Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.apo.nmsu.edu/35m_operations/35m_manual/Instruments/DIS/OLD/dis/dis_quickstart.pdf Дата изменения: Tue Apr 20 01:40:58 2004 Дата индексирования: Sun Apr 10 10:29:31 2016 Кодировка: Поисковые слова: южная корона

April Draft

8, Version 2004

A Quick Start Guide To The Double Imaging Sp ectrograph (DIS)
John C. Barentine Apache Point Observatory jcb@apo.nmsu.edu

1.

Introduction

The Double Imaging Spectrograph (DIS) is a low- to medium-resolution optical-wavelength spectrograph covering the spectral range of approximately 3700-10000°. Two images are creA ated in each exposure; light entering the instrument is sorted into `red' and `blue' beams by means of a dichroic element and directed to complementary pairs of gratings blazed to the red and blue ends of the spectrum, respectively. The spectrum is split into halves at the dichroic break, around 5500°. The instrument can also act as a visible light imager by A substituting mirrors for the gratings and an open slide for the slit; a built-in pair of g and r filters is available for imaging. A still CCD slitviewing camera aids in pointing and guiding exposures.

1.1.

Detector Characteristics

The CCDs in the DIS cameras are Marconi CCD42-20-310 chips. Their linearity, gain and read noise were measured in April 2004. Chip Red Blue Linearity 99.9%60K 99.6%60K Saturation Scale ( pix-1 ) 65K 0.549 65K 0.543 Gain (e- ADU-1 ) Read Noise (e- pix-1 ) 1.692±0.003 4.907±0.819 1.611±0.006 4.502±0.227

1.2.

Instrument Setup Requests

It is especially important when using DIS that the observer send a very clear and specific instrument setup requests to APO at least 24 hours prior to his or her scheduled observing time. We need to know several pieces of information from you: 1) Requested grating set 2) Wavelength centers of requested gratings 3) CCD binning or windowing


April Draft

8, Version 2004

­2­ 4) Slit choice 5) Any additional setup information (nonstandard slit and/or filter requests) Gratings are the cause of most confusion with novice DIS users. There are six gratings available in total; refer to the APO website for full do cumentation. The six gratings are designated by the color for which they are blazed (`red' or `blue') and the resolution (`low', `medium' and `high'). We offer two so-called "default" grating pairs which are useful to the large ma jority of DIS observers: 1) Low Blue + Red Medium and 2) High Blue + High Red. These two setups provide the best pairings in terms of wavelength coverage and spectral resolution following the DIS optics upgrade of 2002. In the case of the first pair, the medium-resolution red grating has approximately the same wavelength coverage as the low-resolution red grating, but a throughput problem with the Low Red grating exists. Throughput information is available on the DIS webpages. We encourage users to be very specific about their grating selections, even if those selections match the "default" setups. Please state BOTH the color AND the resolution of the gratings you want; e.g., don't simply ask for the "low default" or "medium setup". Vague grating requests by observers in the past have led to incorrect instrument setups which adversely affected science. Different spectral window regimes can be imaged on the DIS chips by changing the grating tilts via stepper motors. Each grating has an asso ciated "default center" which is chosen to give the widest possible spectral window, but observers may request alternate centers as their science requires. The default values are listed in the online DIS do cumentation. We can also prepare the instrument with your requested CCD windowing and binning values, but the user can change these values in the control software at any time. Finally, let us know which slit(s) you would like available, along with any additional special requests such as slitwheel filters. User-supplied slits can be readily accommo dated in DIS; please contact the Observing Specialists for details, and be sure to ship your slits with plenty of lead time before your run.

2.

Operating The Instrument

DIS is operated within the Telescope User Interface (TUI) software written and maintained by Russell Owen at UW. More information about the environment is available at http://tycho.apo.nmsu.edu:81/TUI/. First, before taking any exposures, verify your correct setup request with the Observing Specialist. Also check that the grating set, wavelength center, and slit displayed in your TUI matches your request. In TUI, select DIS from the "Inst" menu in the Status window. In the window which pops up (Figure 1), the current status of the shutter, slitwheel mask, etc. are shown at top. Grating information is given in the center of the window, and CCD information toward


