Chapter 4- An Imaging Cookbook

  1. The OBSERVER System
  2. Starting OBSERVER
    1. Starting OBSERVER for use with OFFSET_RUN files
    2. Summary of how to start up OBSERVER
  3. Ending OBSERVER
  4. Running OBSERVER
    1. Setting up the CCD
    2. Types of exposures: RUN vs DUMMY vs GLANCE
    3. Taking Data
    4. Other Commands
    5. When things go wrong
  5. Typical Observing Procedures
    1. Flat Fields
    2. Focus Sequences
    3. Standard Stars
    4. Bias Frames
    5. Pointing Stars
  6. Riding in the Cage
  7. A Typical Night
    1. In the Afternoon
    2. At Sunset
    3. After Twilight
    4. In the Morning
  8. Preparing for your Run
  9. Operational Stuff
    1. Finding your Support Astronomer
    2. Finding the Technician-on-Duty
    3. The Night Assistant
    4. Observing Limits
    5. The Dome
    6. Fault Reports
    7. Observing Run Report



Introduction The Telescope & Optics The Detectors
The Imaging Cameras An Imaging Cookbook The Data you Take Away
Exposure times OFFSET_RUN files CCD Windows Data Catalogs
On-line Reduction Filters Flat-fields Blank Fields Orientation Shutters

Back Contents Next

Last Contents Next

4.1 The OBSERVER System

While the hardware which is used for f/3, f/1 and f/8 imaging may differ, the detectors are all run by the same software - the OBSERVER system.

Figure 4.1 - Schematic Diagram of the Hardware used in Imaging

While OBSERVER is a fairly complicated system in terms of the interactions it controls, it is actually quite easy to use for imaging applications. Because all our current imaging systems have either manual filter wheels and/or guide probes, or (in the case of the f/1 system) a separate filter control system, there is essentially nothing  for the astronomer to control with OBSERVER apart from the CCD. All CCD operations are controlled from a single VT102 terminal.

CCD data is readout by one of two controllers, called either CCD_1 or CCD_2. The particular controller being used for your run will be written on the small whiteboard on the control panel. From there the data is stored in the Large eXternal mEMory (or XMEM) (see Fig 4.1), and saved onto the disk systems of a MicroVAX 4000. These disks are also NFS-mounted on a Sparc 10. Both the Sparc 10 console, and several X-terminals which can be connected to the Sparc 10 and the MicroVAX 4000, are available for the use of observers in the control room for on-line examination of their data.

Occasionally the XMEM may hang up and need to be reset. This can be done by pressing the black reset button on the panel behind the observing terminals.

The MicroVAXes data archive disk has a capacity of about 2 Gb on both DISK$INST and there is a further 2Gb of space on the `non-archival'  DISK$DATA.

4.2 Starting OBSERVER

  1. The OBSERVER system is started by connecting to the MicroVAX (aat40a) from any suitable X-terminal in the AAT control room. You should then log on to the MicroVAX as OBSERVER, following which you'll see some puff. Wait until you get the VMS `40a>' prompt.
    AAT Vaxcluster
    Username: OBSERVER
    Password: *********** (Ask your support astronomer)
            Welcome to VAX/VMS version V5.5 on node AAT40A
        Last interactive login on Friday,  5-SEP-1997 22:09
        Last non-interactive login on Thursday, 31-JUL-1997 15:50
            Welcome to the AAO OBSERVER system
        This software was developed in house by the AAO software group.
        It makes extensive use of the ADAM software system, supplied by
        the U.K. Astronomy Computing support group - Starlink
            Will now run ADAM setup procedures
      ADAM version 2.1 available
      Drama version 1.0 enabled.
            Using generic control task observing system
            Use SYSTEMS to list options, SYSTEM to change system
            Task configuration is 13-Nov-96
            Use VERSIONS to list options, VERSION to change version
            Type CCD_n to run the new Large External Memory CCD
            Type TEST1 for all tasks simulating
            Type TEST2 for only TEL task simulating
            Type TEST3 for all tasks simulating and INST logging
            Type TEST4 for only CCD task simulating
            Type FIGARO to set up for data assessment
         * PLEASE NOTE:                                           *
         *                                                        *
         * Observers are asked to log into the OBSRED account     *
         * to do their data reduction.                            *
      If using RGO, type "RGO" before "CCD_x" command to record RGO details in 
      your data file's FITS header

  2. You are now almost ready to start the OBSERVER software. Before you do so, however, you MUST type

    40a> WARM

    if you intend to run the CCD warm (ie at 200K). This is the default mode for imaging, unless you are using narrow band filters, so you will almost certainly want to type this.
  3. You will also want to start the system which automatically transfers files to the Sparc stations for later analysis.

  4. A second option available to you before you start OBSERVER, is what version of the code do you want to use. Usually you will want to use the latest `safe' version. This is usually called DEFAULT, and if you don't do aything, it will be what you get. Your support astronomer may, however, tell you that you should use a different version. In this case type

    40a> VERSION 37FEB96

    where 37FEB96 is the name of the version you want. A list of the available versions can be printed by typing

    40a> VERSIONS
  5. Now start OBSERVER by typing either

    40a> CCD_1        (or CCD1)


    40a> CCD_2            (or CCD2)

    depending on which controller you are using. (Again, it'll be written on the white board on the control panel.) Be patient, as this step can take a minute or so to get going. Wait for the


    prompt, to appear (see Fig 4.2 below).

