************************************************************************ ***** THIS DOCUMENT IS FOR OBSERVERS AT THE 0.9M ***** ***** WHO ARE OBSERVING FOR CTIOPI ***** ***** (version 16 AUG 2009) ***** ************************************************************************ The most recent version of this document can be found on the GSU CTIOPI website at http://www.chara.gsu.edu/~thenry/CTIOPI/ctiopi.observing.YYYY.MMDD ************************************************************************ ***** WE HAVE BEEN ASSIGNED A STABLE ACCOUNT FOR CTIOPI, v12 ***** ***** ALWAYS USE THAT ACCOUNT! ***** ***** (password can be found inside the CTIOPI notebooks) ***** ************************************************************************ ________________________________________________________________________ CTIOPI HELP FOR THE 0.9M ________________________________________________________________________ Todd Henry thenry@chara.gsu.edu John Subasavage subasavage@chara.gsu.edu Adric Riedel riedel@chara.gsu.edu ************************************************************************ ***** ***** ***** FIRST STOP, LA SERENA ***** ***** ***** ************************************************************************ Upon arrival at the La Serena airport, Tito will most likely be meeting you at the airport to whisk you to the dorms in the AURA complex. If not, you can catch a cab, and ask them to take you to the AURA complex. Make sure to enjoy La Serena while you are there. Say hello to folks at CTIO you know, and make new friends with those you don't (yet) know. Always be sure to stop by to thank Ximena Herreros for her help coordinating your trip. Go to the beach! Have pisco sours! Buy souvenirs at La Recova! ************************************************************************ ***** ***** ***** NEXT STOP, CTIO ***** ***** ***** ************************************************************************ After checking in and having lunch, go up to the telescope and find the large plastic PIBOX (with a red top), which is kept in the 0.9m observing room. In the PIBOX, you should find three 0.9m setup notebooks for CTIOPI targets (0h-7h, 8h-15h, 16h-23h), the 0.9m logbook, a photometry notebook, DAT tapes, DVDs, as well as other useful items. Please return all items to the PIBOX after your run, as the next observer will need them. After your observing run, just leave the PIBOX in the 0.9m observing room. ________________________________________________________________________ HELP ON THE MOUNTAIN ________________________________________________________________________ Observer Support dial 421 Electronics Support dial 417 Computer Support dial 422 Medical Center dial 454 ________________________________________________________________________ WEATHER ________________________________________________________________________ To look at recent pictures of the sky, access the SASCA (SOAR All Sky Camera) website by typing 139.229.3.224 into the browser address line. To check weather in Chile and the rest of South America go to http://www.ctio.noao.edu/site/environment.html or http://www.ctio.noao.edu/environ/environ.html To check recent weather history values for temperature, seeing, etc., type "weat" or look at the website http://www.ctio.noao.edu/htbin/wwweat. ________________________________________________________________________ RADIO ________________________________________________________________________ As of AUG 2009, good radio stations (at least by Todd's standards) include 88.9, 100.9, 101.5 and 102.1 (Horizonte). The second one has been known to play the Pet Shop Boys, so listen at your own risk. ________________________________________________________________________ COMPUTER ADDRESSES AT THE TELESCOPES ________________________________________________________________________ 0.9m ctioa4 observer's computer with ARCON /ua41/v12 home directory on ctioa4 /ua44/v12 alternate home directory on ctioa4 /ua45/v12 alternate home directory on ctioa4 ctio36 additional computer /usr/u361/v12 home directory on ctio36 or /home/u361/v12 You can rlogin from anywhere on the mountain to another computer. I have a file of tricks called .alias where I keep some of my favorite quick commands. You can make your own such file and activate it by typing "source .alias" to make your commands work. If a machine locks up, you can try rebooting by holding the [STOP] and [a] buttons down simultaneously. Then, answer "sync" at the prompt. ________________________________________________________________________ REMOTE ACCESS ________________________________________________________________________ La Serena ssh -l v12 ctio.noao.edu 0.9m ssh -l v12 ctio36.ctio.noao.edu cd /ua41/v12 to get to the files from ctioa4 cd /ua44/v12 to get to the files from ctioa4 cd /ua45/v12 to get to the files from ctioa4 cd /u363/v12 to get to the files from ctio36 ________________________________________________________________________ TAPEDRIVE INFORMATION ________________________________________________________________________ 0.9m from ctio36 workstation (not ctioa4!) DAT is /dev/nst0 ________________________________________________________________________ GETTING STARTED, AND IRAF COMMANDS ________________________________________________________________________ A quick list of how to get information about the telescope and instrument setup: lpar telpars telescope parameters, e.g. focus lpar detpars detector setup parameters, e.g. pixels used, gain lpar instrpars instrument parameters, e.g. current filter positions lpar wheel1 assignments for filters in wheel 1 lpar wheel2 assignments for filters in wheel 2 lpar obspars observing parameters, e.g. observers, frame # A fairly comprehensive list of commands you can use for observing is given here: ccdinfo tells you how the detector is currently set up, but does not allow you to change anything --- to change a parameter, use [detpars]. ctiopi loads useful scripts for CTIOPI (note: smarts loads similar scripts for SMARTS observing, e.g. finishsmarts) detpars allows you to change parameters on the detector. You must use this if