April Draft

8, Version 2004

­3­

Fig. 1.-- The DIS status window in TUI which appears when the user initially selects the instrument. the bottom; if any of these status fields are in pink, the display needs to be synced with current information. Select "Refresh Display" from the TUI menu in the Status window. The salient information provided includes the current grating set, the linear dispersion of those gratings, and the central wavelength values, labeled . The "Show More" button in the TUI status window, as its name suggests, provides further information (Figure 2): the turret detent position, dewar temperatures and grating tilt values. Pressing the button again (now labeled "Show Less") will suppress display of this information. CCD windowing and binning information can be displayed by pressing the "Show CCD" button (Figure 3). Thus far the user is "safe" in the regard of only displaying information; none of the user-configurable aspects of DIS can be changed nor exposures taken. Some users will want to change the configuration of the instrument during the night. For example, the grating central wavelengths can be changed to accommo date different spectral windows, and different slitmasks can be selected that are best suited to changing seeing conditions. To reveal the configurable aspects, click on the "Show config" button (Figure 4). Pull-down menus in this window allow the user to change the status of the slitmask wheel, put filters in and out of the beams for imaging, and Each exposure results in two files; the last letter of each file, r or b, denotes which part of the spectrum it consists of, and affect changes to the position of the DIS turret. Be careful!!!When the configuration aspects are left open, it is easy to inadvertently change a number of instrument parameters in harmful ways. To update the displayed values of the different parameters, click the "Current" button. If you want to change the grating wavelength centers, enter them in the boxed-in areas where the numbers 4602 and 6799


April Draft

8, Version 2004

­4­

Fig. 2.-- Additional instrument information displayed in the DIS status window in TUI on pressing the "Show More" button. appear in Figure 4. These boxes correspond to the wavelength centers for the blue and red gratings, respectively. Last but not least, the "Expose" button in the status window will bring up another window for setting up and executing exposures (Figure 5). Fields in this window control the type, length, file name and sequence number of exposures. During shutter open time and readout, a moving countdown/up bar will appear in the Exp Status field to indicate progress along with an exposure/readout time. Buttons also appear in this window for pausing and aborting exposures.

2.1.

Setting Up Exposures And Taking Data 2.1.1. Spectral Mode

The algorithm for taking spectra with DIS can be summarized as follows: 1) Slew to your target. 2) Take a slitviewer image. The DIS slitviewer has a field slightly more than 2 on a side; if the field of your ob ject do esn't appear, first try increasing the exposure time, then doublecheck your co ordinates. If neither of these remedies helps, ask your OS to check telescope


April Draft

8, Version 2004

­5­

Fig. 3.-- Detector configuration information displayed in the DIS status window in TUI on pressing the "Show CCD" button. pointing. For further information, see section 2.2, "Running The Slitviewer". 3) Request autoguiding, if needed. Ask the OS to start guiding. Wait about a minute for the guider to find a star and keep it reliably, then confirm guiding with the OS before beginning your exposure. 4) Begin exposure.After setting the exposure time, checking the filename convention and number sequence, click "Expose" to begin. Keeping an eye on the guiding via a series of slitviewer frames taken throughout your exposure is advised.

Fig. 4.-- User-configurable information displayed in the DIS status window in TUI on pressing the "Show config" button.


April Draft

8, Version 2004

­6­

Fig. 5.-- User-configurable information displayed in the DIS status window in TUI on pressing the "Show config" button. 2.1.2. Imaging Mode In imaging mo de, the pro cedure is similar, but for obvious reasons leaves out the slitviewer. The instrument is switched from spectral to imaging mo de by rotating the turret to the "mirrors" position and rotating the slitmask wheel to the "open" position. This is done in the DIS Config window of TUI by pulling the drop-down Mask menu to "open" and pulling the Turret menu down to "open". Built-in g and r filters can be optionally moved into the beam by pulling the FIlter menu down to "Gunn-Thuan". If switching between mo des throughout a night, be aware that fo cus is sometimes slightly different between the two mo des. Ask your OS to check fo cus if you're unsure; otherwise, examine your first imaging frame to make sure the fo cus is proper before pro ceeding. One more important note about imaging: don't panic when your images show sky on only a small part of the chip! The detectors were chosen for their optimal size and aspect ratio for spectroscopy. In pro jection, the open slide in the slitmask wheel is rather small compared to the area taken up by spectra. Your field of view in imaging mode is approximately 4 в6 .

2.2.