4.2.1 Starting OBSERVER for use with OFFSET_RUN files

Some programs - especially surveys - require the observation of large areas of sky, or dithering of observations around a particular position. In this situation it may be useful to start OBSERVER such that it can use an `OFFSET_RUN file'. This is a file which can command telescope offsets before each exposure in a sequence is started.

To use these files, you must use a different procedure to start up OBSERVER. In fact, you need to start two windows - one OBSERVER window which controls the CCD, and one ICL window which you use to control the telescope and OBSERVER together

Log in as above, onto a suitable X-terminal, and follow all the steps up to (but not including) unix_server and ccd_1 or ccd_2. Instead use the following

40a> xon aatxth          (where aatxth is the X-term where you
                          want the second window to appear)               
40a> rvicl ccd_1                              (or rvicl ccd_2)

Now wait. The ICL environment will startup in the first window, and OBSERVER will start in a new window which will appear in a bit. This won't work if you don't give a valid display name at the xon command. Once the Idle: prompt has appeared on the OBSERVER window, you are ready to proceed. You can enter any of the usual OBSEVER commands in this new OBSERVER window. The first thing you should do is start the unix_server process by typing

in the OBSEVER window (not in the ICL window!).

You can now use the ICL window to start an OFFSET RUN FILE, with the command

ICL> OFFSET_RUN  filename

The OFFSET RUN FILE is a simple text file (with a default extension of .DAT) which commands telescope offsets and exposures. The exposures commanded can only change exposure times, so you can't use these files to execute a series of exposures with different filters. The default location in which ICL expects to find the offset run file is DISK$RAW:[OBSERVER]. If you create your own files using the OBSRED account, you will usually have to specify the directory as well as the filename


The creation of OFFSET RUN files is discussed in Appendix 2. Note that to execute normal VAX/VMS commands at the ICL> prompt, you must prefix them with a `$' character.

4.2.2 Summary of how to start up OBSERVER

4.3 Ending OBSERVER

You exit observer by simply typing the command

Idle: EXIT

which will prompt you to see if you want to log out of your login session also. Say 'yes' and you're done!













4.4 Running OBSERVER

Once OBSERVER is running, your VT102 terminal will look like so :-

Figure 4.2 - Typical OBSERVER screen on start-up (ie minimal command list.)

The screen is divided into four areas.

Commands to observer are entered by simply typing on the keyboard (OBSERVER has no mice!). All commands can be abbreviated, so long as the abbreviations are unambiguous. For example, MAXIMISE can be abbreviated to MAX, or MINIMISE to MIN. They can't however be abbreviated to M.

Figure 4.3 - Typical OBSERVER screen, but showing maximal command list.

One thing it is worth doing as soon as you start OBSERVER is changing the colour table used by the XMEM to display the CCD data. There are a range of colour tables you can use; GRJT (the default) is a horrible multicolour mess; GREY is a grey table; GRAY (different spelling) is the most useful. It's a grey table, but it displays values above 65535 as RED, so it's easy to see saturated pixels. It is highly recommended to use this table, so type.


4.4.1 Setting up the CCD

The CCD parameters you will need to set up are: (1) the window on the CCD you want to use; (2) the speed you want to read the chip out at; and (3) now to write the data to disk.

  1. Full Windows: Use the WINDOW command to set the CCD window and binning. For most imaging applications, you will want to use the largest window available, with all of the CCD's active area, and some overscan regions for bias estimation. For example for the TEK chip use,


    while for the Thomson, you'll want to use


    The window names actually point to files, which are used to define the window size and binning patterns. Appendix 3 describes how you can construct your own windows by editting an existing window file. For most applications, however, standard windows exist which should suit your needs. These are discussed below, and in Appendix 3.

    Test Windows: It can be very useful to be able to take short exposures using only a window of only few hundred pixels, to quickly check exposure times when taking flats, focussing etc. Special windows exist for this purpose;

    WINDOW TEK1K_DIRECT_CEN           (300x300 in the centre of the chip)


    WINDOW THOMSON_F1_CENTRE          (300 x 300 in the centre of the chip)


    WINDOW THOMSON_F1_TEST            (10 x 1024 for testing flats).

    Note that because of the residual image problems of the Thomson chips, the latter window must be used to test flat field exposures (ie exposures where the entire chip has large numbers of counts). If you use THOMSON_F1_CENTRE you will bin-up the rows you do not keep into the readout register, which could saturate it and produce residual image dark currents for hours!

    Binning: There is rarely much to be gained by binning the CCDs for imaging - at f/3.3 you get 0.391"/pixel with the TEK, so you'd need very bad seeing for there to be much point. It is possible you'd want binning at f/8, however, where the scale is 0.13"/pixel with the Thomson. In this case use the Thomson 2x2  binning window which reduces read-out times significantly :-


    Creating your own windows:
    Should you decide to create your own window, you will not be able to write it into the default area where OBSERVER goes to look for these files (which is DISK$USER:[SYSDISK1.ADAMLOCAL.WINDOW.Q0_1A] and .R0_0]), so you will have to specifiy the full path when you run WINDOW, eg. the following window was created to expose a 100 x 100 pixel region in the centre of the TEK chip (this is even faster than the 300 pixel windows).


