Skip to content. | Skip to navigation

Personal tools
You are here: Home Telescopes Irenee du Pont Instruments Website Direct CCD Camera Direct CCD Users Manual



Patricia Knezek, Ian Thompson, Gaspar Galaz
August 1996
Last modifications: Mark Phillips November 2000


Table of Contents



1. Introduction

2. Getting Started

3. The CCD Command Window

4. Guiding an Exposure

5. Data Storage and Tape Recording

6. Things to Watch Out For

7. Images You Will Need for Data Reductions

8. Observing Hints

1. Introduction

This manual describes how to take direct imaging or spectroscopic CCD data on the Las Campanas 40-inch Swope telescope and the 100-inch du Pont telescope using an acquisition program running on Sun workstations. CCD data can also be taken under the control of Steve Shectman's PC acquisition program ``ccd'' (and this is presently the only way to take drift-scan data). For further information see the appropriate manuals at the domes or contact the individuals listed at the end of this manual.

Please pass along any comments and suggestions for improving the acquisition program and/or this manual to Mark Phillips or Ian Thompson.


NOTE: SITe3 CCD will be soon replaced by a new detector. In the meanwhile, you may find some useful updates on SITe3 operations at the Carnegie Supernova Project web site:


Table 1 and Table 2 describe the 100-inch and 40-inch telescopes, respectively, and Table 3 describes the various chips available for use and their performance characteristics.

Table 1. du Pont 100-inch Direct Imaging Mode
Type: Ritchey-Crétien optimized at f/7.5
Diameter of Primary: 100" (2.540 m)
Primary Focal Length: 300.0" (7.620 m)
Cassegrain Focal Length: 750.0" (19.050 m)
Nominal Plate Scale: 10.8 arcsec/mm
Field of View: 718 mm (~2.1 degrees) in diameter
Sky Shields: 2-element provides complete shielding over 10 armin diameter field and blocks 3% of beam
3-element provides complete shielding over 18 armin diameter field and blocks 11% of beam

Table 2. Swope 40-inch Direct Imaging Mode
Type: Ritchey-Crétien optimized at f/7.5
Diameter of Primary: 40" (1.016 m)
Primary Focal Length: 162.1" (4.118 m)
Cassegrain Focal Length: 280.0" (7.112 m)
Nominal Plate Scale: 29.0 arcsec/mm
Field of View: 372 mm (~3 degrees) in diameter
Sky Shields: 2-element shield

Table 3. CCD Detectors at Las Campanas Observatory
Chip: Tek#1 Tek#2 Tek#5 SITe#1 SITe#2 SITe#3
pixel size (µm) 24.0 21.0 24.0 24.0 15.0 15.0
# pixels 10242 10242 20482 20482 1752x572 2048x3150
40" pixel scale (arcsec/pixel) 0.696 0.609 0.696 0.697 0.435 0.435
100" pixel scale (arcsec/pixel) 0.259 0.227 0.259 0.244 0.153 0.153
40" FOV (arcmin) 11.9 10.4 23.8 23.7 12.7x4.1 14.8x22.8
100" FOV (arcmin) 4.42 3.87 8.85 8.3 4.4x1.4 5.2x7.8
QE(%) at U(3700 Å) 47.0 20.0 60.0 48.0 44.0 51.0
QE(%) at C(4140 Å) 53.0 26.0 73.0 52.0 50.0 55.0
QE(%) at B(4550 Å) 54.5 29.0 75.0 52.0 52.0 56.0
QE(%) at V(5610 Å) 57.0 33.0 77.0 55.0 54.0 56.0
QE(%) at R(6830 Å) 52.0 42.0 74.0 57.0 52.0 55.0
QE(%) at I(8380 Å) 33.0 29.0 59.0 37.0 34.0 36.0
QE(%) at z(9720 Å) 10.0 7.0 ... ... ... ...


2. Getting Started

The day crew will have set up the CCD camera and the control computers. The CCD camera is controlled and the data are gathered by a 486-PC. The 486-PC also controls the instrument mounting base (guider position, filter wheel, and shutter) at the 40-inch and the filter wheel and shutter at the 100-inch. The observer operates the camera from the CCD command window running on a Sun Ultra 1 (named Canopus) at the 100-inch and on a Sparc 5 (named Henrietta) at the 40-inch. Normally the day crew will start the program and make sure that everything is running; the start-up procedure is given here in case you have to do i


2.1 Logging In

To get started from scratch on either the 40-inch or the 100-inch, login to the Sun workstation with username obs40 or obs100, depending on which telescope you are using. Passwords are written on the monitors of the corresponding computers. Otherwise, they are available from the day crew. At the prompt, initiate OpenWindows by typing:
    opw or openwin

One or two windows will come up. There will be one saying ``Console'', and possibly one saying ``/bin/csh'' as well. If a third window, saying ``xterm'' and running IRAF with an error message shows up, kill it by typing `exit' within the xterm window or using the right button of the mouse at the top of the `xterm' window and selecting the `QUIT' option. Then start a fresh xgterm/IRAF window as described in section 2.3.


