It is very important to get the best possible focus with a fiber-fed instrument like Hydra. A modest error will cause a large fraction of light to miss the fiber. Hydra has the tools necessary to help the observer get a good focus, but the unwary observer can easily fool him/herself into thinking the image is well focussed when it is in fact poor.

Hydra has a camera on the gripper which shows the observer a simultaneous view of the tip of the fiber and the image of the target object. This makes it easy to determine if Hydra is positioning well and is properly focussed, right?

Wrong! Let's see how this camera works. Click here [12] to bring up a drawing of how the image is generated. The gripper camera sees the fibers by means of a "periscope", shown in this drawing. The hydra fibers come in from the side of the field, enclosed in a piece of hypodermic tubing to avoid breakage. A small prism is cemented onto the tip of each fiber. In turn each of these prisms are cemented to magnetic buttons. Hydra moves the buttons with a "gripper" which picks them up and places them at the appropriate locations on a flat steel plate. The plate is then warped into a curve which moves the tips of the fibers to the focal surface.

The periscope permits the gripper's TV camera to see both the target and the fiber simultaneously by means of a pellicle mirror. Light from the target hits the pellicle and is reflected directly onto the camera. The fiber buttons are illuminated from above with LEDs on the base of the gripper. The intensity of these LEDs can be controlled using the Hydra GUI. This light reflects off the back of the pellicle, into a collimating lens and then to a retroreflector, which passes back through the lens, through the pellicle and focusses on the TV camera.

If the light is parallel when it goes into the retroreflector and the distances are the same from the pellicle to the camera and telescope focal surface, the relative positions of the images of the target star and the fiber tip will remain unchanged wherever the images are within the camera field. Thus the observer sees star and fiber and can theoretically bring them into focus and see when light from the object is indeed going straight into the fiber.

This is what one thinks he is seeing but appearances can be deceiving. The gripper camera focusses on a plane, which is that of the unwarped plate. Warping the plate varies the position of the focal surface by up to 3mm over the field. This will increase the diameter of a star's image on the focal surface when the plate is warped by as much as the diameter of the large fibers, but someone looking at the view in the gripper camera will not see any change in the image of the target. The image of the fibers will go slight out of focus.. Additionally, as the plate warps, it pulls the fibers slightly off position. This is compensated in the positioning model which results in situations in which a fiber is well positioned, but appears to be off center or out of focus in the gripper camera.

If the telecsope focus is adjusted to make the image of a target the sharpest in the gripper camera, the telescope will be focussed on the plane of the unwarped plate. It will thus be out of focus everywhere in the field save at the very edge. The image of a fiber as seen through the periscope will only appear to be in perfect focus at the edge of the field.

Thus, though the gripper camera is very useful, it is NOT a reliable gauge of telescope focus. Telescope focus can only be judged with the FOPS fibers. These are bundles of seven fibers 100µ diameter placed on buttons and connected directly to a television camera. They are therefore in the same plane as the object fibers. One focusses by centering on one or more of the FOPS and adjusting the focus to get as much light as possible into the central fiber and minimizing the light in the peripheral fibers.

In summary, the rule is use the Gripper camera to see what is happening and to recover a lost fiber, but check focusing and positioning with the FOPS. Additionally, BE ABSOLUTELY CERTAIN that the plate is in the warped position when focussing with the FOPS. If for some reason you absolutely must focus using the camera on the gripper, do so as near to the edge of the field as possible.

There is a television camera in the telescope chimney which allows you to see the position of the comparison lamp mirrors, the corrector and everything else in the cage. It is good practice to turn it on before using Hydra to check and see if everything appears to be in the right place. In particular, the TV camera allows you to verify that the corrector has been deployed.