    The required extension for a window file is .WINDOW. Note that all the windows mentioned above are available as hyper-links to the respective files, so you can play with them before your run if you want to.

  2. The CCD read-out speed is set with the command SPEED, eg


    The available speeds are NORMAL, SLOW, FAST and NONASTRO. See Chapter 2 for the properties of the CCDs at these different speeds. There are also available some detector-specific special read-out speeds (XTRASLOW for the TEK; SLOW_BESTCTE, NORMAL_BESTCTE, and FAST_GOODCTE for the Thomsons), which are also discussed in Chapter 2. For imaging you will usually use

    SPEED NONASTRO                 for the f/1 system, and

    SPEED FAST                          for f/3.3 direct broad-band, and

    SPEED NORMAL                      for f/3.3 direct narrow-band
  3. A third set-up you may require, is how the data will be written to disk. The default is to write the data as FIGARO  FLOAT files (ie as files of real numbers). This is somewhat of a waste of space and time, as at present, our A-to-D converters only produce 16-bit unsigned data. You can tell OBSERVER to write the data as FIGARO `Unsigned SHORT' files (ie 16-bit unsigned integers) using


    This is often a good idea for f/1 observing. However, it does mean that any on-line examination you do will be that much slower, as the code does an internal conversion into FLOATs. The FIGARO command RETYPE will convert USHORT images back into FLOAT.

4.4.2 Types of Exposures : RUN, DUMMY, GLANCE

There are three fundamental types of exposures OBSERVER will take

  1. RUN : a RUN is the basic way to take data. Once the image is read into the XMEM, and displayed on the XMEM's own display, it is immediately written to the archival disk on the MicroVAX. You will find your data goes onto DISK$INST:, in a file called something like


    As you might guess the actual directory used will depend on which controller you are using (CCD_1 or CCD_2) and the date on which your observations are made. Once the data are written to this disk, you can ONLY read them. You cannot delete files from DISK$INST:, as this is the AAT's archival disk. A number of other commands which also take data (eg. BIAS and DARK ) will also put their data on DISK$INST:. Note that DISK$INST: can be read from the Sparc Station, where it is called /vaxinst. So the above file would appear to the Unix file system as


    Since Unix has no version numbers you can only see the highest version from the Sparc. If you are using the unix_server system to copy files automatically from the VAX to the Sparc stations, then you will also find a copy of this file in the disk /data/ssf/1/obsred/ on aatssf.


    This system will also create a FITS and an IRAF copy of this file in the directories

    /data/ssf/1/obsred/fits/ccd_1/37feb96 and /data/ssf/1/obsred/iraf/ccd_1/37feb96
  2. DUMMY : a DUMMY is taken in exactly the same way as a RUN, except that the data is stored onto DISK$DATA:, and the files are called A.SDF, B.SDF, ...


    when you take a DUMMY after Z.SDF, the name wraps around to A.SDF again. The data on DISK$DATA: CAN be deleted by the observer, so it should be used for test data you know you won't actually use later, so that you can save space on DISK$INST:. As for DISK$INST:,  DISK$DATA: is cross mounted to the Sparc Station, ie


    Should you decide that you actually want to keep a DUMMY, you can use the command KEEP, which will copy an image from DISK$DATA: into the archival area on DISK$INST:.

    When unix_server copies dummy files, it places them in the same directory on the Sparc stations as the run files. There isn't a different directory tree for these files on the Sparc stations.
  3. GLANCE : a GLANCE is read out of the CCD and stored in the XMEM and displayed on the XMEM's display, but never saved to disk. GLANCEs are useful for test exposures that you know you don't even want to save on disk. (For example, when testing the twilight sky to see if it's bright enough to take a sky flat). However, if you decide you want to keep a GLANCE, you can again use the KEEP command, which will read and save the XMEM's contents to disk.

Both  DISK$INST and DISK$DATA have a capacity of about 2 Gb each. This is usually more than adequate for even the most prolific of observers, though for f/1 runs (where exposures of 5 minutes all night are not uncommon) the situation can become strained - the telescope staff try to ensure as much of this space as possible is available. Observers should consider using DATA USHORT if they think they are likely to run out of space.

4.4.3 Taking Data

To actually take data, you need to specify an exposure time and (optionally) an object name. This can be done in two ways. You can,

  1. Simply put them all on the one line. For example

    RUN 900 ABELL740/R

    will start a 900s exposure, with the object name ABELL740/R. Similarly,

    DUMMY 900 ABELL740/R

    will start a 900s DUMMY exposure.
  2. Or you can specify the time and object name seperately. For example

    TIME 900

    will achieve the same thing.  However, this second method will enter 900s as the default exposure time (and ABELL740/R as the default object name), so that on the next exposure if you simply type


    you'll get a 900s exposure with the object name ABELL740/R. There is also a COMMENT command which will place an arbitrary remark into the file's header. For example

    COMMENT Boy this really is exciting stuff!