Now look at the 486-PC screen to see which instrument driver is loaded. This information can be found on the last few lines of the boot messages, immediately before the ``user name'' prompt. You will see either ``ccddriver'' or ``irdriver''. If the correct driver (i.e. ``ccddriver'') is loaded, all is fine; if not, see Appendix A for instructions on how to change the driver. If the correct driver is loaded, you are all set to start up the command window as well as an IRAF window to examine your data.


2.2 Starting the CCD Command Window

First, in the console window, go to the directory where you want the data to be stored. This can be any directory where the user obs40 or obs100 has write privileges, normally on one of the data disks. Then start the CCD command window by typing:

This will launch an xterm window with the title ``CCD_Dialog'', as well as a configuration window (see Figure 1). This window has pull down menus (activated with the right mouse button) for selection of the type of shutter to be used (curtain for all large CCD's), the CCD, and the telescope. There is also an entry to allow changing the number of overscan columns from the default of 16. When the appropriate selections have been made, click on "OK" and the CCD program will then start. The CCD_Dialog window will display various messages as the data are collected.



Figure 1. The CCD Configuration Window

If you wish to keep each night's data in a separate subdirectory you will have to kill the CCD acquisition window and restart it from each new subdirectory.


2.3 Starting IRAF

IRAF is used to display the data and to measure simple diagnostics from the images. To start IRAF on either telescope, first open an IRAF xgterm window using the right button on the mouse (while on the screen background) and selecting ``IRAF'' and then ``IRAF-Xgterm''. Then in the IRAF window go to the data directory from which the CCD command window was started.

To start up an ximtool window for displaying your data, use the right mouse button to select ``IRAF'' and then ``Ximtool''. Then within the IRAF window once again, set the ximtool window to the correct size by typing:

    set stdimage=imtN

where N is the number of pixels you wish to display (512, 800, 1024, 2048, etc.). For more information on IRAF data processing, see Appendix D.


3. The CCD Command Window

The CCD command window controls all aspects of taking data and changing filters. Figure 2 shows the layout of the command window.



Figure 2. Layout of the CCD Command Window


The window is divided into four sections. The top section contains various function buttons that control the starting and stopping of data acquisition. The second section contains panels to enter exposure time, the data file name, the filter, and various pieces of information about the exposure that are stored in the header of the data file. The third section contains CCD status information, as well as settings for the exposure that are not frequently changed. Finally, the bottom of the command window has a section where messages on the progress of an individual exposure are displayed.


3.1 Function Buttons

All function buttons are activated by clicking on them with the left mouse button.
  • Start: This starts an exposure or a sequence (loop) of exposures. Various parameters of the exposure are set in the second section of the window: the exposure time (in seconds) is set in the Extime panel, the number of exposures in a loop is set in the Loop panel, and the image is stored in the disk file named ccdnnn.fits, where nnn is the Disk File number. Note that if this file already exists it is overwritten.
  • Focus: The telescope is most easily focussed onto the CCD using this function. A series of exposures is taken on a single CCD image, with the user changing the focus between each exposure. At the end of each exposure, the image is automatically shifted on the CCD, and so the final image has a sequence of star images with different telescope focus settings. Note that the Focus Button is close to the Start button, and if the Start button is clicked during a Focus sequence, then a message is printed in the bottom section of the acquisition window, and no action is taken. Beware that the focus mode does not work properly when the subraster mode is on.

    The user selects the number of images in the sequence (generally 7) by setting the Loop counter in the second section of the command window. The exposure time for each is set in the Extime panel. The telescope focus is set to the first value of the sequence, and the exposure is started by clicking on the Focus button. At the end of the first exposure, the shutter closes, and the charge is shifted on the CCD. The user then sets the telescope focus to the second position (reasonable steps are 0.05 units at the 40-inch and 5 units at the 100-inch if the telescope is close to being in focus) and starts the second exposure by clicking on the Focus button again. This continues until the sequence is finished, and the resulting frame is read out into a disk file called focus.fits. The best telescope focus can be estimated by displaying this image and, for example, measuring the image parameters with the IRAF routine IMEXAMINE. In particular, within IMEXAMINE the FWHM of the stellar objects is measured with the keystroke ``.'' and the ellipticity with ``e''. Note that the first image shift between focus settings is twice the size of the subsequent shifts. All focus exposures are read out into the file focus.fits, so previous focus frames are overwritten. If you want to save them for some reason, be certain to copy focus.fits to another file.

