MOIRCS Instrument Parameters
| Detectors | HAWAII-2 2048x2048 x 2 |
| Pixel size | 18 um |
| Pixel scale(1) | 0.117'' |
| Field of view | 3.94' x 6.90' (2) |
| Gain(3) | 2.95 e-/ADU @ Channel 1 (Eng #40) 3.24 e-/ADU @ Channel 2 (Sci #64) |
| Dark current | < 0.1 e-/s |
| Saturation level | 146 000 e- (chip 2) (4) |
| Read noise | 32 e- rms @ Channel 1 33 e- rms @ Channel 2 (5) |
| Number of amplifiers per detector | 4 |
| Minimum Exposure(6)(7) | 11.5 s (NDUMMYREAD=0) 19.7 s (NDUMMYREAD=2) |
| Filter exchange time | 15~60 sec |
| MOS | |
|---|---|
| Number of mask slots for users | up to 14 (12 for users during S08A) |
| Number of slits per mask | 30~50(8) |
| Days needed for cooling mask | > 2 whole days |
| MOS aquisition time | 25 to 40 min.(9) |
(1) Pixel scale for each detector has slightly different value. And also, the values in raw image also changes with the distance from the optical center due to the optical distortion.
(2) About 0.4 arcmin2 is vignetted in the field of view of 4' x 7' (Please see the FOV page).
(3) New estimate by photon-transfer method in Dec 2007. The values shown here before (3.15 e-/ADU for old chip1 #87, 2.86 e-/ADU for chip 2 #64) may have been affected by the read-level anomaly in partial-read mode.
(4) Newly measured value as of Dec 2007.
(5) Directly measured value using a set of 21-sec dark frames in Dec 2007. The readout noise estimated by the photon-transfer method in Dec 2007 showed a similar values.
(6) These values are for the read of the entire array. We recommend to use about +1.5 sec longer than those values as the "minimum" for actual science data to suppress the reset anomaly (see the Detector Information below).
(7) To achieve shorter exposures than those listed values, a partical read mode is also available (Please see the Exposures page).
(8) Assuming the typical length of slitlets of 10-20 arcseconds.
(9) Please also see the MOS page. The acquisition time for Long slits should be similar.
Detector Information
Engineering Detector for Channel 1
Due to a technical problem in the detector on channel-1, we replaced it to an Engineering-grade chip on October 2007. We have to operate MOIRCS with this array until a replacement is found. The data taken by the chip should be handled with enough caution for scientific use. The detail about the character of channel-1 detecor has been updated to the dedicated website.
Channel-1 Engineering-grade detector (chip-1) has a prominent large ring-like structure as seen in the dark images. This region has significantly-high dark noise and more possible latents. Scientific use for that ring region may be difficult, or should use with enough calibration data. There are also several smaller and patchy structures, which is sometimes difficult to distinguish with the real scientific features. The example of 600-sec dark frames in fits format are available: CHIP 1 / CHIP 2.
The relative sensitivity of the chip-1 (except ring region) to the detector on channel 2 (chip-2) is estimated to 74.4% in Ks (other bands are not known yet). Gain value and readout noise of chip-1 is similar to chip-2. Linearity for chip-1 (good region only) is also as good as chip-2, but saturation level seems to be less for chip-1.
Appearance of the data
There are no dead pixel islands with the size larger than 1.5 arcseconds on both detector, though several scratches or small holes do exist. There also exists one-pixel-width non-data rows/columns between each quadrant. The unreadable regions are [1:1024,1025:1025], [1025:2048,1024:1024], [1024:1024,1:1024], and [1025:1025,1025:2048]. Examples of raw images can be downloaded from here (CHIP 1 [gif, fits] / CHIP 2 [gif ,fits]).
Linearity
The figures below are the result of the detector linearity measurements executed in Dec 2007. The linearity is maintained within 1% up to 85000 e- and 110000 e- for chip 1 and chip 2, respectively. Note that under the CDS readout the full well will vary depending on the position on the chip and the input flux density (above values were measured near the center of detectors: i.e. deepest well is expected). And also, the measurement for chip 1 was done using region that are relatively clean. For high-dark "ring" region on chip 1 the linearity may be poor.
We recommend to keep the sky count under 20000 ADU for science data, because the detector characteristics starts to change.
|
| Figure 11a. |
|
|
| Figure 11b. | Figure 11c. |
Latent Image
A latent image appears at the level of ~0.06-0.6%. The amplitude seems to depend on the background level (under investigation). In imaging mode the latent seems generally weak ( <0.3% ) and in J it is a bit prominent. In spectroscopy mode or during the narrow-band observation the latent sometimes goes up close to 1% due to the low sky level. Though time is the major factor for eliminating the affection of the latents by saturation, taking several short-exposure dark frames may also work.
Low-level signals by latents may occasionally remain even in your fully-reduced data. A special care must be paid about the reality of faint objects that are close to bright stars. If the magnitude or appearance of the object in question largely changes between final-combined images by average and by median, it can be a spurious. Also, the objects in question around a bright star traces the same pattern as the dither, it may be spurious.
Reset Anomaly (Bias Tilt)
Reset anomaly (Bias tilt) is a specific effect seen in some HgCdTe FPAs such as HAWAII and HAWAII-2. It can be reduced by operating an array continuously and sampling by the CDS (Correlated Double Sampling) method. Though the current chip-2 detector shows only a small level (a few %) of the reset anamaly, the chip-1 detector shows relatively large level of reset anomaly. As getting data with dummy-read option (NDUMMYREAD=2) greatly suppresses it, we usually operates MOIRCS under the mode since November 2007. However, the first image after taking a rest for a while (5~10 minutes) generally shows relatively large anomaly, and may need to abondon that data.
The instrumental minimum exposure times are listed in the table at the top of the page. In the minimum exposure data we see relatively strong reset anomaly. This can be suppressed when we set a bit (+1 to +1.5sec) longer exposure times than the true minimum. So we tend to use the minimum exposure of 13.0 sec for NDUMMYREAD=0 and 20.5 sec for NDUMMYREAD=2 for full readout case.
Please note that all data on these pages are subject to change as the evaluation of the performance of MOIRCS progresses.
Created by Yuka Katsuno Uchimoto (Feb 2006)
Revised by Ichi Tanaka (2008-02-07).