    It is also worth remembering that the object name and the comment are only determined by OBSERVER at the end of the exposure. So it is perfectly legitimate to start an exposure with

    TIME 900

    Then after the exposure has started you can enter the object name or a comment

    COMMENT Cloud arrived 800s into exposure

    Though you must do this before the exposure is written to disk (you can, though, do it while the chip is reading out).

    You can also change the length of an exposure during an exposure by simply re-entering the TIME command - though make sure you don't put TIME to less than the length you've currently been exposed!

Whatever you do, it is worth briefly checking the parameters OBSERVER is using after you start an exposure (as opposed to what you think you may have told it) - especially if the exposure is a long one. It can be some what frustrating to finish a 900s exposure and find you are using only a 300 x 300 pixel window, or the wrong speed!

4.4.4 Other Commands

A few other commands are worth taking note of :-

HOLD simply closes the CCD shutter and stops counting down the exposure. CONTINUE recommences the exposure.
stops the exposure immediately, and begins reading  the CCD out.
aborts the exposure and does not attempt to read the CCD out. You can do an ABORT even when the CCD has begun reading out.
this sets the REPEAT counter to the specified number. When you execute an exposure after this, it will actually be executed n times. You can ABORT in the middle of a REPEAT sequence, but the REPEAT counter is left in an indeterminate state. If you do this you should type SINGLE to reset the REPEAT counter.
resets the REPEAT counter to 1 (ie puts you back in single exposure mode).
HELP command
will print a description of what the command does and how to use it. The available commands can be listed with the command MAXIMISE.
This command allows you to set the run number of your exposure. The default operation is to start at 1 on each new night. You may like to have your files numbered sequentially on all nights of you run, in which case you can use NEXTRUN to set the next exposure number to be one higher than the last exposure from the previous night.

4.4.5 When things go wrong

GENERAL GUIDELINES - The follwing general guidelines can be applied to most errors

START UP PROBLEMS - There are a number of things which can go wrong on start up, but they fall into a few categories.

  1. Problems loading OBSERVER. These are rare, unless you are trying to do something very funky. One of the tasks OBSERVER uses, may fail to load, or it may load and fail to initialise for some reason internal to the VAX (such as not being able to find a file that it requires). If this happens you will probably have to consult a VAX system expert. In the absence of such an expert you can try again, first attempting a RESET HARD, and if that fails, logging out and logging back in again.
  2. One of the various D-tasks has a pronlem with the device it controls. This may be some solid problem, like not having the XMEM switched on, or it may be some intermittent problem, such as noise on the serial line that connects the CCD micro control to the VAX. Or there may be some cases where a device seems to be responding normally, but with strange resulrs - a cabling fault on the CCD controller, for example, can cause the D-task to misidentify the detector in use, which can cause serious confusion. The system will respond to these in one of two ways
    1. Normally, the D-task in question will signal an error, and the overall control task will abort the initialisation, chaging to a state called System Failure.
    2. Sometimes, the system will just hang in the `initialising' state. This can happend if one of the D-tasks gets stuck in its initialisation sequence and fails to time out. (In general OBSERVER prefers to avoid having things time out, as some intruments can take a very long time to initialise - partiularly devices like the IRIS spectrograph which may have to wait for slow pieces of hardware to move. This leaves it hard to set realistic and safe time out limits, so OBSERVER often prefers to leave it up to the human observer to decide to abort something that has stuck.) Remember, and initialisation (especially on a loaded machine) can take some significant time. However, once you are convinced that the system is stuck, you can abort the initialisation with the ABORT commend (respond YES to the cautionary prompt - you don't have any data to lose yet) and the system will move into the System Failure state.

SYSTEM FAILURE - Following a major problem, you will usually find yourself in the System Failure state. Generally you should try to RESET the system, and if that fails, reload it - if necessary by logging out and back in again. A reasonable sequence to try is as follows:

DATA RECOVERY -  If the system has failed in such a way that there is data on the chip, or in the external memory that has not been transferred to the VAX and written to a data file, you will be notified of this at the next initialisation time.

If you need to recover data from the XMEM, do this before recovering data from the chip. The reason for this is that the RECOVER command can only recover the most recently created XMEM slot and if you read out the chip you will create a new and even more recent slot. To recover data from the XMEM


and to recover data from the CCD,


Note that data cannot be recovered from the XMEM after it has been rebooted.

4.5 Typical Observing procedures

A number of procedures are carried out so commonly, but have a few wrinkles over straight observing, that they are summarised here.

4.5.1 Flat Fields

Ensure that you always take flat fields with exposures of at least 5s and preferably greater than 10s, otherwise all you will measure is the shutter open-close pattern.

The Thomson CCDs are seriously affected by overexposure to light. While this causes no physical damage to the chip, recovery from saturation is slow, and high dark current and after images may persist for several hours if the chip is grossly saturated. You  must be extremely careful not to everexpose the detector while taking flat fields. This usually means always testing you exposure levels first with very short (0.1s) exposures, or with your most insensitive filter in.

Thomson user must also note that small windows used to test flat field exposures must read entire vertical columns from the CCD - as does the sample 10 x 1024 window THOMSON_F1_TEST. If you use THOMSON_F1_CENTRE you will bin-up the rows you do not keep into the readout register, which could saturate it and produce residual image dark currents for hours!