Running The Slitviewer

DIS is equipped with an Apogee AP7 still CCD camera which images the reflective surface of the various slitmasks. The masks are slightly tiled in the slitmask wheel and their surfaces polished to a mirror-finish; since the slits reside in the fo cal plane of the telescope, images reflected from the slitmasks are in fo cus. A flat mirror directs the reflected image of the sky immediately surrounding the slit onto the Apogee camera; standalone software running on tycho.apo.nmsu.edu allows for control of this camera. Images from the slitviewing


April Draft

8, Version 2004

­7­ camera are used to acquire ob jects, deposit them on the slit, and monitor guiding. The software is maintained by Jon Holtzman at NMSU (holtz@nmsu.edu). To start the slitviewer software, log into the visitor1 account on tycho (call APO for the current password). At the tycho prompt, type slitview; a new window will open which lo oks like the window in Figure 6. A list of commands is available by typing ?at

Fig. 6.-- The slitview control window. the Command:prompt. Two commands are the most frequently used: EXP and REPEAT. EXP takes as an argument the desired number of seconds for slitviewer exposures and will take one image before returning a prompt. The image is displayed lo cally in the software's internal image display client, as shown in Figure 7. Images taken with slitview are also archived to disk on tycho.apo.nmsu.edu:/export/images/dis-slit. Each night's data is identified with the filename YYYYMMDD.nnnn.fits, where YYYYMMDD is the UT date and nnnn is a running file sequence number. It is useful for many observers to copy over dis slitviewer images from the specified directory for context. The REPEAT command takes an endless series of images, each with length equal to that of the last EXP command; the sequence can be stopped by issuing CTRL+C. The REPEAT command is particularly useful for monitoring telescope guiding once an exposure is underway; if an observed ob ject is sufficiently bright, it can also be useful to take a running image sequence while offsetting the telescope to drop the ob ject on the slit.


April Draft

8, Version 2004

­8­

Fig. 7.-- The slitview image display window. 2.2.1. Computing Offsets With slitview Blind pointing with DIS is typically quite go o d, but rarely do ob jects land squarely on the slit. Depending on the lo cal pointing accuracy, ob ject will land somewhere in a slitviewer frame, but rarely outside of it. In other words, if your ob ject do es not appear within a slitviewer frame upon completion of a slew, suspect your co ordinates first. To aid in placing your ob ject on the "sweet spot" of the slit, a feature is incorporated into the slitviewer control software to calculate the appropriate offsets, which the user then feeds to the telescope via TUI. To use this feature, lo cate your ob ject in the Dis Slit Viewer window, place your cursor over the ob ject and type "c". The software attempts to centroid on the ob ject nearest the cursor, then in the slitviewer control window returns two numbers after the word OFFSET. These are the offsets required in the instrument (x, y ) co ordinate system to place the ob ject directly on the sweet spot of the slit. They can be entered in the TUI Offset window. For diffuse or faint ob jects on which the software would struggle to centroid, type "x" instead of "c". The former command calculates offsets directly from the cursor position, rather than attempting to centroid on the nearest discrete ob ject. Once the offset is complete, take another slitviewer image to confirm the correct placement of your ob ject. Depending on a variety of factors, more than one iteration of this pro cess may be required to adequately center the ob ject on the slit. During autoguiding, it is useful to run an image sequence with the REPEAT command. Very small autoguiding errors can accumulate over the course of an exposure several tens of minutes long, causing


April Draft

8, Version 2004

­9­ your ob ject to slowly migrate off the slit.

2.3.

Focusing

The fo cusing of light from the 3.5m onto the DIS slit is accomplished by slightly changing the fo cal length of the telescope until the fo cal plane matches the physical position of the slit. This is done by the Observing Specialist and usually takes about five minutes at the beginning of the night. The OS can update fo cus throughout the night without interrupting science, except in cases of drastic ambient temperature or seeing changes in which a proper fo cus sweep is required. Any doubts about proper fo cus should be brought to the attention of the OS immediately. Please note that fo cus updates o ccasionally cause the optics to "twitch", which may result in your ob ject momentarily wandering off the slit. This is usually corrected within seconds on the next autoguiding update. If you suspect a twitch or bad guiding, let your OS know.

3.

Calibrations

The following figure shows a typical DIS bias. Bias levels are ordinarily quite stable over many months and their pixel value distributions are very Gaussian in nature. Be alert for any pattern noise in your biases. Such noise has been recurrent throughout the life of DIS; depending on the sensitivity of your observations, a large number of biases may be required to sufficiently average out bias patterns. But sure to inform your Obs Spec if you note any pattern noise in your biases.

