ00172160841_1 CSIJ00172162621_1 CSIJ00172164100_1 CSIJ00172165501_1 CSIJ00172170840_1 CSIJ00172172300_1 CSIJ00172173620_1 CSIJ00172175340_1 CSIJ00172182621_1 CSIJ00172184049_1

OBSTYPE
INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL

DATEOBS
16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 16/06/00 19/06/00 19/06/00 19/06/00 19/06/00 20/06/00 20/06/00 20/06/00 20/06/00 20/06/00 20/06/00 20/06/00 20/06/00 20/06/00 20/06/00

EXPTIME (s)
27.4 27.4 30.0 30.0 30.0 30.0 30.0 30.0 7.6 7.6 7.6 7.6 7.6 7.6 7.2 7.2 7.2 7.2 7.2 7.2 0.5 0.5 35.0 35.0 1.0 1.0 4.0 4.0 8.0 8.0 40.0 40.0 30.0 30.0 40.0 40.0

DETECTOR
WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC

FILTER1
CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L G800L G800L F892N F892N F625W F625W CLEAR1L CLEAR1L F555W F555W CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L

FW1 OFF
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 -61 -64 0 3 0 0 0 3 0 0 0 0 0 0

FILTER2
FR505N FR505N FR656N FR656N FR656N FR656N FR656N FR656N FR914M FR914M FR914M FR914M FR914M FR914M FR459M FR459M FR459M FR459M FR459M FR459M CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L CLEAR2L F814W F814W CLEAR2L CLEAR2L F435W F435W FR914M FR914M FR656N FR656N

FW2 OFF
-65 -62 0 3 62 65 -65 -62 0 3 62 65 -65 -62 0 3 62 65 -62 -65 0 0 0 0 0 0 0 3 0 0 0 3 0 3 62 65

CCD TEMP (C)
-76.7 -76.7 -76.9 -76.8 -76.7 -76.8 -76.8 -76.8 -76.8 -76.7 -76.9 -76.8 -76.7 -76.8 -76.8 -76.8 -76.8 -76.8 -76.7 -76.8 -76.8 -76.8 -76.8 -76.7 -76.8 -76.8 -76.8 -76.8 -76.7 -76.8 -76.8 -76.8 -76.8 -76.9 -76.7 -76.8

Instrument Science Report ACS 00-10

10


ENTRY
17039 17040 17041 17042 17044 17045 17046 17047 17058 17060 17061 17063 17064 17065 17067 17068 17070 17071 17073 17074 17076 17077 17078 17079 17091 17092 17093 17094 17095 19114 19115 19116 19117 19118 19119 19120

FILENAME
CSIJ00175032030_1 CSIJ00175034541_1 CSIJ00175040251_1 CSIJ00175041920_1 CSIJ00175051600_1 CSIJ00175053049_1 CSIJ00175054540_1 CSIJ00175060029_1 CSIJ00175154059_1 CSIJ00175163249_1 CSIJ00175170229_1 CSIJ00175174819_1 CSIJ00175175909_1 CSIJ00175181300_1 CSIJ00175190418_1 CSIJ00175193910_1 CSIJ00175210040_1 CSIJ00175215000_1 CSIJ00175225230_1 CSIJ00175232010_1 CSIJ00176000200_1 CSIJ00176001829_1 CSIJ00176003510_1 CSIJ00176010459_1 CSIJ00176035139_1 CSIJ00176041409_1 CSIJ00176045240_1 CSIJ00176052900_1 CSIJ00176060400_1 CSIJ00256194959_1 CSIJ00256195950_1 CSIJ00256201309_1 CSIJ00256202309_1 CSIJ00256203610_1 CSIJ00256204609_1 CSIJ00256205949_1

OBSTYPE
EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL EXTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL INTERNAL BIAS

DATEOBS
23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 23/06/00 24/06/00 24/06/00 24/06/00 24/06/00 24/06/00 24/06/00 24/06/00 24/06/00 12/09/00 12/09/00 12/09/00 12/09/00 12/09/00 12/09/00 12/09/00

EXPTIME (s)
400.0 400.0 400.0 400.0 300.0 300.0 300.0 300.0 100.0 1000.0 1000.0 200.0 50.0 50.0 1500.0 1500.0 2000.0 2000.0 1000.0 1000.0 400.0 400.0 400.0 1000.0 600.0 600.0 1500.0 1500.0 1500.0 3.4 3.4 2.4 2.4 2.0 2.0 0.0

