INT/WFC Operation Manual

Introduction
Areas of the telescope
Safety
Preliminaries
Moving the telescope out of access park
Moving the telescope into access park
Zeroset
Initial TCS Commands
Loading catalogues
Readout speed
Filling the cryostat
Opening up
Twilight sky flats
Single star calibrate
Focusing the telescope
Observing
- 16.1 Lower shutter
- 16.2 Acquisition
- 16.3 Filters
- 16.4 Guiding
- 16.5 Not guiding
- 16.6 Dithering
- 16.7 Exposing
- 16.8 Windowing
- 16.9 Apertures
- 16.10 Observing scripts
- 16.11 Examining data
Special observations
- 17.1 Observations below 33 degrees
- 17.2 Observations of moving targets
Closing down at the end of the night
Closing down in a hurry
Saving data
- 20.1 Recording data on DVDs
- 20.2 Direct to laptop
- 20.3 Writing tapes
Creating the observing log
Filling the logbook
Leaving the building
Observers' handover check
Acknowledgements

1. Introduction

In this guide, the following syntax will be used:

The Systems prompts will appear in the text as type writer fonts, in the following format:
```
 USER> Commands to TCS (window on Telescope Control System)
```
```
  SYS> Commands to ICS after startobssys
       (Instrument Control System, pink window)  
```
```
   cl> IRAF commands
```
```
    $> Any other terminal window eg: xgterm
```
Typewriter text indicates that that exact command should be used. For variable input, curly braces will be used, e.g {your_input}. So, where the example might say:
```
  SYS> sky {time}
```
to take a sky flat of length {time} seconds, at the prompt you would type:
```
  SYS> sky 10
```
to take a sky flat of length 10 seconds.
Words in green correspond to actual buttons to be pressed.
Words as hyperlinks go to corresponding pictures or further details.

2. Areas of the telescope

The control room is located at the top floor of the building, and has the following key areas:

The Engineering Rack, with all manual movement controls.
The Engineering Console, which contains the dome lights, engineering button lights, Emergency Stop and reset buttons, and the Computer/Engineering mode switch. There is also the TV showing the inside of the dome, and the speakers providing the sound.
The ICS Screens, (Instrument Control System). A Mosaic of the various ICS windows can be seen here: ICS Windows
The Meteorological Data Screen.
The Whiteboard, containing lots of details and information.

3. Safety

Before starting work at the telescope please make sure you read this important safety information!
Please print and read this document very carefully. If you agree with all the items, please sign it and hand it to your support astronomer.
Please use a lone worker alarm when you are observing alone at the INT. They can be found on the back wall of the control room (see here), on a black rack. They are specific to each telescope so should never be taken away. The orange buttons should only be pressed in a real emergency.
For security reasons you must follow the instructions given to you by the support astronomer or WHT Telescope Operator (eg. abandoning site in bad weather or regarding the operation of the telescope and instrument).
For safety reasons there is a laser curtain installed in the elevator that will cause the elevator to stop if the light path is broken by a person or object. If this happens you may be stuck in the elevator. Also, due to its age, the elevator is not reliable. For these reasons we recommend that you don't use the elevator outside the Operations Team working hours (9 - 16h), nor on weekends.

4. Preliminaries

In the control room, first check the Operations Logbook for any restrictions that may be still in force. In the upper-left corner of the Operations Logbook, write the names of all observers who will be at the telescope during the night.
Turn on the dome lights (MAIN ON/OFF) from the Engineering Console and raise the control room blind.
Make a visual inspection of the observing floor (and the balcony above the control room if the telescope is at access park) to check that the telescope is free to move e.g. no ladders or steps are close to the telescope or test instruments connected.
Clear any emergency overrides by pressing the RESET button next to the EMERGENCY STOP button on the console. If the light does not clear, check for emergency stop buttons that may have been pressed.
Turn on the OIL PUMP from the engineering rack. The oil pump must be switched off at the end of the night, or when the telescope has not been moved for more than an hour (e.g. during bad weather).

5. Moving the telescope out of access park

Turn the DEC QM variable speed control fully clockwise.
Turn and hold the limit key-switch clockwise in the OR (override) position following the next few steps. The limit key-switch causes the slow motion clamps to re-engage.
Push the DEC QM button to go back to Quick Motion. Note that both the SM and QM lights are lit when in Slow Motion. It is in Quick Motion when only the QM light is lit.
Push the DEC[-] button. This will move the telescope up and away from the balcony.
When declination is below 90 degrees you may release the limit key-switch. If the audible alarm sounds, push the ACCEPT button on the Engineering Console to mute it.
To bring the telescope to zenith manually continue to push the DEC[-] button and stop at DEC=28d 45' (+/- 1 degree). This needs a bit of experimenting because the telescope will not stop immediately when you release the DEC[-] button. The telescope is now at zenith.
Push the DEC SM button to re-engage the slow motion clamps. When Slow Motion is activated only the SM light is lit.
Alternatively, once you are out of the override limit, switch to computer mode by turning the key on the Engineering Console. In order to do this four conditions must be met:
- There must be no alarms pending (press the RESET button next to EMERGENCY STOP on the Engineering Console to clear).
- The oil pumps must be switched on.
- The telescope must be out of the override limit.
- The declination and hour angle must be in Slow Motion (DEC[SM] and HA[SM]).
Now, type:
```
 USER> park zenith 
```
To park the telescope at zenith.