  • Abort: This stops an exposure, or a focus or multiple sequence. The continuous wipe is turned on, and the image collected to that point is lost. Before starting the next exposure, wait some seconds to complete the wipe of the CCD. For example, for the SITe#1 you should wait for ~13 seconds and when using the SITe#3 for ~26 seconds. This allows the continuous wipe to clean the CCD of the image accumulated before the exposure was aborted. The Disk File counter is not updated when an exposure is aborted.
  • Macro: This button starts a sequence of exposures run under the control of a simple macro language. The language allows the user to change filters, and to select exposure times and the number of exposures in a loop. To use this feature, be sure that a copy of the file ccd.macros is in your data directory. Edit this file to build the macros you want to use, and then enter the name of the macro to be run in the ``Macro'' field in the third portion of the CCD command window. Clicking on the ``Run Macro'' button will then execute the selected macro. See Appendix B for more information.
  • Multiple: This function is similar to the focus sequence except that the image is not shifted on the CCD between exposures. The intention is to give the user the option of moving the telescope between exposures to take multiple images of, for example, a standard field on a single CCD frame. The exposure is stored as a regular image. The use of this function is similar is similar to the Focus function: select an exposure time in the Extime panel and the number of exposures in the Loop panel. Then take the sequence of exposures by clicking on the Multiple button, moving the telescope between exposures.
  • Abort/Read: This function aborts the exposure and reads out the CCD with the image accumulated to that point. The Disk File counter is updated.
  • Snap: This function takes a ``quick look'' exposure, automatically rebinning the CCD by 4x4 pixels for the large (2048x2048 pixels) CCD's, or 2x2 for the smaller ones. The data are put into a regular disk file, but the Disk File counter is NOT incremented.
  • Pause: This function closes the shutter and waits until the exposure is continued by clicking on this button again, or aborted with the Abort or Abort/Read buttons.

3.2 Selecting the File Name, Exposure Time, and Filter

  • Disk File: Regular exposures begun with the Start and Snap functions are read out into files named ccdnnn.fits, where nnn is the value in the Disk File panel. This number can have values between 000 and 9999. The number is incremented by one at the end of a regular exposure, and is NOT incremented for a Snap, nor when the Save flag is set to Test (see next section). The Disk File number can be set by the user by clicking the left mouse button on the panel field and entering the new value. (This is the case for all panels in this section with the exception of the menu panels Filter and Imtype.) Note that if a frame with the same number as the current exposure already exists on the disk, it will be overwritten.
  • Extime: This is the exposure time in seconds. The Actime panel immediately below displays the current running time for the exposure. Fractional exposure times (e.g. 1.5 sec) are not allowed. Also, exposure times of less than 5 seconds are not recommended when the curtain shutter is used. If you wish to change the duration of an exposure while it is running, set the Extime panel to the new value and enter it with a <cr>. To take bias frames, set Extime = 0.
  • Loop: This panel controls the number of exposures in a Loop sequence, or in a Focus or Multiple frame. The panel to the right labeled ``Doing'' lists the current exposure in the sequence.
  • Object: The object name for the exposure, entered by the user and stored in the header of the data file.
  • Comment: The user can enter a comment line which is stored in the header of the data file.
  • Filter: This menu driven panel controls the filter wheels at the 40-and 100-inch telescopes. To choose a filter, click the left mouse button on the panel to bring up the menu, then choose the new filter by clicking the left mouse button on the choice. The labels for the filters default to 1 through 6. If you wish to have names attached to the filters, create a file named filter.names in your data directory, with one name per line, and then restart the CCD command window. The names in the filter.names file can be up to 16 characters in length, and can contain spaces. In either case, these names are used in the macro language to refer to the filters (see Appendix B for more information on macro use).
  • Aperture: This controls the aperture wheel on the Wide Field CCD Camera and is not used for direct or spectroscopic CCD observations.
  • Imtype: This is a menu to chose the value of the IMTYPE entry in the fits file header for the exposure. Available image types are dark, flat, fringe, illum, object, zero, none, and unknown.
  • UT and Date: The UT and Date for the beginning of an exposure are read from the Telescope Control System (TCS) at the 100-inch or the control PC at the 40-inch, and are stored in the header.
  • RA and Dec: These are read from the TCS at the 100-inch and need to be entered manually at the 40-inch. The airmass is also read from the TCS at the 100-inch. Note that the IRAF format for RA and Dec (if you are going to use IRAF routines to calculate airmass) is nn:nn:nn.n and snn:nn:nn.n, respectively.

3.3 Selecting the Binning, Gain, and Subrastered Readout

This section contains various instrument settings and status flags that the user will probably not need to change frequently during a typical observing night, as well as some manual controls useful for engineering. The Shutter, Wipe, Subraster, and Save functions are controlled by clicking the left mouse button on the appropriate diamond button. The corresponding function is activated when the diamond is hole-like and dark.

After making the appropriate settings the user can economize on screen space by holding the left mouse button depressed with the cursor on the small square on the line separating the third and fourth sections, ``hiding'' this section by sliding the dividing line up, and then resizing the command window.

  • Shutter: This is a flag that indicates whether the shutter is open or closed. The shutter can be opened and closed manually by clicking on the diamond symbols.
  • Wipe: This is a flag to indicate if the CCD is being wiped. The default condition is that the CCD is continuously read out in a 4x4 binned mode between exposures.
  • Subraster: The CCD can be read out in a mode where up to eight subrasters of the complete frame are saved. Clicking on the subraster button generates the CCD Subraster window (see Figure 3). Each extracted subraster is defined in column number by the values of X1 and X2, and in row number by Y1 and Y2. Note that the values of X1, X2, Y1, and Y2 are in binned pixels if the CCD is read in a binned mode. The final subraster includes the requested number of overscan columns. After entering the values for the different subrasters the setup is applied by clicking the APPLY button, and the window is closed by clicking on the DONE button. The setups can also be saved into a file and loaded from a file. If a single subraster is used, the resultant image has a filename with the normal format, ccdnnn.fits. If more than one subraster is defined, the filenames have the format ccdnnn_1.fits, ccdnnn_2.fits, etc., depending on the total number of subrasters.