Flat fields can be taken in a number of ways.

  1. Dome flats : A white patch on the AAT dome can be illuminated by lamps on the Prime Focus Access landing. The telescope should be moved to the zenith, the dome rated to 0° and the windscreen moved to 21°. (If you want to take flats and need the telescope or dome moved please contact the afternoon technician-on-duty after 2pm to request this. The technician's name will be listed on a small notice over the control panel, and usually the telephone extensions where they can be found are written on the whiteboard on the control panel.)

    There are two lamps used: a small reading lamp, and a quartz-halogen spotlight. The second is much, much brighter than the first. They should be plugged into a power-outlet at the back of the PF Access landing, which is connected to a dimmer in the control room. In practise, it is very difficult to get domeflats in the redder broadbands, because the lamps are too bright. By the time you dim them enough to be able to observe them, the light leaks in the dome during the day can make the whole excercise questionable. Dome flats can be taken in blue broad-bands, but here the lamps are so spectrally different from the sky, that the flats are not of a hell of a lot of use.

    Having said that, domeflats are good insurance in case you don't get any sky flats, and are easy to take.

    You cannot turn off the lights in the dome until the Visitor's Gallery has closed at 4pm. After this there is a delay of ~15 minutes programmed into the lights for the stairwell in the Visitor's Gallery (to stop visitors falling down the stairs in the dark!), so you can't start domeflats until 4:15pm at the earliest.
  2. Twilight sky flats : These are the usual way people get flat field data. You should be ready to take flats 10-20 minutes after sunset for f/3.3, 30-40 minutes after sunset for f/1. The night assistant will usually be at the telescope in time, but it is a good idea to forewarn them. A list of (relatively) blank fields for taking these flats is given in Appendix 8. Two tricks can considerably speed the process up
  3. `Super skyflats' : If you have enough counts in your data, and they are no tdominated by large extended objects, you may be able to construct a flat field (or at least an optical illumination correction) from the suitably scaled and medianed data itself. That will depend greatly on what band you operate in, and how long your exposures are - ie. on how many sky counts you get.

Flattening f/1 Data : The f/1 optics cause some vignetting at the edges of the field, so flat fielding is necessary not only to remove pixel-to-pixel sensitivity variations in the CCD but also to correct for this vignetting. Appendix 7 shows typical flat-fields for each of the currently-used filters. Experience shows that flat-fielding of images taken with the focal reducer can be done to better than 0.5%, but this requires some careful work to created a master flat-field frame from the median of twenty or so sky-limited frames. Observers whose fields contain few bright stars or extended objects can achieve this by making a super-sky-flat.  If most of the data frames contain large extended objects such as bright galaxies, this approach is not usually feasible and there are two possible alternatives. One is to take twilight sky flats (as above), the other is to take a number of `flat' frames on the dark sky, offsetting the telescope by several arcseconds between each. Which of these is chosen will probably depend on the requirements of the observing program. There is no significant night-sky fringing with the current (thick, uncoated) Thomson CCD.

Flattening f/3 Data : Sample flat fields taken with the f/3.3 camera and the TEK ccd are available in Appendix 7

4.5.2 Focus Sequences

A suggested recipe is as follows:

  1. Acquire a field containing at least one star which is suitably bright (a few thousand counts in 5 seconds, or V~15-16).
  2. Start an exposure sequence in multiple mode by typing MULTIPLE, then EXPOSE to start the first exposure  (the CCD will not be read out until the whole exposure sequence is finished and you type READOUT).
  3. After the CCD shutter has closed, ask the night assistant to offset the telescope by about 15 arcsec in right ascension or declination and move the telescope focus. Type EXPOSE to start the second step.
  4. Take a sequence of 6-8 steps, moving the telescope and focus between each step. Before the last step, it is useful to ask the night assistant to move the telescope a double step on the sky. Terminate the exposure sequence by typing READOUT after the last exposure finished. The CCD will then read out and the focus frame will be displayed.

    (This procedure sounds cumbersome, but in practice the AAT's night assistants can carry it out quite quickly. Because focus offsets between filters are constant, and the telescope focus is automatically adjusted for changes in the tube length due to temperature and attitude changes, you rarely need to do this more than once a night.)
  5. You can then pick the best focus value - this usually requires redisplaying the image on the Sun or VAX, as the XMEM's default stretch is fairly uninformative. Either QIKLOOK, IMEXAM in the local Figaro version (the 'S' or 'P' functions), or the IMEXAM program in IRAF will enable you to determine profiles for each star. QIKLOOK and IMEXAM, in particular,  will plot FWHM as a function of focus position, and a evaluate the best focus using a polynomial fit. For a description of the use of QIKLOOK and IMEXAM  for quick look examination of imaging data see Appendix 5. Ask the night assitant for the number of exposures (say n), the first focus value used, and the step between focus values, as the programs will prompt you for these. Then use any key to select the n images of a single star in order. The program will fit for the FWHM in each, and plot them as a function of focus position, from which you can get the best number.

4.5.3 Standard Stars

Standard stars can be taken using regular `one-off' exposures, just like any other frame. However, as the times used for standard stars are usually short (5-15s) compared to the readout time (~60s), a few tricks can save time and get you more data.