6. Moving the telescope into access park

If the telescope has been zeroset since the last time the TCS was re-started then in Computing Mode type:
```
 USER> park zenith 
```
```
 USER> park ap1 
```
If you're not sure then try typing the commands: if the telescope hasn't been zeroset it won't move.
Switch to Engineering Mode by typing:
```
 USER> eng
```
At the Engineering Rack, turn the DEC QM (Quick Motion) variable speed control fully clockwise, to quicken the process.
In order to move to 106 degrees (access park), using your left hand, turn and hold the limit key-switch clockwise in the override position following the next few steps. The limit key-switch causes the slow motion clamps to re-engage.
Push the DEC QM button to go to Quick Motion.
Push and hold the DEC[+] button; the telescope will then stop at access park by itself. There is no danger in keeping the DEC[+] button pressed.
Now you may release the limit key-switch.
Bringing the telescope down may cause an alarm to activate as the mirror nears the vertical. Push the ACCEPT button on the Engineering Console to mute the alarm.

Had the telescope not been zeroset since the last time the TCS was re-started then the entire procedure would have to be done in Engineering Mode. You would first switch the telescope to Engineering Mode, as in point (2) above. Using Quick Motion you would move the telescope first to zenith. At zenith the telescope should be vertical, with the HA synchro indicator pointers at the "12 o'clock" position (HA = 0h), and the DEC indicator pointers around the "9:50" position (DEC = 29d). After this, using only the DEC controls, you would bring the telescope down towards Access Park, still in QM. The telescope should end up pointing in the direction of the control room window. Only when the telescope can no longer move any further down do you need to start using the override key switch, after which the procedure is as above, from point (4) onwards.

To avoid having to do this it is easiest to park the telescope at zenith before re-starting the TCS, should you be forced to do so.

7. Zeroset

The zeroset of the telescope should be done only in the afternoon of the first night of your run, and whenever the TCS is restarted. It is also the first action to take if the telescope pointing is poor.

Park the telescope at the zenith position (see section on moving the telescope out of access park), and switch to Engineering Mode using the command:
```
 USER> eng
```
Check that the pointers on the Engineering Rack are on the correct sides for the zeroset, as indicated by the labels. If they are not you should move them to the correct sides using the HA[+]/[-] and DEC[+]/[-] buttons. The light above each dial will flash as the telescope moves through the zeropoint. This should be done in slow motion (SM).
In the TCS User window, enter:
```
 USER> zeroset ha target         [or: zero ha tar]
```
```
 USER> zeroset dec target        [or: zero dec tar]
```
n.b. These commands need to be typed in pair. If you type only one and then zeroset only that axis, it won't work.
Using the HA[+]/[-] and DEC[+]/[-] buttons with slow motion (SM) selected and the SM variable speed control set to about half-way, move the telescope so that the synchro dial pointers move through the zeroset marks in both axes. This should be done in the same order they were typed, hour angle then declination.
The TCS will beep if the zeroset is successful. The yellow LEDs mounted above the synchro dials will also briefly flash at the same time confirming that the telescope has been zeroset.

Note: Occasionally when moving through the zeroset position the yellow LED will flash but the TCS computer does not beep to acknowledge the zeroset, or vice-versa. If this happens, finish the entire zeroset procedure. Then move the telescope back to the pre- zeroset position, and redo the zeroset.

      USER> cal def                   Loads the default TCS pointing model.
                                      Use on the first night of run.

      USER> cal last                  Loads the previous TCS pointing model. Should
                                      be used on all subsequent nights of the run.

8. Initial TCS commands

These should all be typed now so that the telescope is ready for the night's observing. If the focal station is not the one you expect, contact your support astronomer.

      USER> show focal                Shows the focal station setup.

      USER> focus {x}                 Sets focus to a good value,
                                      where x is the value shown
                                      in the log book and on the whiteboard.

9. Loading catalogues

Start by checking which objects are loaded into the TCS current list of targets.

      USER> out term

If there are objects loaded which you don't need, then the command

      USER> erase

can be used to delete all targets from the current list. Note that the catalogues will still be stored on the computer, but will no longer be loaded into the TCS.

It is possible to load your own catalogue onto the TCS with all your targets in it. They can be created with any text editor, and should be saved as {your_catalogue}.cat, with no capital letters in the filename. Each line of the file should contain one object and its coordinates in the following format:

[object] [ra] [dec] [epoch] [!comments]

So, for an example object SN 1987A:

sn1987a 05 35 27.9 -69 16 10.00 J2000 ! comment

The line must be delimited by spaces (one or more), but no tabs, and there should be no spaces or punctuation other than an underscore in the object name. They can then be loaded into the /int/cat/ directory on the ICS computer. This can be done using the scp command on the data reduction computer next to the ICS:

         $> cd directory/containing/catalogue

         $> scp {your_catalogue}.cat intobs@lpss7:/int/cat/.