    Figure 3. Layout of the CCD Subraster Window
  • Save: If the Test flag is set, the Disk File number is NOT incremented at the end of an exposure. If the SAVE flag is set, then the Disk File number is incremented at the end of an exposure.
  • Xbin and Ybin: The CCD can be read in a binned mode, and Xbin and Ybin define the binning in the column and row directions, respectively.
  • Macro: This entry is the name of the macro defining the sequence of commands for the Macro function. This Macro must be defined in the file ccd.macros. See Appendix B for more information.
  • Gain: The command window allows for the selection of one of 3 CCD gain settings. All of the Tek CCD chips have full well capacities of > 10**5 electrons before any deviations from linearity set in. However, the electronics for the CCDs cannot have output data numbers greater than 32767, and this limits the dynamic range of the data for some observing projects. The CCD electronics have been adjusted so that the gain setting (1, 2, or 3) for each camera roughly corresponds to the actual gain in electrons/data number in the final data. This variable gain setting is accomplished by manipulating the CCD clock timing, and the result is that larger values of the gain setting give a larger dynamic range in the data, faster readout times, and higher read noise values. Note that the bias level rises with LOWER gain value. The cameras have been adjusted so that the bias level is about 100 data numbers for a gain setting of 3 for the Tek CCDs.

    Table 4 presents a summary of the actual gain and read out noise values for the CCD detectors at LCO.


Table 4. Gain and Noise Measurements for the CCDs
Chip: Tek#1 Tek#2 Tek#5 SITe#1 SITe#2 SITe#3
Gain #1 (e-/DN) 1.2 1.4 1.0 ... 0.83 ...
Read Noise #1 (e-) 4.2 6.0 5.6 ... 3.1 ...
Gain #2 (e-/DN) 2.0 2.7 2.0 ... ... ...
Read Noise #2 (e-) 5.6 7.2 6.6 ... ... ...
Gain #3 (e-/DN) 3.5 3.9 3.0 2.2 ... 2.5
Read Noise #3 (e-) 8.4 9.4 7.0 7.0 ... 6.6


3.4 Readout time for CCDs

The read times for unbinned frames are 35 sec, 24 sec, and 20 sec for the Tek#1 and Tek#2 cameras at gain settings of 1, 2, and 3, respectively. The values for the Tek#5 camera are 120 sec, 84 sec, and 68 sec, and for the SITe#1 camera 114 sec, 92 sec and 77 secs. SITe#3 reads out in 128 secs for gain 3, the only available.

Various messages about the status of an exposure and any filter changes are displayed in the bottom section of the command window. Similar messages about the status of the camera, exposures, and filters also show up on the ``CCD_Dialog'' xterm window. Diagnostic messages will also appear in these areas if there are problems with the operation of the CCD, filter and shutter hardware.


4. Guiding an Exposure

Both the 40-inch and the 100-inch telescopes have offset guiders which allow you to guide on stars near the field you are imaging. If you need to search for a guide star, the telescope operators will assist you.


4.1 Selecting Guide Stars

By far the easiest way to find guide stars for either telescope is to use the GMap Guidestar Tool. GMap actually consists of two different tools: Skymap which displays guidestars available in the SAO, PPM, or HST Guide Star catalogs for any position on the sky, and Airmass which plots a graph of airmass vs. U.T., again for any sky position. See the manual for further details on the setup and use of GMap on either the 100-inch or 40-inch telescopes. The output of GMap is the offset guider xy coordinates for the guide star that has been selected. These are entered in different ways at the 100-inch and 40-inch telescopes.


4.2 The 100-inch

The guider/finder assembly for the 100-inch is presently separate from the data systems. See the Instrument Mounting Base manual in the 100-inch observing room.


4.3 The 40-inch

The guider/finder assembly for the 40-inch is controlled by the Sparc 5 and PC. Once you are logged in as obs40, in the console window type:

After the guider has been initialized, the guider window will appear (see Figure 4).



Figure 4. Layout of the 40-inch Guider Control System


The guider provides the following functions:

  • INIT: This moves the guider against the limit switches in the four directions, initializing the instrument.
  • PARK: This moves the guider to a parked position with the pickoff mirror out of the optical beam.
  • KEYPAD: The keypad controlling manual motion of the guide probe is turned off and on with these buttons.
  • CURRENT POSITION: Displays the current position of the guide probe. The units are approximately arcsec.
  • CURRENT CENTER: The defined position of the optical axis in guider units. As of this writing (February, 1999) the centerfield is at X = 88, Y = 1916). Note that this position is needed in the setup file for the guide star selection program described in Section 4.1.
  • SETCENTER: This defines the location of the Current Center by copying the Current Position cocordinates into the Current Center display.
  • CENTERFIELD: Move the guide probe to the centerfield as defined by the Current Center coordinates.
  • OFFSET: Offset the guide probe from the Current Position by the offsets defined by DX and DY.
  • GOTO: Move the guide probe to the position given by X and Y.