DETECTOR
WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC WFC

FILTER1
CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L CLEAR1L F555W F555W F625W F625W CLEAR1L CLEAR1L CLEAR1L

FW1 OFF
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

FILTER2
FR656N FR656N FR656N FR656N FR656N FR656N FR656N FR656N FR914M FR914M FR914M FR914M FR914M FR914M FR914M FR914M FR914M FR914M F814W F814W F814W F814W F814W F814W F814W F814W F814W F814W F814W CLEAR2L CLEAR2L CLEAR2L CLEAR2L F814W F814W F814W

FW2 OFF
30 30 30 30 30 30 30 30 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

CCD TEMP (C)
-76.8 -76.7 -76.8 -76.7 -76.8 -76.8 -76.7 -76.9 -76.8 -76.8 -76.8 -76.8 -76.8 -76.8 -76.7 -76.9 -76.8 -76.8 -76.8 -76.8 -76.8 -76.7 -76.8 -76.8 -76.8 -76.8 -76.8 -76.8 -76.7 -66.7 -66.6 -66.6 -66.7 -66.6 -66.6 -66.6

Instrument Science Report ACS 00-10

11

* Contaminated by straylight from internal lamp.


Instrument Science Report ACS 00-10

REFERENCES
Biretta, J. 1995, in "Calibrating HST: Post Servicing Mission," ed. A. Koratkar & C. Leitherer, (Baltimore:STScI), p. 257. Bohlin, R. C., Hartig, G., Lindler, D. J., Meurer, G., & Cox, C. 1999, Instrument Science Report, ACS 99-01, (Baltimore:STScI), Paper I. Bohlin, R.C., & Tsvetanov, Z. 2000, Instrument Science Report, ACS 00-05, (Baltimore:STScI). Cox, C., Bohlin, R. C., Griffiths, R. E., & Kelsall, T. 1987, "Standard Astronomical Sources for HST: 6. Spatially Flat Fields," (Baltimore:STScI). Sparks, W. B., Jedrzejewski, R., Clampin, M., & Bohlin, R.C. 2000, Instrument Science Report, ACS 00-03, (Baltimore:STScI). Van Orsow, D., Boffi, F. R., Bohlin, R. C., & Shaw, R. A., 2000, Instrument Science Report, ACS 00-07, (Baltimore:STScI).

12


Instrument Science Report ACS 00-10

ACS WFC LP Flat
CLEAR1L F435W rms(%)= 0.97 0.62 0.75

0.90

1.10

lpwsptrln-00172sm02f435w

Figure 1: WFC LP-flat for the F435W filter derived from external illumination by a Spectralon diffusing screen. The image stretch is from 0.9 to 1.1, as indicated by the reference gray scale at the top. The image is the sum of two exposures which were recorded on day 172 of 2000 (00Jun20). The three rms(%) values are for the total, counting, and intrinsic scatter of the flat field in a 100x100 pixel box centered at (2000,1000) in chip 2. There are two 4096x2048 CCD segments with chip 1 at the bottom and chip 2 at the top, where the gap in sky coverage between the segments is about 40 pixel wide. The orientation of the flat is in the prelaunch coordinate frame. The GO science data will be flipped about a horizontal axis so that chip 2 will be at the bottom. The intensity is corrected for the geometric distortion of effective pixel area per Fig. 8 of Paper I.

13


Instrument Science Report ACS 00-10

ACS WFC LP Flat
CLEAR1L F435W rms(%)= 0.83 0.36 0.74

0.90

1.1

lpwint00167sm02f435w-2

BOHLIN: prtimg 23-Aug-2000 15:29

Figure 2: As for Figure 1, except for internal illumination obtained on 167/2000 (00Jun15).

14


Instrument Science Report ACS 00-10

ACS WFC RATIO
CLEAR1L F814W / F625W CLEAR2L

0.95

1.05

lpwint00167sm02f814w-2 / lpwint00167sm02f625w-2

BOHLIN: prtimg 5-Sep-2000 13:49

Figure 3: Ratio of F814W/F625W LP-flats from the internal lamps with a grey scale stretch from 0.95 to 1.05.