The username and password are shown on the label on the computer's monitor.
The catalogue can then be included using the command:

      USER> include {your_catalogue}

Note that the .cat extension is not required.
It is also possible to add and remove individual objects using the following commands:

      USER> source sn1987a 05 35 27.9 -69 16 10.00 J2000  Creates object.

      USER> add                                             Important! Loads
                                                            object into TCS.

      USER> remove {object_name}                            Removes object.

      USER> out file {your_catalogue}                       Writes the contents
                                                            of the current TCS
							    catalogue to file.

Note that to add the object you must include the epoch. No error message will be shown if you do not, but the object will not load properly.

10. Readout Speed

Check that the readout speed being used is correct. This can be changed either from the SYS prompt using the commands:

       SYS> rspeed slow

       SYS> rspeed fast

or from the GUI by pressing the 'fast' or 'slow' button. The fast readout speed is ~10s faster than the slow. However, the fast readout adds read noise to your image, so if looking at faint objects for which it is the dominant source of noise then slow readout is recommended.

11. Filling the cryostat

Important! Wear safety helmet (with visor), apron and gloves when filling cryostats. Example.

The dewars are located at the balcony, and the telescope must be in access park for this procedure (see section on moving the telescope into access park).
On the dewar top, check that the SILVER valve (gas vent) is closed. Vessels must be left pressurised.
Use the indicator on top of the dewar, (by pressing and holding the black rubber button) to check the dewar contains enough LN2 (>25%). If it is lower, the smaller dewar can be used. Take care with the hoses as they are prone to breaking when cold.
Check the pressure gauge: the pressure should be between 0.5 and 1.5 bar. Again, use the smaller dewar if this is not the case.
Wheel the dewar carefully into position opposite the prime focus, with the earth wire, filler tube and pressure gauge at sensible positions.
Connect the dewar to earth; use the earthing point on the hand rail to the right of the telescope.
Fully insert the filler tube into the cryostat hole, going through the metal triangle, under the plastic tie.
On the dewar top, check that the GREEN pressure build valve is closed. Then open the BLUE decant valve.
Vapour will come out of the cryostat continuously while you are filling it. However, it is not full until liquid nitrogen is seen to be spilling out onto the floor. You will hear this as well.
Close the BLUE valve.
Wait a couple minutes for the filler tube to warm up (until it is not rigid anymore) and remove it from the cryostat.
Move the dewar back to the position where you found it.
Always note down in the Operations Logbook what time the cryostat was filled and the initials of your name.
The cryostat must be refilled before the start of the night's observing, and again in the morning before you leave. In the event of leaving the telescope early (e.g. due to bad weather) fill the cryostat, then fill in a fault report so that the cryostat can be filled again in the morning by the operations team.

12. Opening up

Check that the weather conditions allow for observations. Set the humidity alarm to 75%. If it's already above that level set it to 90%.
Check the binning, readout speed and windowing have been set correctly.
Check that the IRAF window and Observing Log are both showing the correct directory for the date you are observing.
Fill in the 'Who Goes There?' window with your proposal number and the name of the PI.
Open the south doors, and the double doors between these and the dome for ventilation, if not too humid (>75%). If the wind speed is >40 km/h, or you close for bad weather the South Doors will need to be shut again. It is also the observer's responsibility to shut them at the end of the night.
Turn off the lights in the area between the Control Room, the Dome and the South Doors and the Control Room Corridor.
Ensure the lift (elevator) is at least one floor down and turn off its lights. Do not ride the elevator after 16:00 or at weekends!
Take the stairs back up and turn off the lights at the far end of the control room nearest the South Doors.
Using the CW (clockwise) and CCW (counter-clockwise) buttons on the lower engineering rack rotate the dome so that it points east, away from the sun. This should leave the dome over the doors on the left as you look through the window. Note that there is a compass laminated onto the desk in front of the Engineering Console.
Turn off the dome lights on the Engineering Console.
On the Engineering Rack, in the Upper Shutter panel, use RAISE MICRO to raise the shutter slowly, until you hear a 'clunk' and the Main Overtravel light turns off.
Now use RAISE MAIN to raise the upper shutter, until another 'clunk' is heard, and the shutter stops moving.
To fully open the upper shutter RAISE MICRO must be used again, until the 'clunk' is heard and the telescope stops moving.
If observing at low elevation (<33??) then the lower shutter will need to be raised as well, in the same way as the upper shutter, using RAISE MICRO then RAISE MAIN. RAISE MICROcan then be used again to raise it the last small fraction. However, as the lower shutter is very large and blocks most of the upper region of the shutter space, this should not be done unless absolutely necessary.
Push and hold the PRIMARY MIRROR OPEN button until all the petals are fully open. The telescope needs to be at or near access park in order to make sure they have opened fully.
Turn the DOME VENT FAN on, on the Engineering Rack, to provide further ventilation. This should be turned off if the humidity rises above 75%.
Move the telescope out of the override position (see section on moving the telescope out of access park).
Close the control room blind.