you are changing setups during the night. You must type "setdet force+" after using this. display is the command that puts the image in the ximtool box. It is best to [epar display] and make sure that the parameter "fill" is set to "yes" so that the image fills the ximtool (or saoimage) box. There are a million subtleties to [display] that you can pretty much ignore for data-taking purposes. To fit Ximtool display box, use "set stdimage=imt800" and stretch the box so that it barely exceeds the edges of the frame when the frame is displayed. doobs is used to set up a series of observations and is very useful for flats (take 11 exposures each in the filter sequence "v,r,i"). epar is the command you type to get inside each package to change parameters. You keep the changes and exit the package by typing d. filter in blue window, changes the current filters finishctiopi when observing for CTIOPI, makes .fits files, generates header files, and emails them. flpr resets window commands after aborting or stopping. Type it three times to be sure. focus is used to focus the telescope at the 0.9m. The charge is transferred along the CCD so that you get several star images in a line to inspect for focus. All you have to do is adjust the focus between each exposure and hit when you are ready for the next one. Remember to delete the file and 'epar obspars' to reset the frame count when you are done. hedit is useful if you misname the file -- use it to change the title to one of the following formats: STAR0088 at R for pi STAR0088 at R for phot STAR0088 at R for refraction STAR0088 at R for pi, phot, refraction STAR0088 at R --- bad frame imarith is used to add/subtract/multiply/divide images, i.e. basic arithmetic. imexamine is very useful when taking data. Type [imexamine] to get a cursor on the image in the imtool box. Type "r" to get a radial profile to see if a star is saturated (64,000 counts) or has enough counts, and to check the FWHM/seeing. Type "e" to get a contour plot to see if the focus is ok. Type "a" to get a readout of the peak value. Type "m" to get counts in a 5x5 box to get a sky noise estimate. Type "v" twice to get a plot along a vector where you mark the two ends. Type "s" to get a surface plot. NOTE THAT YOU MUST HAVE A SQUARE BOX FOR THE IMEXAMINE WINDOW TO LOOK AT THE SHAPE OF THE IMAGE --- A RECTANGULAR BOX WILL FLATTEN THE IMAGE AND MAKE IT LOOK OUT OF FOCUS WHEN IT IS NOT!!! imhead typed alone, this will give you short header names for each file taken and show them on the screen. Typed as "imhead f* > look" will make a file called look that lists all of the frames taken. If you [epar imhead] you can change "longhead" to "yes", d out of it, and then get all the details of the observation. Or, just type "imhead f001 long+" to see the full header of a single file. imrename renames the specified image. imstat is useful to see what the mean is in a flat/bias/image. lpar can be used to list the parameters without going inside the package. mkdir creates a directory. more repeat the previous observation. motor init in blue window, this reinitializes the motor controller; do this when you get a message that says "SMC not responding" motor stat in blue window, returns location of the two filter wheels observe is the command used to take data. Use "object" for a target, "dflat" for a dome flat, and "zero" for a bias. Use filter "cb" for dome flats and "dia" for objects. obspars gives basic observing parameters. Use to set the file name for the night or adjust the incremental number of the frame being taken. File names should simply be "f" offset displays the last image taken and computes the offset necessary to move a star selected by the cursor to a specified X,Y. "offset" first puts you in "imexam", then put the cursor on the star of choice, hit q, and it will then prompt you for the X,Y to move to. Type "q" to get out. pause allows you to stop integrating to let clouds pass, recapture guide star, etc. Use "resume" to start again. qlook displays the last image and puts you in "imexam" in quick look mode. resume to start integrating again after pause. setdet force+ locks in the setup of the CCD. You need to type this after using "detpars" stop to get out of a sequence of exposures. tchange can be used to adjust the length of the integration once you have already begun integrating. "tchange -30" would shorten the exposure by 30 sec, "tchange 30" (no + sign!) would add 30 sec to the exposure. tcs info (must be typed in the Arcon Console window, NOT the Arcon Acquisition window) tells you if the TCS (telescope control system) is talking to the observing computer. temp sets the telescope temperature. test lets you take a frame but does not increment the frame number. This is useful for centering an image, testing integration time, etc. (i.e. any frame that you are likely to throw away). weat gives a weather update on the mountain. zero takes a bias frame with exposure of zero seconds. ************************************************************************ ***** ***** ***** TIME TO OBSERVE ***** ***** ***** ************************************************************************ ________________________________________________________________________ BEFORE OPENING ________________________________________________________________________ You will not have a night assistant at the 0.9m. So, you are on you own for most of each night, although Telops can provide support when needed (Arturo Gomez). Instruction on how to open/close the dome, move the telescope, etc. can be found in the black notebook "0.9M TEL. MANUAL". The directions are given on the white pages covered with plastic near the front. ________________________________________________________________________ A FEW TIPS ABOUT THE ARCON ACQUISITION WINDOW AT THE 0.9M ________________________________________________________________________ cl> d TO SAVE ANY CHANGED PARAMETERS IN IRAF cl> flpr 3 TIMES TO CLEAR ANY ERRORS THAT MAY SCREW UP ARCON cl> ctiopi ANY TIME YOU REBOOTED ARCON, TO RELOAD THE SCRIPTS FOR CTIOPI OBSERVING If you lose the TCS header information, you