If the keypad is turned on, then the guide probe can be moved "manually" by clicking on the keypad, either in the cardinal directions or in a diagonal direction. The center button toggles the keypad speed between slow (1 guider unit per move) or fast (10 guider units per move)


Once the guider window is up and running define the approximate center field by setting X = 88 and Y = 1917 and clicking on the GOTO and SETCENTER buttons. If you have predefined guide stars then you can enter the positions in the X and Y fields and go to these positions with the GOTO button.


If you do not have predefined guide stars or if you can't find any of your predefined guidestars in the field you should PARK the guider and then set DX=200 and step the guider using the OFFSET button until you come to a star. If this fails put in one step with DY= -200 and then repeat the steps with DX= -200 to scan back across the field. The field is unvignetted for all Y values < 520, and you should try and keep X between ±2000. If Y > 520, then X must be in the range -1570 < X < 1690. Apart from that the hard limits on the guider stage are around -3000. An error will show in the guider window if you try to exceed these.


Be careful not to ask the guider to do too much at once, including changing the guider keypad speed while doing a GOTO move, or by issuing guider commands while changing the filter. Such conflicts on the serial control line can hang the guider window. If the guider window hangs then quit the guider control window. Log into the pc (the name of the PC is simply ``pc'', therefore type ``rsh pc'') and type:

    ps -ax | grep gdr

This will output a list of jobs, with the process number (pid) in the first column. Then type:

    kill -9 pid

and then restart the guider window.


5. Data Storage and Tape Recording


5.1 Data Storage

The data are written in fits format on the data disk and are called ccdnnn.fits where nnn is the number of the image.

NOTE: It is especially important to keep track of space if you are using one of the 2048x2048 CCDs in unbinned mode! Each image is ~8.5 Mb, and if you have biases, flats, standards and/or many short exposures, you may fill up your data directory in one or two nights!


At the du Pont 100-inch telescope, there is one partition large enough for data storage (/data1 in Canopus), of ~8.8 Gb. We recommend that data be taped on a night-by-night basis.


On the Swope 40-inch telescope, there is currently one ~16.6 Gb partition (/data1) for data storage on Henrietta.


5.2 DAT Tape Recording

DAT drives are available on the Sun workstations at both the 100-inch and 40-inch telescopes. Basic storage capacity, which depends on the type of tape you have, is given below. Note that the DAT drives currently available on Canopus and Henrietta support only DDS and DDS2 tapes. A DDS3 drive is available on Andy, and a DDS4 drive on Rigel (the Sun Ultra-2 in the du Pont console room which is part of the WFIRC data acquistion system).


Type tape length Capacity
DDS 60m 1.2 GBytes
DDS 90m 2 GBytes
DDS2 120m 4 GBytes
DDS3 125m 12 GBytes
DDS4 150m 20 GBytes

To write a tape, put a DAT data tape in the DAT drive and run the following commands in a `/bin/csh' window. If you need to open a `/bin/csh' window, press the right mouse button against the background, selecting `Programs' and then `Command Tool'. (If you are on the 40-inch telescope, substitute ``obs40'' wherever you see ``obs100'' below.)


cd /data1/obs100/ccd_data (or wherever your data happen to be)
mt -f /dev/rmt/0ln rewind set the tape at the beginning
tar -cvf /dev/rmt/0ln *.fits & fits data to tar archive file 1 in the background
mt -f dev/rmt/0ln rewind reset the tape at the beginning
tar -tvf /dev/rmt/0ln check that data are on the tape by listing tarfile contents to screen
mt -f /dev/rmt/0ln rewoff to rewind tape and eject it
rm *.fits to delete data if you want to or need disk space
Now suppose you want to append data from night 2 to the same tape. This procedure requires caution because it is very easy to erase your data if you make a mistake!



cd /data1/obs100/ccd_data (or wherever your data happen to be)
mt -f /dev/rmt/0ln rewind set the tape at the beginning
mt -f /dev/rmt/0ln fsf 1 fsf $=$ `forward skip files'
tar -cvf /dev/rmt/0ln *.fits & write the fits data to tar archive file 2 in the background
mt -f dev/rmt/0ln rewind reset the tape at the beginning
mt -f /dev/rmt/0ln fsf 1 forward skip to tar archive file 2
tar -tvf /dev/rmt/0ln check that data are on the tape by listing tarfile contents to screen
mt -f /dev/rmt/0ln rewoff to rewind tape and eject it
rm *.fits to delete data if you want to or need disk space


  1. Use `mt -f /dev/rmt/0ln fsf n' to skip over `n' tar archives. The `n' tells the tape drive not to rewind at the end of the command. The commands `rewind' and `rewoff' should override the `n', but to be safe, don't use `n' when rewinding or taking the tape drive offline. The above sequence of instructions can be written in a file (we will call this file a ``script''), which can be executed in order to proceed with the instructions. This file contains one command per line (for example: tar cvf /dev/rmt/0ln ir*.fits). To run your script (named for example save.csh), you simply type in the csh window (be sure to be in the correct directory): csh save.csh &. The & symbol put the procedure in batch mode. This allow to you to run the script, save all the tars as necessary, while you are eating or sleeping or whatever.
  2. Again, WATCH the disk space, especially if you are using one of the 20482 chips in full format. Each image is ~8.5Mb!
  3. To remove from the computer disk data that you have successfully backed up on a DAT tape, you can use the command `rm ccd*.fits'. This will remove ALL the ccdnnn.fits files in whichever directory you are located in. Make sure you are deleting the correct data first!