13. Twilight sky flats

Include the blank fields catalog ("Table of 38 blank fields for sky flats" - newblank38 catalogue) using:
```
 USER> include newblank38   
```
Move to an appropriate star field from the catalog using:
```
 USER> gocat {field name}
```
The field should have its RA within 1 hour of the current sidereal time, which is shown on the TCS Display window. If possible choose a wfc or wfs field, as these were especially chosen to be blank over large areas for the wide field camera. If one is not close enough, then one of the other fields will do.
Select the appropriate filter using the GUI, remembering to first use narrow filters, then broadband filters, and moving from blue to red within each group in the evening. In the morning this will be the opposite way around.
Use:
```
  SYS> ftest 2
```
to perform a two second windowed glance. Note that you must be in the /home/intobs directory to use this command. This will read out almost instantly, allowing the sky levels to be determined using IRAF on the data reduction machine using:
```
   CL> display s1[1] 1; imexam
```
By hovering the pointer over the background and pressing m the pixel values will be displayed, for the area covered by the pointer. The ideal level for the flats is 22000 counts (including the bias level of ~2000), though anything from 15000 to 30000 counts should be useful.
When the background drops to an appropriate level a flat field can be taken using the command:
```
  SYS> sky {time}
```
with the time expressed in seconds. Times shorter than 2s or longer than 120s are not recommended for flats. Remember that with the same filter, assuming the level was good for the flat just taken and you are using slow read out mode, the exposure time will need to be increased by a factor of 1.5 to account for the dropping background brightness in the evening. In the morning, of course, the exposure will need to be decreased by a factor of 1.5.

The sky level can now be checked in IRAF using the command:
```
   CL> display r{number}[4] 1; imexam
```
again, using m to check the sky level. The {number} can be seen on the GUI. The fourth extension is viewed because this shows the middle chip.
It is customary to offset the flats taken in the same filter, so that the images can be combined to remove stars later. This can be done using the command:
```
  SYS> offset arc {x} {y}
```
with {x} and {y} being the number of arcseconds you want to offset by. Note that x and y are absolute values and not relative values. Therefore a different offset will be needed for each flat in the same filter to avoid having multiple flats in the same place.
When changing filters the ftest script can again be used to check the sky levels so the correct exposure time is used for the flat.

14. Single star calibrate

On the first night, the default pointing model should have been loaded in the initial TCS commands, and is likely to be accurate. However, it is good to check it and make an additional correction with the following procedure.

       SYS> agwin acq pos 1032 512    Sets readout to the centre
                                      of the autoguider chip.

       SYS> autotv on                 Turns on continuous readout mode.

On the autoguider window mark a cross in its centre by selecting TOGGLE CENTRE CROSS from the SETUP menu.

      USER> cal faint                 Sends the telescope to a
                                      suitable calibrate star.

When the HANDSET menu appears use <ctrl-z> to break out of it.

      USER> aper 5                    Changes the telescope pointing
                                      to the autoguider centre.

If the star appears right on the cross-hairs then the pointing is okay, and the rest of this section can be skipped. Otherwise, press the HANDSET button (F6) located on the TCS keyboard to centre the star on the cross-hairs in the centre of the autoguider window.

Points to consider:

Bear in mind that arrows move the telescope and not the sky. So think of it as moving the cross to the star, and not the star to the cross.
Pressing a direction button once moves the star by the increment size value shown in the TCS window, near the top. The increment can be changed by using the '<' and '>' buttons.
After pressing a key, wait until the autoguider window has refreshed showing the new position.
In case the star is either too faint or saturated, then the exposure time can be changed by going to SELECT -> INTEGRATION TIME -> SELECT FROM MENU.

Once the star is well centred, press handset and type:

      USER> point calibrate           Writes the new pointing model to file.

      USER> cal anal zero             Loads the new pointing model

The rms of this solution is typically 0.01, because only one star is being used. Type 'Y ' to accept.

       SYS> autotv off                Turns off continuous readout mode.

Note: If the pointing is still bad after the single star calibrate, then refer to the troubleshooting guide. First try zerosetting the telescope again, the default calibrate and then the 7 star calibrate if necessary.

15. Focusing the Telescope

To determine the best telescope focus select a suitable field (e.g. one of the WFC photometric standard fields). Choose either a field close to your first target, or continue using the same field as the single star calibrate. Sometimes these fields can be too crowded and in that case try a suitable blank field. Run the focus script:

       SYS> gocat {suitable standard field}

Note: Complete the focus run in a suitable filter. The reddest filter in the wheel is desirable, since it will have the smallest FWHM for a point source.

       SYS> focusrun &

A dialog box will pop up and the appropriate values may be filled in and a focus run started. The default values should be good, but the recommended values are:

Number of exposures (n)      =        9

Focus start value            =        0.2mm below last known focus (see white board)

Step size                    =        0.05 mm

Exposure separation          =        10 arcsec

Exposure time                =        15 sec

Hit RUN. All n exposures will be taken on the same frame, giving lines of repeated stars. There is a double space between the first and second exposures to make them easily identified.
Accept any error box messages and close the focus run window.
Next, type:
```
   CL> wfc_starfocus
```
to compute the best focus value from the focus run. It will ask you which run number the focus run is, and the number can be found on the GUI. It will know all the values you selected for the run. If the command is not recognised then you may need to load the package for it by typing:
```
   CL> wfc_ql
```
It will ask you to identify as many first stars as you want, 10-12 being typical. This is done by placing the cursor over the first star of a set and hitting 'm'. Make sure that the whole line of stars is on the CCD, and that the stars you pick are clear single stars and not obscured or saturated. Pressing 'q' when finished produces a graph showing what it believes to be the best focus value. Quit display window with 'q' before closing it.