can type "tcs info" in the blue ARCON Console window (NOT in the ARCON Acquistion window). If you get a bunch of numbers, you are ok --- only the last frame had a problem. It is wise to take a test frame after any TCS problem to make sure the TCS header is being written. Loggin in/out of ARCON is the usual solution. The definitive method involves turning off/on the entire TCS using the switches below the guider. ________________________________________________________________________ GETTING STARTED, CCD CAMERA SETUP ________________________________________________________________________ First things first --- it is usually quite dry at CTIO, so TURN ON THE HUMIDIFIER (high setting) in the observing room. This will help prevent ARCON crashes due to static charge. As soon as you login, you should load various observing commands you will need. You must be in the home directory (the "top" directory). If you are using the v12 account for CTIOPI observing, type "ctiopi". Make sure the chip is setup correctly by typing "ccdinfo" in the ARCON Acquisition window. You should get the following parameters: gain = 2 (gain 1 saturates at 40K as of March 2009) xstart = 513 ystart = 513 xsize = 1024 ysize = 1024 amplifiers = "ul" detname = "Tek2K_3" This gives us one quarter of the array, which equals 6.8 arcmin on the sky, centered on chip center. The "ul" amplifier reads out in 30 seconds via one amplifier, and provides low noise. To change the parameters, type "detpars", fill out the correct settings, followed by "setdet force+" to activate the new chip settings. ________________________________________________________________________ DIRECTORIES AND FILENAMES ________________________________________________________________________ Each night has a separate directory. For working through the night, you can name the directory "n1" for the first night, etc. Eventually, the name of the directory should be YYYYMMDD-09, i.e. 20060201-09 for 01 FEB 06. At the end of the night, the directory name MUST match the local date at the BEGINNING of the night. To move into the desired directory to take data for the night, type "cd n1". The name of the files MUST have the single letter name "f". Type "epar obspars" to setup the file name and number 1 (at the start of the night). Once you have entered this, type " d" to save the parameters in IRAF. Numbers will be appended during the night, so that the first frame will be f001, the second will be f002, etc. ________________________________________________________________________ CALIBRATION FRAMES --- BIASES AND DOME FLATS ________________________________________________________________________ Only one set of calibration frames is taken each night. The set of calibration frames should be for either the full-chip or quarter-chip mode, BUT NOT BOTH. Science images may still be taken in both setups during one night, but calibration frames from a different night will be used during data reduction. BEFORE taking the biases and flats, repeat three times the command "preview 1". This takes a frame of 1 sec exposure, but does not save the image. This "flushes" the CCD from any remaining charge traps or electronic noise accumulated during the CCD idle time. For both the full-chip and quarter-chip setups, bias frames and VRI dome flats should be done. You should take 17 bias frames, 11 V dome flats, 11 R dome flats, and 11 I dome flats. This makes a total of 50 calibration frames. Before taking any type of calibration frame, it is best to take a test frame to see if everything is working as expected. Type "imstat f*" to get the number of counts in frames to see if everything is ok, or load them in qlook and use the v command to look at the counts. BIAS FRAMES Take bias frames before flat frames because the flats take a long time, during which you can go to dinner. To take the bias frames, go upstairs to the telescope and push the dark slides in --- there are two of them and they have gold-colored "handles". Then, back in the computer room, type the following into the ARCON acquisition window: > observe Exposure type (|zero|dark|object|comp|pflat|dflat|sflat|focus): zero Number of exposures to take: 17 Title of picture: bias You should get about 1400 to 1500 counts. DOME FLATS Typically, the Telops folks will point the telescope to the white spot for you to take dome flats. If for some reason you want to take flats on your own (in the morning, for example), you can do so by setting the telescope to zenith, then turning off the auto dome, dome tracking, and telescope tracking (but leave the drives on). Then, set the slew position to be hour angle 02H 28M west by subtracting that amount from the ST (sidereal time), and set the DEC = -28 00 00. That should point the telescope at the white spot. Go upstairs and check. As of August 2009, we will be using only the oV = old V = Tek#2 V = cracked V filter. However, the nV = new V = Tek#1 V = replacement V filter will be the standard one used for SMARTS programs because most people (a) don't care which V filter is used and (b) we don't know for sure if the crack in the corner affects FULL CHIP observations. The filter wheel that holds the standard optical filters should therefore be in the following configuration for the 8 available slots: slot1 slot2 slot3 slot4 slot5 slot6 slot7 slot8 named u b nv r i ov U B nV R I oV - - Tek1 Tek2 Tek2 Tek2 cracked ************************************************************************ * ASK THE TELOPS FOLKS TO CLEAN THE FILTERS AND DEWAR WINDOW, AND * * WATCH THEM DO IT SO YOU CAN BE SURE WHERE THE CRACKED V FILTER IS ! * ************************************************************************ For dome flats choose an exposure time that gives about 30000 counts in each filter (a range from 20000 to 40000 counts is acceptable). Integrations of 120-180 sec are usually fine, giving about 30000 counts in the flats (highest in V, lowest in I). To take dome flats, turn on the lamps with the switch in the computer room, then go back upstairs and pull the dark slides out. While you are up there, check that all three lamps