Both the Landolt (1992) and Graham E-region standards star lists are available in the control room.

4.5.4 Bias & Dark Frames

BIAS: There's not much special to mention about bias frames (or zero frames if you're of an IRAF bent) , except that there is a special command for taking them


which is functionally the same as


except that your frames will get object names like BIAS_0001, BIAS_0002, etc. In both cases the shutter will never open. Sample BIAS frames are available in Appendix 7

DARK: Dark frames can be taken with the DARK command.

In both cases (but especially for DARKS) the dome should be dark, so you can't start these until after 4:15pm either.

4.5.5 Pointing Stars

The AAT points and tracks superbly, so there is no real need to check pointing. However, if you are carrying out a program where you are pointing blind (say, looking for optical IDs of radio sources), you may like to know where on the CCD is the position you have requested from the AAT. Or, you might be pedantic.

The night assitant will usually acquire what's known locally as a SNAFU star at the beginning of the night, just to check eveything is working and to make a seeing measurement with the telescope-mounted DIMM. You can take a very short exposure of this star (0.1s) or put in an insensitive filter (U or Halpha) to take a snapshot of this object.
Users of the Thomson chip should avoid this procedure, or be VERY CAREFUL. If you overexpose the centre of the chip, you'll have a residual image which may not go away for hours.

4.6 Riding in the Cage

Section 3.1 contains a description of the equipment an observer riding in the cage will find around him - please familiarise youself with its contents. The observer riding in the cage will be required to acquire guide stars and change filters.  

The filter wheels, filter slides and filters are kept in the dark room on the sixth floor (next to the toilet). Also in this dark room are materials to clean the filters - but, please ask the AAO staff to do this for you if its necessary.

4.7 A Typical Night

The following is a summary of the commands you're most likely to use on a typical night's observing. It is not a guide to how to carry out you SCIENTIFIC program, but a guide to the OPERATIONS you will probably use.

For general information on preparing for your run see Section 1.10 of the AAO's Observer's Guide.


Unlike other CCDs in use at AAO, the Thomson CCD is seriously affected by overexposure to light. While this causes no physical damage to the chip, recovery from saturation is slow, and high dark current and after images may persist for several hours if the chip is grossly saturated.

Observers are therefore urged to avoid illuminating the Thomson with bright lights either before or during an observing run. Since the dark current also takes several hours to stabilize after the CCD is first powered on, the system should be powered up as early in the day as possible, and observers should be aware that the dark current will increase substantially if for any reason the electronics have to be switched off and powered up again during the night. See Section 2.2 for more details on the dark current produced by gross saturation.

4.7.1 In the Afternoon

Because you can't turn all the lights out in the dome until after 4:15pm, there's not a lot of actual data taking you can do in the afternoon. However, it is worth check out your setup (have the right filters been put in the wheels, is the wheel mounted in the filter slide and installed in the camera head ...) and taking a few test GLANCEs to see if the CCD is working properly.

Hardware setup : The filter wheels, filter slides and filters are kept in the dark room on the sixth floor (next to the toilet). Also in this dark room are materials to clean the filters, if necessary - please don't do this yourself however ... ask the AAT staff. If you can't find your filters or filter wheel here, they may also be on the table on the PF Access landing, or actually mounted in the camera head. Your instrument scientist should be able to help you with all this setup stuff.

You should also check the CCD setup. This will be shown on the small whiteboard on the AAT control console in the Control Room. Your should check that the CCD is running `warm' (ie at 200K), unless you specifically want it to run `cold' for narrow-band observing. You should also check which controller (CCD_1 or CCD_2) the CCD is connected to.

Software startup : Now startup the OBSERVER system on the rightmost of the VT102 terminals.

Calibrations : Once the lights have been put out you can take the bias, dark and domeflat frames.
Commands (user input is marked in
BOLD font)
C Conect to the MicroVAX
LOGIN: OBSERVER Login as observer
PASSWORD:*********** Your support astronomer will give you the current password
WARM Skip this if you are not running the CCD warm (ie 200K)
CCD_1 or CCD_2 Depending on which controller you're attached to
Now wait for the Idle: prompt
Usually used for f/3.3
Usually used for f/1
COLOUR GRAY Put in the useful colour table
GLANCE 0 You should get a full size bias frame.
REPEAT 15 Take (for example) a series of BIAS frames
DARK 3600 Or perhaps a 1 hour dark

4.7.2 After Sunset

The CCD will need to be filled with liquid nitrogen before observing starts. This is usually done before taking skyflats, but can be done in between taking sky flats and starting real observations if time is short.

You should be ready to start taking sky flats about 15 minutes after sunset for f/3.3, 35 minutes after sunset for f/1. By this time your night assitant hould have arrived. He will startup the telescope, dome etc; open the dome for you; and move to a hopefully blank area of sky.

You should put in a small window for taking test exposures and wait until the sky becomes faint enough to start observing. You'll usually start taking flats in your bluest filter, and work redward.