Remember to set the best focus value at the end of the focus run using the command:

      USER> focus {value}

If the target is needed back at the original position then the gocat command will need to be repeated.

Enter the value obtained on the whiteboard and in the log book.

If, after running the focus script, the stars are still very unfocussed or doughnut-like instead of point-like, then the focus run can be done again using a larger step size and more steps as appropriate. A 0.1mm step size with 11 steps may be a good next try.

16. Observing

This section acts as a glossary of standard commands, but is organised into progressions of tasks and order of use. Note: it is strongly recommended to use guiding for exposures of longer than two minutes.

16.1 Lower shutter

Remember that the lower shutter starts to vignette at an elevation of 33??. The system does not warn you. If you need to observe at elevation <33?? then the lower dome shutter must be raised (see section on opening up).

16.2 Acquisition

Loading catalogues and objects is described in the section Loading Catalogues. To move to one of these objects:

       SYS> gocat {object}            Telescope will slew to that object,
                                      which must be loaded into the TCS,
                                      and track it, e.g.

       SYS> gocat 92-248              Moves the telescope to the Landolt field
                                      92-248 which is stored in the system catalog.

16.3 Filters

       SYS> filter {filter}           Changes filter. Use the same format
                                      as seen in the GUI wheel, e.g. Ha, r.
                                      Alternatively, just click on the filter
                                      wanted in the GUI.

A de-focus term will automatically be applied when you change filter. It can take a few seconds to finish changing focus, and this can be seen on the TCS information screen.

16.4 Guiding

Check that the autoguider tv-mode is off. If it is on, type:

       SYS> autotv off

When the gocat command has been executed to slew to a new field then a list of guide stars will appear in the Talker window. This list can also be seen by entering:

       SYS> more /tmp/gsc.out

If this does not occur then check that the auto predict is on (see section on finding guide stars in the troubleshooter).

The guide window size can be set in the GUI under the settings tab. A size of 120x120 pixels (resulting in a sky field 45"x45") is recommended.

A guide star of appropriate magnitude should be selected from this list. To guide on this star use the command:

       SYS> guide on pixel {x} {y}

with {x} and {y} being the same values as before. If the guide star doesn't appear within the window you can try a larger window size (for instance 180x180 pixels under "Settings" on the GUI) or do the following:

       SYS> guide off

       SYS> agwin acq pos {x} {y}

       SYS> field

This centres the AG on the position where the guide star should be and searches for potential guide stars within that field displayed. They will be marked and numbered on the autoguider. Sometimes this command needs to be issued twice before it works.

To guide on one of these stars, use the command

       SYS> guide on star {n}

where {n} will be a number assigned to each star, being 1 the brightest. Be aware that the field command often picks up artifacts, so you should always check that you are guiding on a real object. For this reason it is recommended not to use scripts for setting up guiding.

Before starting the exposure it's important to leave the autoguider to settle. When guiding, the TCS will display A/G on rather than tracking, and show the x and y errors in the guiding. These should be allowed to drop below 0.3, though if the seeing is bad the errors may not get lower than 0.5. It can take a few readout cycles for the errors to be displayed in the TCS. If you do not see them then the guiding is not working.

NOTE: When using long autoguider integration times, then increase the wfc_ag_sequencer time before readout to be integration time + 9s.

It usually takes some time to turn the guiding off (SYS> guide off). If it takes too long, then use <ctrl-c> to exit it, then check guide job ID number and kill this process:

       SYS> autotv off

       SYS> jobs

       SYS> kill 'job_number'

The command to turn the guiding back on is:

       SYS> guide on

16.4.1 Autoguider sequencer

The aim of the so-called "guiding sequencer" (the wfc_ag_sequencer program) is to avoid extra noise (interference) on science images if they are read out while autoguider is reading out. The following two cases are possible:

Observing programs using guiding (exposure times longer than 2-3 min), the sequencer must be always used, so the sequencer should remain active (red flag checked) during the whole run. Also note that any TCS restart will start/default the observing system with the sequencer active.

Observing programs not needing guiding (but only tracking) for their short science exposures (less than 2-3 min) should un-mark the sequencer by pressing in the small window on red click button (un-press it) to dis-activate. Also, observers should un-mark this option after each TCS restart.

16.5 Not guiding

For short exposures (under 2-3 min) you may want to turn off the autoguiding. This can be done using the command:

       SYS> guide off

Remember also to turn off the sequencer by turning off the enable button on its GUI.