are on. If they aren't, call the the Telops folks. ************************************************************************ * USE THE COLOR BALANCE FILTER "cb" IN FILTER WHEEL #1 TO BALANCE * * THE QUARTZ LAMP EMISSION TO SKY COLOR * ************************************************************************ * CAREFULLY CHECK THE DOME FLAT COUNTS FROM THE oV,R,I FILTERS * * TO BE SURE YOU HAVE THE CORRECT FILTERS. * * FOR 120 SEC ON 2009.0817, THE FOLLOWING WERE MEANS FOR 5 FRAMES EACH * * * * U 2018 5.5 % * * B 20247 55.4 % * * nV 35417 97.0 % * * R 31931 87.4 % * * I 27694 75.8 % * * oV 36517 100.0 % * * * * SO, IT IS IMPORTANT TO DOUBLECHECK WITH THE TELOPS FOLKS TO SEE THAT * * THE oV FILTER IS WHERE YOU THINK IT IS BECAUSE IT WILL BE DIFFICULT * * TO TELL FROM FLAT FIELD COUNTS ALONE * * * ************************************************************************ Back in the computer room, use the command "doobs" to take the flats. > doobs Exposure type (|object|dflat|sflat|): dflat Number of exposures to take in each filter: 11 Title for pictures: dflats List of filters in wheel1: cb List of filters in wheel2: ov,r,i List of exposure times: 120,120,120 For both full-chip and quarter-chip setups a set of 33 dome flats at VRI takes about 2 hours. If ARCON crashes during the dome flats, you can restart just the parts you missed (remember to "epar obspars" to reset the frame counter) DON'T FORGET TO TURN THE LAMPS OFF WHEN YOU ARE DONE. ________________________________________________________________________ WRITING DOWN THE OBSERVATIONS ________________________________________________________________________ For CTIOPI observations, keep the log in the logbook from the CTIOPI box, as has been done since 1999. ________________________________________________________________________ AFTER OPENING --- REAL OBSERVING! ________________________________________________________________________ 1. Turn on the fans over the primary mirror using the green cylindrical knob in the observing room (turn it to the vertical mark on the dial). 2. Make sure that the large fan that ventilates the dome is on (and the door to upstairs is open). The switch for the large fan is in the electrical control box in the northeast corner of the first floor. It is marked with a piece of white electrical tape. 3. Make sure the dome lights are off, the windscreen is down, the slit is open, and the platform has been lowered. 4. Back in the control room, make sure the UTs are synched between the DFM controller (Windows machine) and the display above the guider computer. 5. Turn Tracking, Drives, Auto dome, Aux Track, and Dome Track on. Make sure Track/Aux Track (green) is set to Track. 6. Choose a star from the bright star catalog to zero point for the night. Take exposures and center the star on the chip, using "offset" as needed. X. You may also want to check the chip alignment as follows, although we don't do this any more --- the chip rotation has been minimal throughout CTIOPI. Check the CCD alignment to be sure it is oriented properly. Simply turn off the tracking and take a ~60 sec image to check. [display] and [imexam] the image. Go to one extreme of the trail. Position the cursor in the middle of the trail and type "k". This will plot a cut through the trail, and give you the center of the intensity profile on the trail. In order for the command "i" to work properly, zoom the image until you see individual pixels on the frame. Write down the center of the trail. Then go to the other extreme of the trail on the CCD, and type "k" again. Take the difference between the two values for the intensity center of the trail. If you are within ~5 pixels of directly aligned (i.e., if the above difference is smaller than 5 pixels), that is good because 5 pixels corresponds to 0.28 degrees at the 0.9m (1024 pixels). We probably can't rotate the instrument to better precision. 7. Initialize the telescope using the pull-down on "Telescope" in the Telescope Control System window. Adjust three things, as needed: (a) time, (b), zero point (fixes the pointing of the telescope --- be sure you are using the correct epochs!), and (c) dome initialization. 8. To see if all is well with the telescope pointing, choose a second star from the bright star catalog, slew to it, and take a frame. You can skip this step and go to the next one if you are confident that the telescope is pointing properly. 9. Go to the first target star. 10. Focus using [focus], which relocates a star image on the CCD so that you get a series of several star images at different focus values. Choose seven integrations of 10 seconds in the filter of choice with focus increment 50 (30 for seeing of 1.2 arc-sec or better). You have to change the focus by hand at the 0.9m, so change the focus value and then hit to take the next exposure. V and R are expected to have the same focus at the 0.9m. The focus in I is approximately 35 units to larger numbers. When you change focus from a larger value to a smaller value, remember to offset the focus some 400 units "in" (e.g. 18600) and then go "out" (e.g. 19000) to the desired focus value. This way you will always be moving the focus against gravity, to prevent backlash problems that are known to be present in the focusing mechanism. The focus is a strong function of temperature, and also of telescope pointing in the sky. It is therefore best to check the focus for each target before taking observations that you plan to keep. To check the image shape type [imexam], put the cursor on a non-saturated but well-exposed single star, and type "e" --- this will display a contour plot on the graphics window, look for elongations in X or Y. Take special care to be sure that the focus is excellent --- the parallax reductions are much better with round images! ************************************************************************ * * * DELETE ALL FOCUS FRAMES TO MAKE DATA ORGANIZATION EASIER * * * ************************************************************************ ************************************************************************ * * * 0.9m TARGETS HAVE R = 9 