This camera is turned on from the TCS operator's console by going into the "command mode" and issuing the cryptic command:

local cfadc outlet1 move on

Be sure and turn it (and its associated light) off when you are finished! Use the command:

local cfadc outlet1 move off

David James (Hydra scientist): x358

Rolando (Rolo) Cantarutti (computer software engineer): x373

Andres Montane (mechanical engineer): x309

Mountain support:

Observer Support x400 (Mauricio Rojas)

Observer Support x422 (Hernan Tirado)

Electronicos x417 (Humberto Orrego, Javier Rojas, David Rojas, Enrique Schmidt)

For setup questions prior to your run, contact Hydra scientist David James. Once on the mountain, contact Observer Support for help. ObsSup is on call from about 11am-midnight. Call them if you need to change a tilt, a filter, open the dome, etc. The telescope responsibility is handed off from ObsSup to the night assistant around sunset. If you need to do calibrations right at sunset (and you do!), you must communicate this to ObsSup in the afternoon, otherwise they may all be at dinner just at the time

LINK to hydrapro.tar file not found in the server

The best way to prepare for Hydra is to become familiar with the Hydra assignment program at home, and to come to Tololo a day in advance to see Hydra being used by the previous users.

Unlike most other instruments, Hydra requires that you come to the telescope with extremely accurate positions and a well planned program. We provide you with a simulator ("hydrasim") to do the assignments by hand or a Fortran program ("hydraassign") for a brain-dead way of doing the assignments.

See hydra software [13] for both of these programs.

While "hydraassign" works well, you may want to check the assignments with "hydrasim" and maybe tweak up the final assignment.

You need to have sky positions if you want to do sky subtraction. You can add sky positions by hand using the "hydrasim" program. In addition, I have a Fortran program that will make circles of sky positions at selected radii. Contact me (Nick) if you want the program. Knut has an IDL program that also does this. You can get it as part of a tar file of other programs from http://www.ctio.noao.edu/~olsen/IDL/hydrapro.tar [14].

The hardest part of the preparation is the measurement of astrometric positions. If your data come from Mosaic II [15], then you probably already have derived good positions for your targets. Just make sure that you've checked the accuracy of the astrometry in the header against the USNO-A2, USNO-B, or other catalog. The IRAF MSCRED package has tools to do this. If your images have no astrometry in them, then you'll have to make it from scratch. The IRAF IMCOORDS package will get you started. Frank Valdes also has a web page with nice instructions [16], which were written with Mosaic data in mind, but which are helpful for other kinds of data.

The format of the coordinate files that are input into the Hydra programs are Fortran fixed format. You must follow the formats exactly (in terms of column number) and *do not* go beyond the columns noted. Don't be creative here. Some common gotcha's during assignment:

1. Do not go beyond col=53. The space beyond this column will be used by the assignment programs to store the STATUS information. Any information written beyond col=53 will be misinterpreted by programs downstream. See the hydraassign page listed above for the *exact* format to use. Please also note that hydraassign is a Fortran program and has a maximum number of lines of data input. Presently it is 10000, which means that only 10,000 stars can be input into the program. If you exceed this, the remaining list will not be output to the *.hydra file.
    
2. Do not use the same star number twice. This is very important. The GUI assigns the colored dots as a function of object number. If you have FOPS and object with the same number, the FOPS GUI dot will not appear.
   Do not have two entries in the table with exactly the same position. I have evidence that if you, say, have two sky positions with different object numbers but the same RA and dec, this will confuse the hydra program but not the hydrasim program. The hydra program will tell you that it can't perform the transformation, but will give you no other information as to what the heck is the matter. Believe me. Do not assign different numbers to the same objects. (Why would you do this? Maybe you want the FOPS stars to also be objects, so you just copy some of the objects to FOPS).
    
3. Do not use the number 0 or negative integers.
    
4. For assignments using "hydrasim", enter the RA of the object as the "SIDEREAL TIME" parameter unless you really know what ST you will be observing at. Leaving ST=0 will cause strange things if the RA is 12hrs in the simulator.
    
5. There are 4 parts to your coordinate file (which I call a *.coo file): the field center (which may be an object of interest or a FOPS), your objects, the FOPS stars, and skies. These are denoted as "C", "F", "O", and "S".
    
6. The FOPS stars should be between 10 and 15th magnitude. However, there should not be much more than 1 mag difference between the assigned FOPS otherwise you will not see the faint FOPS along with the bright FOPS. Also note that ideally you want to have fairly bright FOPS in preference to faint ones. In case of bad conditions (seeing, thin cloud, etc.) faint FOPS may not work. Play it safe and choose 10-13 mag. FOPS if they are available.
   There is a problem with the FOPS display. The FOPS display should show 12 boxes for the 12 FOPS signals. For some reason, our tv clips off the right hand boxes (FOPS 3,7,11. FOPS 0 is the S one, and the numbers go ccw). While you can guide with these FOPS, you can't really see them. I would strongly recommend you use the other FOPS which we can see.
    