NOTE: By default in guiding mode, the sequencer stops autoguiding 10s before the WFC science image readout, in order to avoid this noise, but you can change this value (thought, it should not be less than 8s). If you do not need guiding for your science, then you should stop the autoguiding and sequencer using the above two actions, or using the following sequence commands:

       SYS> guide off

       SYS> jobs

then look for which job is connected with wfc_ag_sequencer, and type:

       SYS> kill %n

where n is the corresponding job number. If you decide to guide later because you are doing longer exposures, if the sequencer GUI is not on screen then it can be started using the command:

       SYS> wfc_ag_sequencer &

16.6 Dithering

Dithering can be done in several ways, and which is most useful will depend on if you're using guiding and the size of the offsets you wish to use.

Offsets can always be applied using the command:

       SYS> offset arc {x} {y}

Note that the {x} and {y} positions are absolute, with the original position at {0,0}. Therefore, if you keep issuing the same offset the telescope will not move.

If you're not using guiding this will be the easiest way to perform all your offsets. If you are, however, then complications can arise, because guiding must always be turned off before using the offset command, otherwise the guiding will move the telescope back to its original position.

If you're guiding and making offsets small enough to keep the star within the field of the autoguider, then the field command may be used to locate it once again:

       SYS> guide off

       SYS> offset arc {x} {y}

       SYS> field

       SYS> guide on star {n}

Where {n} is the number assigned to your guide star.

If your dither positions extend over a range >1 arcmin then you will probably find the guide star moving out of the field. In this case the easiest thing to do is calculate multiple dither positions for the object you're looking at beforehand, and then load them from a catalogue. Doing this allows you to gocat to each new position, and when the gocat command is used new positions for the guide star will automatically be calculated. There is a tool on the NASA Extra-Galactic Database (NED) to help you convert offsets into RA and Dec.

You would then use the commands:

       SYS> guide off

      USER> gocat {new_dither_position}

       SYS> more /tmp/gsc.out                  Find suitable guide star.

       SYS> guide on pixel {x} {y}

Remember to let the guiding settle before starting your exposure, so that the errors shown on the TCS display are <0.3.

Please note that if you want to mosaic images using all four chips then, because of the chip gaps, you will need to offset by at least 30 arcseconds in both x and y. This is due to the layout of the chip, which can be seen here: offsetting in just x or y merely causes you to move along the gaps.

Should you have any questions about dithering please direct them to your Support Astronomer, preferably in advance of your run.

16.7 Exposing

       SYS> glance {time}                      Take a glance of {time} seconds. 

            Stored as s1.fits (which will be overwritten). Use it to check you're 
            on the right field or to estimate exposure times before the actual run.

       SYS> bias                               Take a bias frame.

       SYS> flat {time}                        Take a dome flat of {time} seconds.

       SYS> sky {time}                         Take a sky flat of {time} seconds.

       SYS> run {time} {"comments"}            Take an exposure of {time} seconds
                                               with the {"comments"} added to the
                                               header of the fits file.

For multiple biases, flats, skyflats, runs or glances, simply prefix the command with 'mult' and put the number desired directly after the command, e.g.

       SYS> multrun {n} {time} {"comments"}    Take {n} exposures of {time}
                                               seconds, adding the {"comments"}
					       to the files' headers.

16.8 Windowing

For time resolved photometry windowing data can be a useful way to reduce readout time. Note that only chip 4 of the WFC can be windowed. This link shows how the readout time varies with window size (and binning). The window size can be set with the command:

       SYS> window {n} "[{x1}:{x2},{y1}:{y2}]"

To enable this window type:

       SYS> window {n} enable

and to disable type:

       SYS> window {n} disable

The window size and whether it's enabled can be seen on the GUI.

16.9 Apertures

With a gocat command the telescope is sent to the RA and DEC specified for the object, so that this object appears in the middle of the centre chip (chip 4). Occasionally, it may be desirable to centre the object in a different position, and for this apertures can be used.

Aperture 0 is an absolute measurement and sets the centre, from which point all other apertures will be measured. This is the default pointing for the telescope and is set at -251" 326" (expressed in arcseconds), the centre of chip 4. All other apertures have values relative to aperture 0. Apertures 1-5 point to the middle of chips 1-5 (where 5 is the autoguider). This means that aperture 4 is defined as 0 0.

To see a list showing the different apertures set, use the command:

      USER> show apertures

If, for example, you wanted to have the object you're pointing at to appear in the centre of chip 1, then simply gocat to that object and then type:

      USER> aperture 1

This pointing will only be used for one command before reverting to its default pointing, so the aperture command should only be issued just before you want to use it.

If it's necessary to reset the pointing of the telescope for more than one target, and you don't want to keep re-typing the aperture command, then it is possible to redefine aperture 0 so that it has this new value as the default pointing. Extreme care should be taken when doing this, however, to return the telescope to its normal pointing afterwards. Also note that when this is done apertures 1-5 will no longer point to the centre of chips 1-5, since those values are set relative to aperture 0.

The command for doing this is:

      USER> enter aperture 0 {x"} {y"}

In general, this should only be necessary if a seven star calibration has to be performed, and this procedure is explained in detail in the Troubleshooting Guide.