to 16 * * * * MAXIMUM ASTROMETRY INTEGRATION TIME SHOULD BE 600 SEC * * MINIMUM ASTROMETRY INTEGRATION TIME SHOULD BE 30 SEC * * * * MAXIMUM PHOTOMETRY INTEGRATION TIME SHOULD BE 1200 SEC * * MINIMUM PHOTOMETRY INTEGRATION TIME SHOULD BE 5 SEC * * (although there are a few exceptions to the minimum time) * * * ************************************************************************ ________________________________________________________________________ WHEN ARCON CRASHES (and it will): ________________________________________________________________________ Symptoms of an ARCON crash are: * a painful zap * the CCD preview monitor has gone completely black * the countdown timer freezes * trying to take pictures spits out endless errors. The solution is to right click, exit from windowing system, and log out of ctioa4. When you log back in as v12 (password can be found inside the CTIOPI notebooks), choose option 2 (IRAF with ARCON acquisition) and once OpenWindows starts, answer yes when it asks if you want to synchronize parameters. From there, run ctiopi in both iraf windows to load the ctiopi tools, and move into the current night's directories in both iraf windows. ________________________________________________________________________ GOLDEN RULES FOR ASTROMETRY OBSERVING ________________________________________________________________________ MAKE SURE FILTER 1 IS dia, NOT cb MAKE SURE FILTER 2 IS ALWAYS ov, r, OR i 1. Plan your observing strategy, depending on the quality of the night. We typically throw out frames with seeing > 2.5". CONDITIONS SPECIAL CONCENTRATION < 1" seeing, clear faintest targets, close binaries 1-3" seeing, clear regular targets > 3" seeing, clear photometry < 1" seeing, cloudy regular targets, close binaries 1-3" seeing, cloudy regular targets > 3" seeing, cloudy sleep! High humidity often comes with poor seeing. 2. In general, use the hours between 10pm and 2am for ASPENS stars, source investigation, or binary star programs, and the rest of the night for parallaxes. Middle of the night parallax frames are useful for high declination targets, proper motion determination, and perturbation searches (e.g. ASPENS stars). 3. Stay within 30 minutes of the meridian for parallax frames. This reduces the need for differential refraction corrections. Choose the next target when it is 20-60 minutes to the east to allow time to set up. 4. Choose an object. Make sure to check the observing list to see if more frames are needed, and what kind of frames are needed --- evening or morning. The goal is to have at least 30 evening frames and 30 morning frames on each target, and to keep the number balanced. There will be few "golden opportunities" to observe a target. These are times when the star is at high parallax factor and crosses the meridian within an hour of opening or closing. Try to take at least 5 frames when you get a chance like this. Consider observing it again several nights later if you have a long observing run. 5. Check the notes on individual objects (ctiopi.objectlog) to see if there are any special observations needed for that object. For example, close doubles may require some special short integration frames. Plan your set of observations accordingly. ************************************************************************ * * * FROM HERE ON OUT, YOU CAN USE THE FFF RULE --- "FORMATION" TO GET * * THE PROPER POSITIONING ON THE CHIP, "FOCUS" TO GET GOOD IMAGES, * * "FLUX" TO MAXIMIZE THE NUMBER OF COUNTS IN THE STAR THAT SETS THE * * EXPOSURE. * * * ************************************************************************ 6. If it is the first observation of a star, take short integrations to find the target and set up a reasonable reference star field. Choose a set up that has at least 5 reference stars. You can use the IRAF task [test] instead of [observe] to take frames but not increment the frame number. Remember that the astrometry is best with the target near the frame center. 7. If it is not the first observation of a star, place the parallax star within 5 pixels (2.0 arcsec). In the red Arcon window, use [offset] to calculate the offsets needed to put it in the final location. 8. Estimate an appropriate filter/integration time to reach about 55,000 counts for the target star or brightest useful reference star. 64,000 counts is saturated (use imexam and "r" to get the peak value). The star that sets the exposure should be marked on the finder, or a comment at the top stating "pi star sets the exposure." R is the first choice of filter, although V might be used for bright stars and I for faint red stars. If the star has already been observed, the filter choice should be clearly marked on the setup frame, as well as a guideline for the integration time. USE THAT FILTER ONLY, UNLESS DOING PHOTOMETRY. IN GENERAL, 30 SECONDS IS THE MINIMUM INTEGRATION ALLOWED, IN ORDER TO PREVENT ASTROMETRIC ERRORS CAUSED BY SEEING EFFECTS ON THE STAR IMAGES. We have a few bright stars that require shorter integrations that are important enough for us to try anyway. 9. Check the positioning, focus, and counts. Using [imexamine], type "r" for a radial plot to check the counts. Remember that you must choose the integration time so that the faintest useful reference star has SNR > 100, while the brightest useful star (target or reference) is not saturated, but has about 55,000 counts at the peak. To estimate the noise level type "m" in imexamine on a part of the image with no sources --- the standard deviation of the pixel values is an estimate of at least the sky noise. Maximize the counts in the target star or the brightest useful reference star. Be willing to let possible reference stars saturate and be discarded because the final error in the parallax is dominated by the SNR of the target star. ************************************************************************ * * * NOTE THAT YOU MUST HAVE A SQUARE BOX FOR THE IMEXAMINE WINDOW TO * * LOOK AT THE SHAPE OF THE IMAGE --- A RECTANGULAR BOX WILL FLATTEN * * THE IMAGE AND MAKE IT LOOK OUT OF FOCUS WHEN IT IS