5.3 Exabyte Tape Recording

A single Exabyte drive is currently available at LCO. It is attached to a Sparc 5 (Andy) in the computer room at the 100" telescope. All the commands defined in Section 5.2 are valid with the Exabyte system. The capacity of an Exabyte tape is 2 Gb.


6. Things to Watch Out For


6.1 Telescope Focus

Once the telescope has been focussed, it can change as a function of airmass and temperature. These variations are tracked at the 100-inch by the telescope control program over a reasonable range of airmasses. Note that the focus is azimuthally dependent at airmasses of greater than about 1.7. This amounts to some 20 units, and is not included in the control program. See the manual in the 100-inch observing room for more information. At the 40-inch, frequent monitoring of the focus is recommended; see the observing hints in Section 8.1.


6.2 Bias Levels of the CCDs

It is very important that the bias level for any of the CCD chips be set higher than zero for whatever gain setting you are using. For example, with the Tek#1 at a gain setting of 3, the background level present in a zero-second bias exposure should be ~100 DN. This corresponds to ~400 DN at a gain setting of 2, and ~800 DN at a gain setting of 1. For the SITe#1 the bias level (and sigma) are ~400 DN (~13 DN), ~830 DN (~9.0 DN), and ~970 DN (~6.0 DN). For the SITe#3 the bias level is ~280 DN (~5.0 DN). If there appears to be a bias level problem, please talk to the staff scientist or the electronic technicians At the time of this writing, Tek#1 definitely seems to have a bias drift. See the discussion of bias frames in Section 7.1.


6.3 Precession of Astrometric Coordinates

Both the du Pont 100-inch, and the 40-inch Swope telescopes can work with coordinates at the epoch you want. However, whatever the reason, if you need to precess your coordinates to another epoch (for example from B1950.0 to J2000.0), you may use the IRAF routine PRECESS, which is found in the directory noao.astutil, in all computers at LCO. Quick instructions on how to use this package can be found in Appendix C.


6.4 Saturation

All of the Tek and SITe CCDs chips have saturation levels (in electrons) above the digital 32767 saturation, even with a gain of 3.


6.5 Bad Pixels

The Tek#5 CCD chip has a number of bad columns, at [773:773,1:2048], [1012:1012,1:2048] and [1015:1015,1:2048] and [1992:1992,1:2048]. The second set of bad columns are very near each other, and very near the center of the chip, so be careful.


6.6 PC Reboot

If, for any reason, you need to reboot the 486-PC while it is running Lynx, make sure that Lynx is stopped before turning off the 486-PC. Failure to do so leads to a very long reboot. In order to make sure Lynx is stopped, you need to log in as superuser (username su) on the 486-PC. Then type:
    reboot -a

You may then switch the computer off when the screen goes blank or when the rebooting starts. Always wait ~10 seconds after power cycling before switching the computer back on.


6.7 Sun Workstation Problems

If you have problems with any computer (crash problems, R.I.P. procedures, etc...), please contact the staff astronomer on the mountain. Otherwise, contact the electronic technicians.


6.8 Dewar Temperature

The dewar temperature is displayed on digital voltmeters in the observing rooms. No record of the temperature is saved in the image header.


6.9 Tek#5 Imaging at the 100-inch

The Tek#5 is physically large! In order to avoid vignetting, on the blue guider control box in the 100-inch control room, make certain that the diagonal mirror is in Position A, NOT C.


6.10 Updating the Universal Time on the PCs

The UT that appears in the frame headers is the UT at the start of the exposure, and comes from a clock in the data acquisition PC computer, and can be used to compute airmass. Take a short exposure while watching the UT display in the control room. Then use IMHEADER with the ``l+'' option in IRAF to compare the header UT with the telescope's UT clock. If the two differ, you can set the UT in the PC. Log onto the PC as root. (Passwords can be obtained from the day crew.) Then type (choosing a UT time 30 seconds or so from the current time):
    date -z0
    date -d0

Hit the return button for the last command just when the telescope's UT clock reaches the correct time. Then log out by typing:


For example, to set the UT to 15:09:10 on July 11, 1996, you would type:

    date -z0
    date -d0
    date 199607111509.10 Hit return at 15:09:10!

7. Images You Will Need for Data Reduction

This section describes the basic images you will need to correctly reduce, calibrate, and analyze your data.


7.1 Biases

Information about the bias level is included in each image in two ways. First, at the end of every readout, 32 extra lines are read and averaged and then stored in the last row of the picture. This gives some information about the average variation in the bias structure across the columns of the image. In addition, each image contains the traditional extra columns of overscan which monitor the mean level of the bias during the readout. The default number of overscan columns is 16, this can be set to any value when the command window is started (see Section 2.2).