16.10 Observing scripts

One could use observing scripts to observe a given sequence, given an object (ex. obj1) previously defined in the catalog, using standard observing commands (example of guiding with small dithering here: wfcsamplescript).

First, you have to sftp intobs@lpss7 your script (ex. johnscript) to the default SYS observing directory (/home/intobs/), then change its permission for execution there (chmod a+x johnscript). Then, in the same /home/intobs/ directory you can run the script by simply entering its name at the prompt, namely:

      SYS> johnscript

16.11 Examining data

If IRAF is not open on the data reduction pc then type 'iraf' into a terminal window. This will open up an xgterm window, a DS9 window and execute IRAF automatically.

The data being collected will be stored in the directory shown on the GUI, which is /obsdata/inta/yyyymmdd with the last directory being the date on which the night starts.

Data can be looked at in the /obsdata directory but not reduced, and must also never be moved or deleted. If you plan to reduce the data on that machine then it should be copied to the /scratch directory, in which you may make your own sub-directory.

Data can be displayed in the ds9 window using the command:

        CL> display r{number}[4] 1

This loads the contents of extension 4, corresponding to CCD 4, of image r{number}.fits to the ds9 display. The {number} for the exposure is found on the GUI, or in the night log. If the exposure was windowed then the image will be written into extension 1. If a glance was taken instead of a run, bias, or flat then the image will have been written to file s1.fits.

The image can then be examined using the imexam task:

        CL> imexam

Useful imexam commands include:

'v' for vector cut; click start and end
'c' for column cut
'l' for line
'e' for elliptical plot
's' for surface plot
'a' for text statistics
'm' to obtain pixel statistics on section around cursor
'r' for radial profile
'q' to quit imexam

Imexam commands are listed on the desk.

Alternatively, the display and imexam tasks can be combined in one line by using a semi-colon:

        CL> display r{number}[4] 1; imexam

Important: To match correct sky orientation (North up, East left, needed to correctly match star charts), in DS9 one should rotate CCD1, CCD3 or CCD4 images by 270 deg (Zoom/270), while CCD2 should be rotated by 180 deg (Zoom/180).

If you want to display all 4 different buffers (corresponding to the four CCDs) at once then this can be done using the command:

        CL> wfc_display r{number}

The 'tile' option must be selected under frame.

17. Special observations

17.1 Observations below 33 degrees

If you wish to observe at a zenith distance between 57 deg and 70 deg (33 deg - 20 deg elevation), then you need to raise the lower shutter (see section 12 of the INT/WFC Operation Manual).

There will not be any warning given from the TCS if you are observing the inside of the dome, so you will have to check the zenith distance (ZD) on the TCS monitor from time to time.

Please consider if opening the lower shutter is essential. Opening up takes around 10 minutes and the image quality will decrease due to observing through higher air masses.

Remember to bring the lower shutter down afterwards, as you will not be able to observe above 45 degrees elevation if it is left in the raised position.

17.2 Observations of moving targets

If you wish to observe a moving target, then you need to compute its motion rates in the equatorial coordinate system. The TCS requires the differential tracking rates to be expressed as:

Sometimes, ephemerides are given as

(1)

(2)

Thus, to transform these for TCS input, compute

(1)

(2)

In case the ephemeris server gives you two options, chose "coordinate motions" instead of "sky motions", so you don't have to divide da by cos(delta) anymore.

The differential tracking rates are then input to the TCS using the following sequence of commands:

 USER> gocat {object name}

 USER> diff_rates da dd

 USER> next

To return to sidereal tracking (e.g. for standards) issue the TCS command:

 USER> diff_rates 0 0

18. Closing down at the end of the night

If running, stop the autoguider:
```
  SYS> guide off
```
The sequencer should also be turned off by depressing the enable button on its GUI.
Bring the telescope close to access park by typing:
```
 USER> park zenith
```
followed by:
```
 USER> park ap1
```
Push and hold the PRIMARY MIRROR CLOSE button. Check the mirror petals close fully (petals can be seen moving through a gap in the telescope structure).
Push the DOME VENT FAN STOP button to turn off dome fans. The red light will go on when the fans are stopped.
Once the telescope is stopped, switch the TCS to engineering mode by typing:
```
 USER> eng
```
The ENG/COMP push button should now be illuminated with a yellow light.
If lower shutter has been raised close the shutter using LOWER MICRO, then LOWER MAIN, then LOWER MICRO until you hear a 'clunk' at each step.
Close the upper dome shutter pushing LOWER MICRO until you hear the 'clunk', then LOWER MAIN until the shutter stops moving, and makes a 'clunk'. To finish closing, press LOWER MICRO until the shutter stops moving.
Bring the telescope fully down into access park. (See section Moving the Telescope into Access Park).
Turn off the OIL PUMP from the engineering rack. The light will show green.
Close the South Doors if they have been opened.
Fill the WFC cryostat following the same procedures as at the start of the night. Leave the telescope at access park.