NOT!!! * * * ************************************************************************ 10. Find an appropriate guide star on the AUTOGUIDER. If you are having problems with the AUTOGUIDER, you can reboot it using login: camera2000 password: ev500gate The autoguider controls are simple. Use the arrow keys to move the white box around a bright guide star, hit F9 to start guiding, and F8 to stop guiding. The 0.9m's tracking is usually perfect for exposures of 300 seconds or less. The 'Camera Head Controller' window is useful for the following items. It is accessible from the CTIO PC GUIDER 5.0.1' window, under the 'Window' menu. If the arrow keys control the mouse, click the 'snap' button. If you cannot see stars in the autoguider, set 'Erase' to 'off'. Alternately, try LOWERING the gain on the guider camera. Try to avoid integration times longer than 1000 ms. 11. Sometimes you need to change the display parameters in IRAF in order to see, e.g., a faint source next to a really bright star. You can do this by typing "display xxx 1 zr- zs- z1=aaa, z2=bbb" in the red Arcon window, where xxx is the frame name. You get aaa and bbb by setting them, typically, to the z1 and z2 values displayed when doing a normal "display xxx 1" (i.e., you have to do this first in order to get the approximate values for aaa and bbb). This will help supress the large wings from a nearby bright star, while keeping the target visible. For more info on controlling z1 and z2 do "help display" in IRAF. 12. Take a parallax frame. Name the star in the frame header with all capital letters and no blanks EXACTLY AS IT IS GIVEN ON THE OBSERVING LIST, i.e. WT0001, GJ0752B, LHS4009, G041-014AB, CE207-061, SCR0000-0000. CONSISTENCY HERE IS ABSOLUTELY CRUCIAL FOR CATALOG GENERATION SOFTWARE! The maximum number of characters for target names is 16. ********************************************************************* * * * ALL FRAMES SHOULD HAVE NAMES IN THE FOLLOWING FORMAT: * * * * STAR0088 at R for pi * * STAR0088 at R for phot * * STAR0088 at R for refraction * * STAR0088 at R for pi, phot, refraction * * STAR0088 at R --- bad frame * * note that you should really have ZERO of that last type * * it is better to just delete those frames during the night * * * ********************************************************************* 13. IF THE SEEING IS CONSISTENTLY MUCH WORSE THAN 2" IT IS PROBABLY BETTER TO BEGIN DOING PHOTOMETRY INSTEAD OF ASTROMETRY, IF IT IS CLEAR. See the photometry directions in the next section. 14. Take about 5 frames of the field for normal CTIO targets, unless otherwise specified in the observing list or on the setup finder (fainter targets often specify only two or three frames). Try to time the observations so that the target passes through the meridian some time during the sequence. Check for saturated stars in the raw frames using [imexamine]. Note any saturated target or reference stars in the logbook. Keep all frames that have the parallax star near the correct position, even if it's just a setup frame more than one hour from the meridian, or if a reference star is saturated. 15. If it is the first observation of a star, print out a finder of the final setup for other observers to use. To make a printout, configure the Ximtool window so that there is very little black around the edges of the frame --- when you print out the setup frame there will then be lots of room to write at the top and bottom. Use the drop-down menu on the File button on the Ximtool window and choose Print Setup. Change the Image Scale from 100% o 85%. Then, in the top box, type lpr -Pnp36 (for the 0.9m printer) and hit the Print button in the lower left. Once it's printed, write the necessary information on the setup chart. An example is given in the front of the 0.9m notebook. Most important are the pixel coordinates for the target star and the filter selection! 16. IF THE SEEING IS GOOD AND THE SKY CLEAR (OR CLOSE TO IT) CONSIDER DOING PHOTOMETRY. SEE THE DIRECTIONS IN THE NEXT SECTION. ________________________________________________________________________ GOLDEN RULES FOR PHOTOMETRY OBSERVING ________________________________________________________________________ MAKE SURE FILTER 1 IS dia, NOT cb 1. It must be clear! No clouds of any kind are tolerated. This is the most anxiety-ridden decision of observing. Don't forget to use weather maps on the web to try to predict the future. Also, don't be afraid to ask around at dinner to see what others think. Best of luck! 2. There is a separate PHOTOMETRY notebook that gives "setup" charts, finder charts, and magnitudes for standard stars. In the same notebook there is also a cookbook of observing methods to get good photometry. The setup charts for the standard stars are very useful to see how to arrange some fields on the 0.9m chip, where they have already been observed. 3. THE PLAN FOR 0.9M TARGETS IS TO DO VRI PHOTOMETRY FOR ALL STARS AT THE 0.9M. In all cases, you must position the telescope carefully to be sure that the target and all reference stars are on the chip. You must also be sure that neither the target star nor any of the reference stars saturate! But, the counts should be at least 30,000 for the target star or brightest useful reference star. There are some rather faint stars where it would take hours to get that many counts. A good rule is that the longest integration in any filter for a photometry frame should be 15 minutes --- we will take whatever counts we can get for these faint stars, but integrate no longer. 4. Photometry should be done all of the night when it is photometric. You might do a few parallax frames at the beginning and end of the night, when you're already pointed at a star that also needs photometry. 