Users of the Thomson CCDs beware! The Thomson chips are seriously affected by overexposure to light. Recovery from saturation is slow, and high dark current and after images may persist for several hours if the chip is grossly saturated. You  must be extremely careful not to everexpose the detector while taking flat fields. This usually means always testing you exposure levels first with very short (0.1s) exposures, or with your most insensitive filter in. You must also use a test window (eg THOMSON_F1_TEST) which reads entire vertical columns, or else you may saturate the readout register.
Commands (user input is marked in
BOLD font)
WIN TEK1K_CEN_100 or
300 x 300 pixel window in the center of the chip, or a
100 x 100 pixel window in the center of the chip, or a
10 x 1024 pixel window in the center of the chip.
GLANCE 0.1 Test the sky brightness and wait. Make sure you
have the correct filter in place.
GLANCE 1 When you start to get about  a fifth of the available
counts (10000 counts for FAST speed, 5000 counts
for NONASTRO) you're ready to take data.
Switch to the full window
GLANCE 5 Flat fields should never be shorter than 5s and
preferably not shorter than 10s.
 Now examine the field and see if its OK. Is the frame unsaturated?
KEEP If it's OK the KEEP it. Otherwise try again
RUN 7.5 Increase the time, move the telescope, and take another one.
...... Repeat until done

Once you have finished your skyflats, the night assistant will want to do a few things with the telescope. These involve measuring the seeing with a telescope mounted Differential Image Motion Monitor (DIMM),  (possibly with a finder telescope also), acquiring a SNAFU star, etc. Please be patient. If a SNAFU star is acquired you may want to take a quick snapshot to get the fiducial location of the telescope's pointing on your CCD image.

One this is done you are ready to check the focus (you may have to wait for the sky to get faint enough) - see Section 4.5.2.
Commands (user input is marked in
BOLD font)
  Acquire a field with some brightish stars in it.
SPEED NONASTRO Use the fastest speed available.
TIME 10 Set the exposure time to use per EXPOSE
MULTIPLE Start Multiple mode
EXPOSE Take an exposure, then move the telescope and adjust the
EXPOSE Repeat this step 6-8 times.
  When done, readout the chip
SPEED FAST, NORMAL, .. Put the speed back to what it was before,  if necessary.
  Now display the image on the Sun or VAX and choose the
best focus value.

Once you know the focus for one filter, the focus for the other filters is determined if the focus offsets are known. The offsets for the AAO's filters are well determined. It is very important you remember that the focus must be adjusted everytime you change the filter. This means you must tell the night assistant when you change filters, so that he can change the focus.

Standards can be acquired in a similar way (though obviously without moving the focus).

4.7.4 After Twilight

Now you're ready to observe.
Commands (user input is marked in
BOLD font)
  Acquire your target object. Get the observer in the
cage to put the correct filter in. Set the focus value
appropriate to your filter. Get the night assistant to
find a guide star from the GSC and give the X-Y
guide probe position to the observer in the cage.
Acquire the guide star with the eyepeice. Then
switch the guide probe to `PMT' and see if the
star is visible on the ST4 CCD camera (you may
have to focus the eyepiece and the ST4 CCD at the
start of the night). Start guiding.
OBJECT ABEL9999/B Object name
TIME 1800 Time
OBJECT ABELL372/B You can change the name before the exposure ends if
you have to.
RUN 900 ABELL372/V You can also put all the stuff on the one command line
.... From here on, you're on your own.

The `cage riding' observer will usually demand to be changed out somewhere in the middle of the night. This is usually a good opportunity to top up the liquid nitrogen in the dewar.

The data can be examined on-line with either the VAX or Sun workstations. Remember the VAX data goes in files like

DISK$INST:[CCD_1.960237]37FEB0001.SDF         or

which you can see from the Sun in files like

/vaxinst/ccd_1/960237/37feb0001.sdf           or

You should use the OBSRED accounts for looking at your data. Appendix 7 has details on how to start the IRAF and FIGARO packages, and some hints on how to display the data.

4.7.4 In the Morning

Simply reverse the above precedures for sky flats (if you want them), staring with the reddest filter and moving to the bluest as the sky gets brighter. You can log out of the OBSERVER system if you like, but it isn't necessary. The CCD will have to be topped up with liquid nitrogen in the morning.

On the weekend, the afternoon technician will not come in until 2pm. It is therefore the observer's responsibility to get the dewar filled before then if necessary (if say, you go to bed early because its snowing). The night assitant or support astronomer will show you how to do this.

If you plan on saving your data to your own tape (as opposed to having the AAO extract a tape for you from the AAT archive), you should start that going before you go to bed also.

If this is your last night, make sure you fill in an Observing Run Report form!

4.8 Preparing for your run

For general information on preparing for your run see Section 1.10 of the AAO's Observer's Guide.

4.8.1 f/3.3 Observers

Photometric standard stars should be chosen carefully, as many are too bright to observe with a 4m telescope. The minimum recommended exposure time is 5 seconds, as set by the shutter timing. Stars brighter than about V=12 may saturate the CCD in less time than this - though this will depend somewhat on seeing. You can also observe somewhat brighter stars by de-focusing the telescope. Both the Landolt (1992) and Graham E-region standard listings are available in the AAT control room.

It is also worth giving some advance thought to flat fielding. If sky flats are to be taken at the start of the night, there is only a relatively narrow time window in which this is possible and it is best to forewarn the night assistant that twilight sky flats are needed so that the telescope is ready at the appropriate time (see Section 4.5.1).