15


Instrument Science Report ACS 00-10

ACS WFC RATIO
CLEAR1L F814W / F625W CLEAR2L

0.95

1.05

lpwsptrln-00172sm02f814w / lpwsptrln-00171sm02f625w

BOHLIN: prtimg 11-Sep-2000 11:21

Figure 4: As for Figure 3, except for external illumination from Spectralon. The large scale doughnut pattern is the same as in Figure 3 to 2-3%; and this difference may be due to non-uniformities in the Spectralon illumination.

16


Instrument Science Report ACS 00-10

ACS WFC RATIO
CLEAR1L F814W / CLEAR1L F814W

0.95

1.05

lpw850-00176sm02f814w / lpw700-00176sm02f814w

BOHLIN: prtimg 11-Sep-2000 12:16

Figure 5: Ratio of monochromatic images at 8500 and 7000е with external illumination from the RAS/HOMS and mylar diffuser. The ratio is boxcar smoothed to remove the fringing, except for 2% of the width around the edges of each CCD chip. Again, the doughnut of Figures 3-4 appears for light of a similar difference in effective wavelength. This L-flat pattern matches Figure 3 to ~1%.

17


Instrument Science Report ACS 00-10

ACS WFC RATIO
CLEAR1L F435W rms(%)= 0.76 0.74 0.18

0.98

1.02

lpwsptrln-00172sm02f435w-2/lpwint00167sm02f435w-2 1024x1024 center: 2000x1000

BOHLIN: wbloup 28-Aug-2000 19:52

Figure 6: Ratio of the F435W external/internal subimages. The major difference in the two flats is that the external illumination has a spurious flocculent aspect that is not present in the external/internal ratios at longer wavelengths. The two circular features near the bottom center are caused by dust on the detector window and are ~100px in diameter vs. only ~30px for the HRC.

18


Instrument Science Report ACS 00-10

ACS WFC LP Flat
CLEAR1L F814W rms(%)= 0.73 0.46 0.57

0.98

1.02

lpwsptrln-00172sm02f814w-2 central 1024x1024

BOHLIN: wbloup 23-Aug-2000 16:14

Figure 7: Subimage F814W flat showing prominent dust motes and the orthogonal stepand-repeat pattern from the CCD manufacturing process.

19


Instrument Science Report ACS 00-10

ACS WFC RATIO
F625W CLEAR2L / CLEAR1L F435W rms(%)= 0.81 0.72 0.36

0.98

1.02

lpwsptrln-00171sm02f625w-2 / lpwsptrln-00172sm02f435w-2 1024x1024 center: 2000,1000

BOHLIN: wbloup 5-Sep-2000 14:01

Figure 8: Ratio of subimages F625W/F435W, both with continuum external illumination by the RAS/HOMS and spectralon. The most important difference is that the dust motes are not as deep at the longer wavelength. The freckles that appear as tiny white dots in the corresponding Figure 4 of Paper I are largely absent in this WFC ratio, which has a different backside CCD treatment than the build-2 HRC with its Lesser coating.

20


Instrument Science Report ACS 00-10

ACS WFC LP Flat
CLEAR1L F814W rms(%)= 5.30 0.84 5.23

0.90

lpw930-00175sm02f814w-2 central 1024x1024

BOHLIN: wbloup 23-Aug-2000 20:44

Figure 9: Subimage for 20е bandpass monochromatic light at 9300е. The rms fringe amplitude is 6.4% for the whole frame, while the 5.3% listed on the Figure is for the central 100x100 pixels.

21


Instrument Science Report ACS 00-10

ACS WFC LP Flat
CLEAR1L FR914M rms(%)= 5.66 0.67 5.62

0.90

1.10

lpw930-20A-sm02fr914m-2

BOHLIN: prtimg 24-Aug-2000 18:17

Figure 10: FR914M filter with monochromatic illumination at 9300е. The fringe pattern of the CCD is evident at the two field positions that have high throughput at 9300е. There are three ramp segments oriented at 45deg at the FR914M setting: the outer FR931N at the top right, FR914M in the middle, and the inner FR853N at the lower left. FR853N cuts off beyond ~8910е and does not transmit 9300е light. The FR914M has a double peaked sensitivity function at 9300е.

22


Instrument Science Report ACS 00-10

23

Figure 11: Heavy solid line - average of rows 1050:1060 through chip 2 of Figure 10 with the 20е bandpass illumination. Dotted line same as heavy solid line, except for an image with a 40е bandpass. Light solid line - 20е bandpass through the F814W filter instead of FR931N.