5. If you are doing photometry of a star that's already been set up for parallax, try to align the field according to that setup, so the relevant frame can also be used for parallax. ************************************************************************ * * * YOU MUST OBSERVE EACH FIELD AT LEAST THREE TIMES FOR THE STARS * * THE FIELD TO BE TRULY USEFUL FOR EXTINCTION CURVES. * * * ************************************************************************ 6. Take a minimum of 20 total observations of standards each photometry night in each of the VRI bands. These 20 observations must include one or two very red stars done 3 times each for the color transformations. The red standards are marked on the standard star observing list. One way to easily get the 20 total observations is to do a Landolt field that has several stars in it, e.g. one Landolt field with 5 stars done at three different airmasses = 15 observations. 7. We MUST do extinction curves every night we want to do photometry. Arlo Landolt has data illustrating that the CTIO extinction at 2.00 airmasses changes from 10% to 33% regularly. So, we cannot assume a standard extinction for a given night. To calculate extinction corrections, observe one Landolt field starting at airmass up to 2.00 to the east. Catch it 3-5 times as it moves up in the sky during the night. 8. THERE IS A PHOTOMETRY WORKSHEET IN THE NOTEBOOK THAT WILL HELP IN PLANNING OBSERVATIONS. Use it to make your life easier. ________________________________________________________________________ GOLDEN RULES FOR REFRACTION OBSERVING (RARELY DONE) ________________________________________________________________________ MAKE SURE FILTER 1 IS dia, NOT cb 1. The seeing must be better than 2.0 arcsec. 2. It works best if a series of refraction frames on a field is done in a single night. You can probably do only one to three fields in one night. It is a very time-consuming process because you have to do refraction frames at VRI and and photometry (VRI), including standards, in the same night. 3. Observe each star 4 or 5 times for a series. One observation should be made about 4 hours east, 3 hours east, 2 hours east, 1 hour east and on the meridian. Each observation should be done as if it were a parallax frame --- properly positioned, focused, and with about 55,000 counts in the target star or brightest useful reference star. A complete sequence of frames on a field would be: VRI at 4 hours east, VRI at 3 hours east, VRI at 2 hours east, VRI at 1 hour east, VRI on the meridian --- the last meridian set is crucial, so plan accordingly. 4. It is possible to use the refraction frames taken in the parallax filter for parallax, as well as doing photometry for the field because you must do standards as well. 5. There is a refraction worksheet in the notebook that will help in planning observations. Use it to make your life easier. ________________________________________________________________________ AT THE END OF THE NIGHT ________________________________________________________________________ THERE ARE THREE THINGS THAT NEED TO BE DONE AT THE END OF THE NIGHT: 1. *** SHUT DOWN THE TELESCOPE *** In the computer room: * Turn track and dome track off (TCS black box below guider) IF YOU DO NOT, THE TELESCOPE WILL CONTINUE TO MOVE WHILE YOU'RE FILLING THE DEWAR. * Turn down the fans over the mirror (dial above ARCON) In the hallway: * Turn off dome fans (in the circuit breaker box) In the dome: * Slew to the cover position (button on Windows computer), replace the cover. * Close the slit and windscreen (old console, or hand paddle) * Slew to Zenith position (button on Windows computer) * Fill the dewar * Push in the two dark slides * Make sure the platform is down. * Turn off master power * Turn off all the lights In the computer room: * Turn drives, autodome, and master power off * Fill out end-of-night report * Log out of ARCON * Turn off all the lights 2. *** CHECK HEADERS *** Check header titles using [imhead f*] to make sure they are correct --- use IRAF task [hedit] to change titles, just type "title" when it asks which fields you want to edit. In particular, make sure that the names of the objects are perfect --- the software that does the catalog requires exact matches. Delete unwanted frames, such as focus frames. To make it easy to identify what kind of frames were taken (pi, photometry, refraction) or bad frames (bad) during reduction, make sure all frame titles have the following format: STAR0088 at R for pi STAR0088 at R for phot STAR0088 at R for pi and phot STAR0088 at R for refraction STAR0088 at R --- bad frame 3. *** MAKE .FITS FILES, SEND HEADERS *** If needed, in the home directory for CTIOPI on the v12 account, type "ctiopi" to load the CTIOPI scripts (originally written by Pat Seitzer). First, rename the directory so it has the format YYYYMMDD-09 i.e. 20060201-09. Then, move into that directory. Type "df" to check that there is enough room to write .fits files for all the frames you took. Each .fits file on the 0.9m takes up 2301120 bytes. Type "finishctiopi". You will be asked for a file name --- USE THE EXACT FORMAT OF THE DATE AND TELESCOPE, USING THE LOCAL DATE AT THE BEGINNING OF THE NIGHT, i.e. "20060201.09" (note there is a period not a dash in this name). The "finishctiopi" task makes .fits files (takes about 2 seconds/file on ctioa4, about 16 seconds/file on ctio36), generates header files and emails them automatically to GSU. The only questions you will have to answer are about who was observing, and how many hours were observed/lost due to the weather and the telescope. The files will be renamed with the date and telescope so that they will be 20060201.09.001.fits, etc. Therefore, the first file taken on 01 FEB 06 is found in directory 20060201-09, has the name f001.imh during the night, but has the name 20060201.09.001.fits when written to tape. THE FINAL NAME MUST BE IN THE FORMAT 20060201.09.001.fits !!! Once the .fits files have been written to tape, DO NOT IMMEDIATELY DELETE THEM. The GSU folks will get the headers via email automatically, and will then rsync the data back to GSU. Do not delete the files until someone from GSU confirms that the .fits files have been rsync'd. If you wish to rsync them yourself (not recommended): * Open up a terminal on ctio36 and cd to your folder (should be /ua41/v12/YYYYMMDD-09) * type the following: rsync -avz *.fits @ssh.chara.gsu.edu:/nfs/recons/disk3/SMARTS/YYYY.MMDD/ this will only work if you have permission to write there. (recons group). It is also prone to hanging, so if no file has transferred for a while, c and retry the command. It will pick up where it left off. You will probably want to go to bed rather than wait for it to finish. Note that if you do this, GSU people will not be able to restart it for you. ________________________________________________________________________ AT THE END OF THE RUN ________________________________________________________________________ *** BURN A DVD OF ALL YOUR DATA *** 1. On ctio36, open the Fedora menu, go to 'Sound & Video', and open K3b. 2. At the bottom of the resulting window, click 'New Data DVD Project' 3. In the upper left, go to root, then find /ua/v12, drag and drop all the relevant folders to the bottom part. You don't need the hidden files. 