If you plan to use filters other than the standard UBVRI, you should contact your support astronomer well ahead of your run to check that this is feasible.

Details of how to prepare a catalogue are given in section 5 of AAO TM 7 ( Telescope Control System) and in Appendix 4. Observers need only create a suitably-formatted file on the VAX or Sun - the night assistant can then transfer this to the telescope computer at the start of the night's observing.

4.8.2 f/1 Observers

Since most f/1 observers have a long list of program objects with only a relatively short exposure time on each, some advance planning of the run will help increase efficiency. Most importantly, it is advisable to prepare a co-ordinate list (data catalogue) which can be entered into the telescope computer before the start of the run. This allows the co-ordinates of each object to be recalled automatically when required without the night assistant needing to enter them from the keyboard. Appendix 4 has a description of how to create these data catalogues. Observers need only create a suitably-formatted file on the VAX or Sun - the night assistant can then transfer this to the telescope computer at the start of the night's observing.

Photometric standard stars must also be chosen carefully, as many are too bright to observe with the f/1 system. The minimum recommended exposure time is 5 seconds, set by the shutter timing because the large shutter of the f/1 system takes a finite time to open and close. Stars brighter than about V=14 may saturate the CCD in less time than this, and are therefore unsuitable.

It is also worth giving some advance thought to how the data are to be flat-fielded. If sky flats are to be taken at the start of the night, there is only a relatively narrow time window in which this is possible and it is best to forewarn the night assistant that twilight sky flats are needed so that the telescope is ready at the appropriate time (see Section 4.5.1).

If you plan to use filters other than the standard glass V, R and I filters supplied with the system, you should contact your support astronomer well ahead of your run to check that this is feasible.

Since an autoguider cannot be used with the f/1 system, there is no need to supply guide star positions.

4.9 Operational Stuff

For general information on the AAO, see the AAO's Observers Guide.

4.9.1 Finding your Support Astronomer

Your support astronomer will usually be either already be on the mountain, or will arrive the night before your run. So the most likely place to find him will be at dinner at the Lodge. If you need to track down your astronomer in the afternoon, the most likely locations for hime are (a) asleep, (b) having breakfast at the Lodge, (c) in one of the offices on the 1st floor, (d) in the computer room on the 2nd floor, or (e) rattling around on the 6th floor (including possibly in the dome, on the catwalk or at PF access).

4.9.2 Finding the Technician-on-Duty

The name of the Technician-on-Duty (also known as the Afternoon Technician) will be shown on the console in the control room. The technicians usually leave the phone numbers where they can be contacted on the console whiteboard. If there are no numbers listed there, try extension 6283 or 6284 (the electronics lab on the first floor).

4.9.3 The Night Assistant

During observing runs, a night assistant is in charge of the telescope. This assistant has absolute power to shut the telescope down and/or close the dome for reasons of telescope safety. The night assistant will operate the telescope and control system to acquire and track an object and will assist the observer as directed. The observer will normally be responsible for the actual data taking and control of the instrumentation.


4.9.4 Observing Limits

Under all circumstances, the night assistant has the final say over decisions to open or close the dome.

4.9.5 The Dome

Observers should be aware that the dome floor and console level walkways can be dangerous places at night when the telescope is in darkness. No-one should enter these areas after dark unless they are already familiar with this part of the building.

For new observers, the support astronomer can arrange a tour during the afternoon or before the lights go out. Torches (or flashlights) are available next to every entrance to the dark areas of the dome. Astronomers entering these areas must take one of these torches with them, and return it when they leave the dark areas. Astronomers entering these areas must also let the night assistant know that they are going out into the dome.

No food or drink may be brought into the 6th floor control room, but the small observers' lounge on the same floor has a refrigerator and microwave oven as well as facilities for making tea and coffee. Toilets are located on the console floor (level 6), the office floor (level 1) and the main floor (level 4).

4.9.6 Fault Reports

A `Fault Report' is submitted by the night assitant whenever something goes wrong during the night. CCD imaging astronomers should be aware that if the rectification of a Fault Report requires the CCD to be removed from the camera head, the AAT's staff will go ahead and remove the CCD without further reference to the astronomer. Astronomers are therefore advised that if it important to their program that the CCD not be removed (eg. to retain the previous night's flat fields), they make sure to inform the night assitant to make such a note in the Fault Report.

4.9.7 Observing Run Reports

Each observer is required to fill in an Observing Run Report form at the end of their run. This form is used to provide feedback to the observatory on its facilities, how successful your run has been, and any problems you may have encountered during your run. The contents of these forms are sent directly to the Director, as well as the responsible AAO staff. As well as being required of all observers, it is also strongly to your advantage to help us improve the AAO's facilities by providing us with your feedback!

The form can be found at

Introduction The Telescope & Optics The Detectors
The Imaging Cameras An Imaging Cookbook The Data you Take Away
Exposure times OFFSET_RUN files CCD Windows Data Catalogs
On-line Reduction Filters Flat-fields Blank Fields Orientation Shutters

Back Contents Next

Last Contents Next

This Page maintained by : Chris Tinney (
This Page last updated:  7 Sep 1997, by Chris Tinney