7. In the hydra and "hydraassign" program, there is a subtle difference between "move" and "assign". Move moves the active fiber to EXACTLY the xy position of the cursor, while "assign" moves the fiber to the object position you click on. If you are in doubt of what the fiber is pointing to, just look at the "Field Display Tool" (this pops up when you select SETUP on the toolbar) which tells you the name of the fiber and the object it is pointing to.
    
8. Similarly, there can be confusion in the "drag and drop" method of assigning. To active a fiber, just left-click on it. Now keep the left button down and drag the cursor to the object. While you are dragging, the cursor changes to a "circle with a dot in it." If you release the left button on the graphic symbol of the object, the fiber is assigned to the object position. However, if you drop it far enough away from the object, the fiber is assigned to a "RANDOM SKY" position. So make sure your fiber is assigned to an object and not sky. Again, by clicking on the fiber you can see the assignment in the "Field Display Tool". Or if you look real closely at the GUI under zoom, you can see the fiber is not quite on top of the object.
   You can also assign fibers by:
   1. left click on the fiber
   2. right click to get menu. Chose "Assign (move to fiber)"
   3. move the cursor to the object.
   4. left click over the object.
       
9.  GSC coords are not good enough for Hydra. They may be good enough for space data, but for the more wide-field applications like Hydra, the errors in rotation and distortion are too large for accurate astrometric positions. You should tie your system to the USNO-2.0 catalog (or the latest version). This catalog by Dave Monet is tied to the TYCHO and USNO ACT astrometric systems. Your astrometric grid should be based on this, or an equivalent (FK5 for instance) system. A convenient access to the USNO-2.0 that I use is VizieR [17].
   The USNO-2.0 is pretty complete to O=21 and E=20, so for many stellar projects, all your stars are in the USNO-2.0.
    
10. The fibers are 2.0 arcseconds, so you need coords good to 0.3 arcseconds. The measured rms error of the fiber positions is about 0.3 arcseconds.
     
11. Here is what I do to quickly assign fibers if I am not using the "hydraassign" program:
    1. Make a *.coo file with field center, fops, objects, and skies.
    2. Load the file into hydra using the "Setup" button on the toolbar and select the "Setup Field" tool
    3. Enter your *.coo file.
    4. Once it is loaded, enter a reasonable value for "How long to you plan to observe?" and "Anticipated sidereal time at midpoint" in the "Field configuration tool". This is important.
    5. Drag and drop the objects. This has been described as the world's slowest video game. Note that all fibers that are not active are marked by a black dot in the GUI. The color of the dot can only be seen at higher zoom.
    6. Assign at least 3 FOPS and ~8 skies. They should be spread evenly around the field you are actively observing. The more sky fibers you have, the better the sky subtraction. Knut recommends 20.
    7. Save the field. This gets written to the fields directory and whatever other directory you want.
    8. If you want to clear the screen of the fiber assignments, use: File: Run Script: configparkall. See the following web page for the description of the script commands [18].
        
12. The guts of the assignment is controlled by the "concentricities" file. This is defined in the ".fiborg" source file as $ECCENFILE in the $FIBINITS directory. This file tells Hydra what fibers are active, which are locked, which are FOPS, etc. From time to time we change this file if fibers become bad or get fixed. Presently we are using the file called "concentricitieslarge" which activates only the large fibers. Go the "Overview" page if you want the latest concentricities file.
     
13. Note that on the engineering night prior to each run, we update the rotation of the Hydra field. Occasionally, this means that some of your fields may not be able to configure all of the fibers. Also, bear in mind that if you are observing at a sidereal time very different from the one at which the assignment was made, the differences in atmospheric refraction may prevent you from configuring all of the fibers.
     
14. Summary of preparation.
    1. Download the "hydrasim" program and become familiar with using the Hydra interface. Make sure you are using the correct concentricities file.
    2. Measure astrometric coordinates for your objects and place them in the appropriate format for input into the Hydra program. Verify that "hydrasim" can read this object file.
    3. Make some assignment files.