Fig. 8.-- A DIS bias. Due to the thinning of the DIS chips, fringing is observed on the red-side chip. It is advisable to collect flat field images at both the beginning and end of your observing time in order to remove the fringing from your ob ject images properly. The instrument is sufficiently unstable mechanically such that fringes in images taken on one given night will notcorrectly line up with those taken on adjacent nights. It is therefore important to obtain an independent set of flats on each observing nights.


April Draft

8, Version 2004

­ 10 ­

Fig. 9.-- A typical DIS flat, taken with the medium-resolution red grating and bright quartz lamps. Note the fringing visible on the left side of the image.

Fig. 10.-- A typical DIS flat, taken with the low-resolution blue grating and bright quartz lamps.

Fig. 11.-- A typical DIS red chip HeNeAr wavecal taken with the medium-resolution red grating. Note that wavelength increases toward the left in this image. Arc lamp spectra for wavelength comparison are obtained via illumination of the primary mirror covers with discharge lamps mounted on the telescope truss. Two sets each of helium, neon and argon lamps are mounted on opposite ends of the truss for proper illumination of the mirror covers. The lamps can be controlled independently by the user either via an MC no de connection or the xlamps widget on tycho.

3.1.

Controlling Truss Lamps From An MC Node

Some observers maintain a connection to the MC to issue lower-level commands than those available in TUI. The truss lamps can be actuated directly from the MC with the lamps command. A list of available lamps and their on/off status can be displayed in the window by issuing a lamps list; each lamp has both a number assigned to it as well as


April Draft

8, Version 2004

­ 11 ­

Fig. 12.-- A typical DIS blue chip HeNeAr wavecal taken with the low-resolution blue grating. Note that wavelength increases toward the right in this image. a string describing the type of lamps. The lamps are commanded on and off with lamps off [lamp] and lamps on [lamp], where [lamp] can either be the number identifier or the string name.

3.2.

Controlling Truss Lamps With xlamps

An Xwindows-based widget has been written which allows the user to control the truss lamps through a simple, easy-to-use GUI. From an Xwindows (e.g., UNIX, Linux, Mac OS X) environment, ssh to the visitor1 account on tycho.apo.nmsu.edu with the -X flag, allowing for the graphical widget to be displayed on the user's lo cal machine. At the tycho prompt, type xlamps(optionally with a & to run the program in the background). A window should pop up which lo oks like the one in Figure 13.

Fig. 13.-- The xlamps widget. Control of the lamps is different lamp names. Red green indicates lamps on; for off. Status updates are done accomplished by left-clicking on the colored boxes labeled with boxes indicate lamps off, yellow indicates lamps cycling, and example, in the figure, the neon lamps are on and all others are automatically; in other words, what you see is what you get.

3.3.

How Many Of Which Kind?

The following tables suggest quantities and exposure times for various calibrations. Low and Medium Resolution Gratings


April Draft

8, Version 2004

­ 12 ­ Type Bias HeNeAr Arcs Flats High Reso Type Bias HeNeAr Arcs Flats N Exp 3- 5 3- 5 3 osure(s) 0 60 180

lution Gratings N Exposure(s) 3- 5 0 3- 5 60 3 180

Odd numbers are suggested to help median out spurious defects in individual images1 .

4.

After Your Shift Ends the instrument or Remark. In t, e.g., you are DIS. Ask your

Be sure to turn off any calibration lamps you have used, then hand over by simply quitting out of the DIS instrument control window in either TUI some circumstances, you may continue to use the instrument after your shif the first-half observer and the scheduled second-half observer is not using Observing Specialist for permission to do so if circumstances warrant.

4.1.

Data Storage And Retrieval

Your data will remain on tycho.apo.nmsu.edu:/export/images/your subdirectory for 7 days before being automatically deleted; observers with special circumstances requiring more than this length of time for data storage should contact Craig Lo omis (clo omis@apo.nmsu.edu) at APO to make arrangements.

4.2.

Data Reduction

A useful set of IRAF-based DIS reduction routines called distools has been put together by Gordon Richards (Princeton). See message 90 in the apo35m-general archive at http://www.astro.princeton.edu/APO/apo35-dis/msg.90.html.
NOTE: Pattern noise in DIS biases from the blue chip is a recurring problem. As of April 2004, the strength of this noise has been diminished by a factor of ten. However, because the noise is fixed spatially, it does not disappear upon median combining stacks of biases.
1


April Draft

8, Version 2004

­ 13 ­ 5. Further Information And APO Contact

For any further information regarding DIS, please contact Russet McMillan (mcmilan@apo.nmsu.edu).