calibration results by item AUTHORS : A.R. Martel and G. Hartig
GOAL :
Characterize the shape and intensity of the LED illumination for the three modes (LOW, MEDIUM, and HIGH) for both blades of the shutters and for both MEBs on the flight build detectors HRC#1 and WFC#4. The repeatability of short LED exposures will also be verified on the HRC.
LOCATION AND DATE :
All the images were acquired at BATC on Nov 2-5, 2001 while ACS was configured in RAS/HOMS.
METHOD AND DATA :
ACS is configured in RAS/HOMS and the vortex chillers are turned on. The detectors are at their nominal temperatures (WFC at -77 C and HRC at -80 C) and gain/amplifier settings (WFC : ABCD, gain=1, offset=3; HRC : C, gain=2, offset=3). The flash SMSs JGCH35A and JGCW35A, described in detail in the SMS Master List, were executed for both MEBs 1 and 2. Bias frames from the same MEB and shutter blade sides were subtracted from each FLASH frame to insure proper removal of any stray light and to account for possible side-dependent electronic levels. The bias levels were adjusted with the virtual and leading physical overscans for the HRC and WFC, respectively. The count rates of the "flat fields" of a given shutter side and MEB are measured and the images are ratioed and compared to detect differences in the illumination pattern. Comparison between images from the two shutter blades is also performed.
RESULTS :
1. HRC
The images and their count rates are listed in Tables 1 and 2 for both shutter blades and MEBs. When the HIGH, MED, or LOW images are paired, the average image and its count rate are given. Exposure times of 400 sec were chosen since the maximum allowed FLASH integrations are 409.5 sec. For these exposures, the median count rate over the whole HRC field is tabulated. The short 1 sec exposures serve to verify repeatibility of the LED illumination and were acquired on shutter side B only. Their mean count rate and error were measured in a 100x100 box at the peak of the illumination (Amp A corner, see below) for better statistics. In orbit, typical FLASH exposure levels will be about 100 e-/pixel, which can be reached in a few seconds. The level will depend on the degree of CTE degradation over time and will likely increase as ACS ages.
The HRC illumination peaks in the Amp A (lower-left) corner for the HIGH, MED, and LOW settings (see Figs 1-3) and smoothly decreases towards the opposite Amp D corner. The drop is roughly 65% from one corner to the other. The images on shutter side A and B are essentially identical (Figs 4 and 6) for both MEBs and show a very weak decrease of ~0.2% from edge AC to edge BD (Figs 5 and 7). This behavior is expected since the LEDs are mounted near the window heater tube and the light is primarily reflected off this assembly and the light baffle tube (the HRC shutter is too far to contribute significantly to the reflection).
In general, the count rates on MEB 1 are a factor of 1.3 greater than on MEB 2 for the HIGH, MED, and LOW settings. No difference is seen in the count rates between the two shutter blade sides, at least for the long 400 sec HIGH exposures. The short term variability of the 1 sec images is plotted in Fig. 8 for both MEBs. If the first data point of each set is ignored, the observed fluctuations are of the order of +/- 2%, with the MEB 2 data apparently suffering slightly more variability than the MEB 1 set. Please note that these data are likely limited by small-number statistics (~10 DN/pix) and corresponding large uncertainties.