You are now at the 4m observing room. Although the first two computers listed below are the ones that actually control the instrument, they should be hidden from you. We access them through virtual "VNC" windows (see "Starting VNC"), which we can run from any computer, although you are likely to run them from either ctiozm, ctioa7, or both. The benefit of doing things this way is that it allows us to see what you are doing from La Serena, and could even be used to allow collaborators to check up on you from their home institutions--please talk to us first if you would like to do this, however.

• ctioa6: A Linux PC running the Hydra software and VNC server. Your assignment files will need to be copied to this computer. Log in through ssh as "hydra", and put your assignment files in a subdirectory of "/home/hydra/fields".
• ctioa1: Sparc10/Hypersparc, SunOS 4.1.4, 128 Mb RAM, CTIO 4.0m A Sun OS workstation running the ARCON CCD controller VNC server. Your images will be stored on this computer. You can access the disks by automounting /ua1?.
• ctiozm: Dual 2.8GHz Intel, Linux, 2 Gb RAM, CTIO 4.0m. We use this computer to run the Hydra VNC windows and is your best choice for on-the-fly data reduction. Login with the "hydra" username, and try not change any of the setup files such as .cshrc, .login, README, etc.
• ctioa7: Ultra2/Dual 200Mhz, Solaris 5.5.1, 512 Mb RAM, CTIO 4.0m A Sun Solaris workstation for data reduction. Can also be used to run any of Hydra's VNC windows.

Generally one person will be the observer and operate the VNC windows. The other person will be doing assignments, bookkeeping, and totally useless web surfing (when the observer is not looking).

All the computers are auto-mounted and can see common disks. Thus, if the ctiozm or ctioa7 user has made an assignment file on the ctioa7 disk, this can be copied directly to a local disk on ctioa6. The only weirdness is that the ctioa6 disks cannot be seen by other users via automount. They can be accessed by scp, though. ctioa6 can see the disks of ctioa1, ctiozm, and ctioa7.

Logging into the computers:

The computers should be set up for you. If not, here is what you do. Log into ctiozm, username "hydra". Then type "startx" to bring up the window manager.

==> After clicking on the Hydra GUI button located along the bottom panel on ctiozm, which brings up the VNC window, type "hydractio" in one of the black NTERM windows. There is no need to source it into background (with the "&"). The Hydra GUI will come up. It will ask you if you want to reload the previous coords. Unless you left the Hydra program with the fibers at a configuration position and want to continue to use this setup, answer this "NO." You will now be told to type "coldstart" in the white Hydra control window. Hydra is now ready for use.

TO LEAVE the program:

Under file, use QUIT.

1. Do you really want to exit the application? YES
2. Do you want to move the fibers to the transport position? NO (note, though, if you have used a different concentricity file, you really need to stow the fibers first -in this case, answer YES.)
3. Do you want to turn off the drives? At this time, TURN OFF the drives. At present, we have found that the motors can burn out if left on, so the drives must be turned off.

==> Bring up the Arcon VNC window on ctiozm:

If it's not already running, right-click to bring up the menu and click on "(Re)Start Arcon". At this point it takes about 60sec for a dizzyingly large number of designer colored windows with mini-fonts to open. You will be asked a single question half-way through the process "Do you want to synchronize the parameters?" In the 6 years I have been using ARCON, I have always said "yes." I have no idea what happens if you say "no," but it probably involves pain and shouting. I think the question is there to see if you are awake.

I HATE the default fonts for the IRAF window. I edit the following line into IRAF section of the .openwin-menu:

/local/combin/xgterm

==> You will probably want to run the Hydra simulator on ctiozm or ctioa7. If the latest version is not already installed, the user should immediately install the "hydrasim" software using the instructions at the hydra software page [19]

Load your coord and assignment files in some subdirectory in ctiozm or ctioa7 - I use "fields." Try entering one of your files into the simulator software to make sure everything is working.