19. Closing down in a hurry

If it starts to rain or snow, the humidity rises above 90% or the winds above 80km/h you will be forced to shut down in a hurry. If this is the case:

Stop the current exposure and read out. To do this break out of the command and return the SYS prompt using <ctrl-z>. Then type:
```
  SYS> finish
```
Park the telescope using the command:
```
 USER> park ap1
```
While the telescope is moving hit the DOME SHUTRS button next to the Engineering Console, so the dome shutters may be moved at the same time as the telescope.
Close the mirror and the dome in that order (as in the section Closing down at the End of the Night), remembering to close the lower shutter before the upper if it is open.
Shut off the fans.
Close the South Doors.
If you were using a 'mult' command then typing finish will not have stopped the remaining exposures from being taken, and the run must be aborted. This can be done by hitting 'abort' on the GUI, and then by typing:
```
  SYS> abort
```
Release the DOME SHUTRS button on the Engineering Console for the dome shutters. If you forget to do this the dome will not track with the telescope.
Visually check that the telescope has returned to ap1 correctly without twisting. If it hasn't issue the commands:
```
 USER> park zenith
```
followed by:
```
 USER> park ap1
```
If shut down for a long period of time then turn off the oil pumps. Do not forget to turn them back on before you move the telescope again.
If abandoning the telescope due to bad weather then fill the cryostat before leaving, and fill out a fault report with the time you filled it so that the Operations Team can refill it first thing in the morning. Note that only time lost due for technical reasons should be included in the fault report. If abandoning due to high humidity turn the alarm off (middle setting) to stop it going off all night.

20. Saving data

With WFC, you can save your data either on DVD, on a DAT tape, or directly to your laptop or hard-drive. You only need to write your copy disk/tape, there is no need to write an archive tape.

20.1 Recording data on DVDs

DVDs can be burnt on either of the two Linux public computers in the control room (INTDRPC1 and INTDRPC2). Please refer for details to the following document (also shown on the wall to the right of INTDRPC1): http://www.ing.iac.es/Astronomy/computing/recording.html.

20.2 Direct to laptop

Data can be transferred directly to your laptop through the INGEXT network. This can be done file by file using the scp command from your laptop:

         $> scp intobs@lpss7:/obsdata/inta/yyyymmdd/r{number}.fit /your/directory

Of course the * can always be used as a wildcard. Alternatively the rsync command can be used to copy all the files from the directory on lpss7 to a directory on your laptop. Running the command another time then updates your laptop's directory only with the new data.

         $> rsync -av intobs@lpss7:/obsdata/inta/yyyymmdd/ /your/directory

The password can be found on the screen of the computer.

20.3 Writing tapes

Insert the DDS3 (125m) tape in the drive /dev/nst0, on top of the data reduction machine INTDRPC1. A DDS3 tape (12GB) should be able to store 150 runs in fits format, or about 300 using tar.

Open a new xterm on the Data Reduction machine and change directory to where the data are stored, e.g.:
$>cd /obsdata/inta/20040419

$> tar cvf /dev/nst0 *.fit
this writes all '.fit' files in that directory to tape. If you expect that this will take long time, leave a note on the keyboard to prevent the day-time staff from closing the window, stopping the tar before it is completed.

21. Creating the observing log

Fill in the night report (even if the weather was bad and you could not observe!) by selecting 'night report' which is situated in the observing logger window on the ICS monitor. Fill out all fields, and when done hit 'create log'. A copy called run_log_yyyymmdd.int is put in the observation directory so you can copy it when copying your data, and there is an option to email it to a comma-separated list of addresses. Please note that from March 2011 the online logs no longer show information on the targets observed for a proprietary period of one year. They only show the names of the observers, causes and amount of time lost and weather information for the night. If you arrive home and find you have lost your copy of the log please contact your support astronomer to obtain another.

22. Filling in the log book

The printed logbook in the control room contains fields which must be filled out during the night:

CCD cryostats : initials and local time for filling the cryostat.
Breakdown of observing time: fill in any down-time that was encountered (bad weather, technical problems, etc.) and note the time lost.
The last focus used during the night. This can be found next to focus on the TCS display.

23. Leaving the building

Before leaving the building, please switch off all major lights in the control room and in the kitchen area. Check that no electrical devices are left on in the kitchen, e.g. sandwich toaster, coffee machine, etc.

24. Observers' handover check

Ready to go it alone? Please check you are familiar with the following tasks:

Use of telephone: numbers on right edge of whiteboard.
Safety precautions when using liquid nitrogen.
Weather precautions, knowing when the dome must be closed (also on whiteboard).
Observing at zenith distance between 57 and 70 degrees (see troubleshooter).
Opening and closing the dome.
Moving the telescope in engineering mode.
Starting up and shutting down the observing system.
Knowing about the most common faults and how to solve them.
Writing your data to an external hard drive or dvd at the end of every night.
Filling in fault reports.
Filling in the log book in control room.
Producing the observing log.
Filling in the observing feedback form at the end of your run.

If, after reading through this guide and the Troubleshooter, you are still having problems you cannot solve then you can ring the telescope operator at the William Herschel Telescope. Numbers are found on the whiteboard.

25. Acknowledgements

Credits should be given to the following ING students who contributed to improve this manual: Hugo Ledo, Dora Fohring, Lee Patrick, Liam Hardy, Mansour Karami and Yudish Ramanjooloo.