Table 1 : HRC FLASH Calibration : Shutter Side A
| LED | EXP (sec) | COUNT RATES (DN/sec) | |||
|---|---|---|---|---|---|
| ID | MEB 1 | ID | MEB 2 | ||
| HIGH | 400.0 | 34903+34904 | 12.4 | 34926+34927 | 9.3 |
| MED | 400.0 | 34905+34906 | 9.5 | 34928+34929 | 7.2 |
| LOW | 400.0 | 34907+34908 | 1.5 | 34930+34931 | 1.1 |
Table 2 : HRC FLASH Calibration : Shutter Side B
| LED | EXP (sec) | COUNT RATES (DN/sec) | |||
|---|---|---|---|---|---|
| ID | MEB 1 | ID | MEB 2 | ||
| HIGH | 400.0 | 34910+34911 | 12.4 | 34933+34934 | 9.4 |
| HIGH | 1.0 | 34913 | 18.93 +/- 4.19 | 34936 | 13.73 +/- 3.92 |
| HIGH | 1.0 | 34914 | 18.89 +/- 4.09 | 34937 | 14.15 +/- 3.98 |
| HIGH | 1.0 | 34915 | 18.16 +/- 4.12 | 34938 | 14.02 +/- 3.94 |
| HIGH | 1.0 | 34916 | 18.53 +/- 4.15 | 34939 | 14.47 +/- 3.91 |
| HIGH | 1.0 | 34917 | 18.59 +/- 4.12 | 34940 | 14.19 +/- 3.99 |
| HIGH | 1.0 | 34918 | 18.77 +/- 4.15 | 34941 | 14.33 +/- 3.94 |
| HIGH | 1.0 | 34919 | 18.21 +/- 4.17 | 34942 | 14.58 +/- 3.93 |
| HIGH | 1.0 | 34920 | 18.18 +/- 4.25 | 34943 | 13.81 +/- 3.94 |
| HIGH | 1.0 | 34921 | 18.50 +/- 4.24 | 34944 | 14.11 +/- 3.89 |
| HIGH | 1.0 | 34922 | 18.87 +/- 4.19 | 34945 | 14.29 +/- 3.92 |
2. WFC
The WFC LED light distribution is centrally concentrated. The cuts shown in Figs 9-11 show a decrease of ~23% and ~30% from the central peak to the AC and BD edges, respectively, for the HIGH, MED, and LOW settings. From Figs 12 and 14 and their associated cuts in Figs 13 and 15, the light distribution between the two blade sides is clearly different for both MEBs. A strong gradient is observed from the Amp A corner (lower-left) to the Amp D corner (upper-right), typically of the order of ~20%. This difference is also confirmed by the count rates : on shutter blade B, the rates are a factor of ~1.3 greater than on shutter blade A. Such a strong dependence on the shutter blade side is expected for the WFC since the LEDs are mounted on the detector assembly and the light is primarily reflected off the backside of the shutter. We also find that the count rates on MEB 2 are typically a factor of ~1.4 greater than on MEB 1, contrary to the behavior of the HRC rates.
We note that digital saturation (>65535 DN/pix) is reached in 400 sec for shutter blade B on MEB 2 (ID 35067, 35068) in HIGH mode at the field center, primarily in the Amp A quadrant (see Fig. 14). Two 300 sec exposures (ID 35070 and 35071) give peak counts of 54500 DN/pix (with no bias subtraction) so to avoid saturation on shutter blade B and MEB 2, the exposure times in HIGH mode should be <360 sec. There is no risk of saturation in MED and LOW modes.
Table 3 : WFC FLASH Calibration : Shutter Side A
| LED | EXP (sec) | COUNT RATES (DN/sec) | |||
|---|---|---|---|---|---|
| ID | MEB 1 | ID | MEB 2 | ||
| HIGH | 400.0 | 35049+35050 | 67.1 | 35060+35061 | 95.0 |
| MED | 400.0 | 35051+35052 | 51.5 | 35062+35063 | 72.2 |
| LOW | 400.0 | 35053+35054 | 7.8 | 35064+35065 | 10.4 |
Table 4 : WFC FLASH Calibration : Shutter Side B
| LED | EXP (sec) | COUNT RATES (DN/sec) | |||
|---|---|---|---|---|---|
| ID | MEB 1 | ID | MEB 2 | ||
| HIGH | 400.0 | 35056+35057 | 88.9 | 35067+35068 | 126.2 |
CEI SPECIFICATION :
Because of the late installation of the FLASH LEDs, there is no formal CEI specification for this item. However, some design goals were defined and we can verify some with the data in hand. The exposure uniformity across the CCDs is defined as <50%. As shown above, this goal is met only for the WFC. The shot-to-shot variation is defined as <1%, except where limited by photon statistics. Our series of short 1 sec HRC exposures yields variations of ~2% for both MEBs, although the uncertainties are large because of small number statistics. An exposure range of 10 to 1000 photoelectrons/pixel is stated as a goal. These levels are easily reached for both the HRC and WFC with the LOW and HIGH settings at the standard gains.
CONCLUSION :