While this monitoring of the bias level will suffice for most imaging projects, more detailed observations of the bias level are suggested for spectroscopy and narrow-band imaging. The two most frequently used CCD's (SITe#1 and TEK#5) both show structure in the bias that changes across the face of the CCD in a ``bowl'' shaped pattern. This is actually not bias structure, but charge that is injected by the parallel clocks during the readout of the CCD. This structure seems to be stable from frame to frame and amounts to a few electrons on the edges of Tek#5 when it is read out in a 1x1 binned mode. The overscan columns can still be used to correct any line-to-line instability of the actual bias level.

7.2 Flats

There are several different types of flat fields, and what combination you choose to use depends on the type of observing program you have. Twilight flats are suggested for observations blueward of the V-band at both telescopes, and indeed work best at all wavelengths.


7.3 Darks

In general, the dark current values for the current generation of CCD chips are small enough that dark exposures are not necessary. However, if you are doing doing very narrow-band imaging, you will need darks.


7.4 Standard Stars

Even with the 40-inch, the quantum efficiencies of the thinned chips are such that short exposures are required to not saturate the brighter Landolt standards, especially in the red. Nevertheless, it is advised that exposure times shorter than 5 seconds be avoided. Coordinates and finder charts for Landolt standards are provided at both telescopes.


8. Observing Hints


8.1 Focusing the 40-inch Telescope

Here are some notes from Ian Smail on focusing the 40-inch:
  1. If you use IMEXAM in IRAF to look at your images you'll find that the telescope astigmatism means that you have PA > 0 (PA = position angle) when you are below true focus, PA < 0 above. The image becomes significantly elliptical far from focus - so if you are seeing ELLIP > 0.07 you should consider refocusing - if only by stepping the focus for the next science exposure and seeing if things get better. When seeing is good you should be able to make significant improvements with steps of only 0.01-0.02 units. The best seeing I've seen is around 1.6 pix FWHM - which is the limit of the sampling at 0.696 arcsec/pix with the SITe#1 CCD camera.
  2. The claim is that if the CELL temperature goes up you should move the focus up by 0.02-0.03 units per C°.
  3. Another claim is that as the airmass increases you should move the focus up. This is especially true at high airmass. Be warned that the secondary can `stick', resulting in small changes in the focus producing large FWHM variations.
  4. We tend to just use large changes in temperature/airmass as signals to be aware of possible focus variations. Then use the shape of the PSF from the data frames to correct the focus interactively.

Appendix A: Changing the Driver

If the incorrect driver is loaded on the 486-PC, you will need to change it. This is accomplished by changing the pointers on the 486-PC to the correct word in a directory, either ``ccddriver'' or ``irdriver'', and then rebooting the 486-PC. To do this, first login to the 486-PC with the username obs. (Same password as the obs40 and obs100 accounts on the Sun workstations.) Then type:
    cd /data1/obs

Now check to see which driver is loaded by typing:


You should see four files or directories, i.e.

eng ``drivername'' rebootpc rccd

Now you want to change ``drivername'' to ``ccddriver''. Type:

    rm ``drivername''
    touch ``newdrivername''

At the end of the reboot, the correct driver should be loaded. Check the last line before the ``user name'' prompt. You will see either ``ccddriver'' or ``irdriver''. If ``ccddriver'' is loaded, you are ready to go back to the Sun workstation and start the command window.


Appendix B: Writing Macros

The macro language allows the observer to string together a list of commands to the CCD command window, including setting the loop counter, changing the filter, and setting the exposure time. All the macros are defined in a file called ccd.macros which must be located in your data directory (working directory). The macro name is entered in the Macro field of the CCD window, and is read from the ccd.macros

file and executed when the Run Macro button is clicked in the CCD window. All exposures run under macros take the values for the next disk number, the binning, the sub-raster, etc., from the CCD acquisition window, updating them as with a normal exposure.

There are four basic commands recognized by the macro interpreter which control data acquisition:

  • time: this sets the exposure time in seconds, format ``time=nnn'' or ``time nnn''.
  • f: this sets the filter, using the names as defined in your filter.names file, format ``f=filternumber'' or ``f filternumber''. If no filter.names file exists, the filter names are 1 through 6 for the 6 positions of the filter wheel.
  • loop: this sets the loop counter, format ``loop=nnn'' or ``loop nnn''. Note that even if only one exposure is required, you must set `loop 1'' in the definition of your macro.
  • start: this starts the exposure, format ``start''.
A macro definition consists of the macro name followed by a string of commands, each separated by commas. One macro is defined per line in the ccd.macros file. For example, if your filter.names file has filters U, B, [OIII], H-alpha, and I, in the positions 1, 2, 3, 4 and 5, respectively, you might define
    blue loop 2, f=2, time=120, start
    halpha loop 4, f=4, time=90, start

where f denotes the filter to use, in the example filters B and H-alpha. Exposure times are in seconds.


To execute the macro ``blue'' you would then enter ``blue'' in the Macro field in the CCD window, and click on the Run Macro button.