4. Go into each folder in the lower panel and delete the .imh and headers files. Keep the .fits files. Sorting by type (click the header) helps. To delete, click on a file and hit the delete key. Hold down the key and keep clicking to highlight lots of files you want to delete. 5. Put a blank DVD in the top of the ctio36 computer in the back room. It's the tower with a black mesh front that says "Cooler Master" at the top. The DVD tray button is near the top. 6. Click burn, make sure it doesn't say "Please insert an empty or appendable DVD+/-RW medium" at the top of the window. 7. On the filesystem tab, make sure 'Linux/Unix + Windows' is selected, and enter a volume name of the form 01-07 FEB 2006 8. On the writing tab, check 'create image' and 'remove image'. This will transfer all data from ua45 to ctio36 temporarily, so you don't have to worry about transfer speeds while writing the DVD. Checking 'verify written data' is optional but recommended. It will make sure all data on disk matches what's on the computer, but it will almost double the time it takes to burn the DVD. 9. Now click 'burn' at the bottom of this window, and wait. The process should take roughly 30 minutes; the longest step should be image creation. 10. Label the DVD with a felt-tip pen (ex. 13-19 AUG 2009). If you are not with GSU, put the finished DVD in a case, and leave it in the PIBOX. Someone at GSU will collect it. If you are from GSU, check the PIBOX for burned and labeled DVDs to bring home with you. ************************************************************************ * * * DETAILS FOR WRITING TO TAPE --- AS OF 2009, WE USE DVDS * * * ************************************************************************ X. *** MAKE ONE COPY OF DATA TO DAT TAPE *** First, clean the tape drive! In the PIBOX there is a DAT cleaning cartridge. It has a green strip on the side, brand name Imation, and it says "Head Cleaning Cartridge" on it. Prior to backing up the first night of data, use the cartridge to clean the DAT drive. You just put it in, it cleans the heads automatically, and then it is ejected. The whole process takes about 30 seconds. It lasts for 50 cleanings, so make sure that when you use it, you put a date on the cartridge so we know when to get a new one. Write one copy of data to DAT tape using the ctio36 workstation. Move to the disk where the data resides (for example, cd /ua41/v12/20060201-09), then get into an xterm window mt -f /dev/nst0 status (to check status) tar -cvf /dev/nst0 *.fits (to write, watch it start writing) mt -f /dev/nst0 rewind (to rewind) note 1: if you get a "write protected" message, go to IRAF and allocate the device [allocate mtd] and then go back to UNIX and proceed note 2: it takes about 12 seconds per file to write data to tape YOU SHOULD BE ABLE TO FIT SEVERAL NIGHTS OF DATA ON ONE DAT TAPE AT THE 0.9M BECAUSE YOU CAN FIT A TOTAL OF 869 REGULAR SIZED CTIOPI FRAMES ONTO A SINGLE TAPE. If you took any full chip frames, a lower number of frames will fit on a tape. To add files to a tape that already has data on it: mt -f /dev/nst0 eof (eof = end of files) tar -cvf /dev/nst0 *.fits or, if you are writing more than one night at a time to tape, don't remove the tape from the drive, move to the directory for the next night's data, and just type again: tar -cvf /dev/nst0 *.fits To check that data has been written to tape: tar -tvf /dev/nst0 (for night 1 data) mt -f /dev/nst0 fsf (to get to night 2 data) tar -tvf /dev/nst0 (for night 2 data) mt -f /dev/nst0 fsf (to get to night 3 data) tar -tvf /dev/nst0 (for night 3 data) mt -f /dev/nst0 rewind (to rewind) If for some reason you want to unload data from a tape: tar -xvf /dev/nst0 (for night 1 data) Label the tapes using the following concise format. DO NOT WRITE ANYTHING ELSE ON THE TAPE LABEL. If you are observing on the night with the local start date of 01 FEB 2006, write 01 FEB 2006 on the tape, NOT 01-02 FEB 2006! Examples: For one night of CTIOPI CTIOPI 0.9m 01 FEB 2006 For three nights of CTIOPI CTIOPI 0.9m 01-03 FEB 2006 We are writing one copy of data to tape. The tape should be left in a tapebox in the PIBOX labeled "TO GSU ..." for pickup by someone from GSU --- it does NOT need to be mailed. If YOU are from GSU, please pick up all the tapes that belong at GSU and haul them back. We are ftp-ing the data back to GSU from the telescope, but still want to have tape copy, just in case. ________________________________________________________________________ NON-SIDEREAL TRACKING ________________________________________________________________________ Non-sidereal tracking is a three-step process. 1.) Confirm whether you have the non-sidereal tracking rate, or the CORRECTION to sidereal tracking rate. 2.) The tracking rates are controlled on the DFM control window (Windows computer) under the Telescope -> Rates menu option. The entire non-sidereal rate should be entered in the bottom set of boxes, not just the correction. Hit apply. For reference, the sidereal tracking rate is 15.0411 arcsec/sec (RA) and 0 arcsec/sec (DEC) 3.) On the black TCS control box, flip the green tracking switch to 'Aux Track' ************************************************************************ ***** ***** ***** THE END ***** ***** ***** ************************************************************************