You can also define some macros for your convenience, and then use these definitions in other macros. For example:

    bflat loop 10, f=2, time=30, start
    uflat loop 10, f=1, time=70, start
    oiiiflat loop 10, f=3, time=50, start
    doflats bflat, uflat, oiiiflat

If you wish to have the integration time, filter name or loop count input as a variable, then define that variable in the macro as in:

    int time=$, start
    bx loop 1, f=1, int $
    vx loop 1, f=2, int $

Then when running the macro in the command window you need to provide the value of the variable in the macro name field, for example by entering ``bx 20''. Note that multiple variable macro definitions DO NOT WORK. For example, the following is not permitted:

    dotwo bx $, vx $

There are two additional macro commands enabling macro control of programs running externally to the CCD command window: ``run'' and ``runlp''. These each take an argument which is the name of a file located in the working directory of the Sun -- i.e., the directory from which the acquisition window was launched. This file name should not include a path, and must be no more than 20 characters long.

  • run: this command works as follows. Suppose the macro ``testmac'' is defined in the ccd.macros file:

      testmac run syscmd, start

    The file given as the argument to ``run'' executes on the pc when the observer clicks on the ``Run Macro'' button. This file, then, should contain commands interpretable by the Unix c-shell csh. As an example, suppose the file ``syscmd'' located on the Sun had the following contents:

      rsh ss ps -a
      ps -a

    When the macro ``testmac'' is executed, the following occurs. First, the Unix command ``ps -a'' executes on the Sun. That is, the Unix report process status command executes showing the status of frequently requested processes. The output is directed to the stdout of the acquisition window, which physically is the CCD_Dialog xterm window displaying on the Sun. Then the command ``ps -a'' executes on the PC, displaying again on the CCD_Dialog window. These processes will all run with blocking unless otherwise requested (by using ``&'' arguments or programmatically) -- i.e., the macro processor will wait till the ps listings are complete before continuing. Thus in this example we see how to synchronize the execution of both local and remote processes with macro operations, and if desired with picture taking.

  • runlp: this command works in a similar, but somewhat different fashion. The argument to ``runlp'' is also the name of a file on the Sun that holds cshell commands to be executed on the PC. However the specified file is executed only inside a loop started in a macro. The command file is run before every integration in a loop (including loops of 1). Also, the current value of the loop counter is passed as an argument into the command file from the macro processor. This would be very useful for operations indexed to the loop counter, such as manipulating a Fabry-Perot, offsetting the focus, offsetting the telescope, or any other synchronous operation. Note that the command file execution will be a blocked operation if the commands contained therein are blocked -- i.e., run with wait, as opposed to shedding off into the background). As an example:

      testmac1 loop 4, f=3, time=10, runlp syscmd, start

    Note that only one "runlp" can exist in any nested stack of macros.


Appendix C: Precessing Coordinates in IRAF

To use this program, load the package by typing (in the IRAF-Xgterm window):



If you wish to precess a few coordinates, the easiest thing is simply to do it interactively from the terminal. If, for instance, you want to precess coordinates for three objects to the current epoch (say 1996.5), two of which are in epoch 1950.0 coordinates, and one of which is in epoch 1900.0 coordinates, you would type:


    precess STDIN 1950 1996.5


And then you would type:


    12:30:10.12 10:18:27.5


The program will return:


    12:32:31.56 10:03:04.10 1996.5


Or you could type:


    12:30 -20


The program will return:


    12:32:26.06 -20:15:23.47 1996.5


Or you could even type:


    12:30 -20 1900


The program will return:


    12:35:03.38 -20:31:55.10 1996.5


And when you are finished type:




The program will return:


    ERROR: software terminate (interrupt)


If you have many coordinates, you will probably want to read in a file containing the coordinates you want to have precessed. For instance, suppose the file ``coords'' contains the coordinates from the previous example -- i.e. if you type:


    page coords


The printout of the file would look like:


    12:30:10.12 10:18:27.5
    12:30 -20
    12:30 -20 1900


In this case, you would type:


    precess coords 1950 1996.5 > output_file


where ``output_file'' is the name of the file you wish to save the precessed coordinates to. If you type:


    page output_file


The printout of this file would look like:


    12:32:31.56 10:03:04.10 1996.5
    12:32:26.06 -20:15:23.47 1996.5
    12:35:03.38 -20:31:55.10 1996.5


Another way to precess coordinates is using the guidestar program. See the section on the guider cameras for more information on this program.


Appendix D: Image Reduction Facilities at LCO

It is possible to do some basic image reduction on the mountain using the computers located in the computer room in the first floor of the du Pont telescope building. Two Sun workstations are available for data reduction, both with several gigabytes of disk space: an Ultra 5 (named Dziadzio) and a Sparc 5 (named Andy). The visitor's username is ``cruncher''. In addition to IRAF, the ``sqiid'' and ``dimsum'' packages, useful for reducing near-infrared observations, and SExtractor (version 2.0.8, July 1998), specifically designed for faint galaxy photometry, are also available. Manuals are located on the shelves in the computer room.


Appendix E: Filters at LCO

A limited number of 3"x3" filters are available for direct CCD imaging at LCO. See the filter web page for details.


Appendix F: Responsible Individuals

Table 5. People to Contact
Person E-mail Phone Number Phone Location
Mark Phillips mmp at 56-51-224680 office
56-51-243848 home
Bill Kunkel kunkel at 56-51-224680 office
56-51-241721 home
Miguel Roth miguel at 56-51-224680 office
56-2-2289049 home
Ian Thompson ian at 1-626-304-0225 office
1-626-798-5751 home


Document Actions