THE CCD:

ccdinfo:
cl> ccdinfo

Spurious Charge is signal per pixel generated by clocking
the CCD. Read Noise is quoted for overscan pixels. In the image:

Noise floor = sqrt{ Read_Noise^2 + xbin * ybin *
              (Spurious_Charge + Dark_Current * Exposure_Time) }

Dark current = 0.7 e-/hr (per unbinned pixel) in Jun, 2001,
               at CCD_TEMP= 150K (pumped N2, heater off)

For gain 2, bin 2x2, total noise in 1800 sec dark = 3.5 e-

Read times are quoted for binning (spatial x spectral).
When binning 2x2 the ADC saturates before full well is reached.

NB: A peculiarity of this CCD is that full well is highly
    variable from pixel to pixel causing vertical trails to
    to appear at high signal levels. The full well figures are
    very preliminary estimates for the worst pixels.

*** Select gain setting from the first column

*** The current gain setting is 1

During the night, at every slew position and every configuration: (these are typical numbers) echelle: etalon, 180s; quartz, 120s rc mode: penray 10s, quartz 10s for KPGL3

Note that "penray" is actually a rack of a number of lamps, including neon, helium, argon, and xenon. It is supposed to emulate a more traditional (but much dimmer) he-ne-ar lamp. The line ratios are very different from those in the old lamps. The henear lamp is available but is dim and not very dense in lines.

During the day: (do all these with the fibers at the large circle position:
select "File->Run Script->largecircle (or configlargecircle)->Open" from the
menu bar in the Hydra GUI, or type "execfile largecircle" in the command window)

25 biases
echelle:
3 daytime skies @ 45s + etalon (with dome closed!)
3 th-ar, 600s
3 penray,henear, 300s
darks (I ususally leave darks running at the end of the night when I
go to bed)

rc mode:
10 dome flats (make sure that you tell ObsSup that you want the color
balance filters removed to do these. We have some blue color
balance filters on the flat-field lights on the ring of the telescope
for direct CCD work. These must be removed)
3 penray, 1s
3 twilight skies (right after sunset)

During the run once:
milky flats

The ObsSup will be more than happy at the chance to install the milky flat diffusing screen in the spectrograph. To do the milky flats, you have to bring in the fibers to the large cicle. Do this by selecting the "largecircle" script from the File->Run Scripts menu. Do the milk flats.

You may want to take some short exposures of the milks to get data for the mask image. The division of the short milk and the long milk is a good way of revealing the traps and bad pixels.

I STRONGLY suggest you worry about the dark frames. You should verify that the dark count is low. The CCD should have only a few e-/hr/pixel behind a dark slide. The spectrograph is wide open, and although we try as hard as possible to keep stray light out, sometimes doors aren't closed and light gets in. The dark count should be no more than 10e-/pix/hour. The Mosaic CCD chip should have less than 3e- read noise, so any dark more than ~10e-/pixel in your exposure is going to reduce the S/N of a faint object.

USING THE COMPARISON LAMPS:

In the Arcon parameter file "instrpars", set the "lampsys" parameter to "new" and select the comparison lamp you want to use in "complamp", using "pen" for the HeNeArXe penray. When you take a comparison lamp exposure (through "comp" or "observe"), the Arcon will automatically move the system into place and turn on the lamps. You can watch all this happening from the "Lamps" menu in the Hydra GUI. Note that Arcon will turn off the lamp but leave all the mirrors, etc. in place after the exposure is finished. If you go to a field and can't see any stars through the gripper camera, you've likely forgotten to park the comparison lamp system. Do this from the Hydra GUI by clicking the "Park" button.

MAKE SURE THAT THE GRIPPER IS OUT. IF THE SOME OF THE SPECTRA ARE MISSING, IT IS PROBABLY THE GRIPPER IN THE WAY.

FOCUSING:

It is not too difficult to focus the telescope using the FOPS. You can ask the night assistant to focus for you and he (she) will use the following procedure. Remember that the FOPS are fiber bundles of 7 1 arcsec fibers. Choose a single FOPS with a star on it near the field center. The idea is to get most of the light into the central fiber by changing the focus.

The object will move during focus and the FOPS will lose centration. I found to focus, you must do the following:

1. Move the focus about 10,000 units above nominal focus in IN direction.
   Check centration.
2. Move 2500 units in OUT direction. Check centration.
   ...etc.
3. Go through about 10 focus steps and pick one with best intensity and smallest xy movement through image.
4. When the focus is set, note the temperture of the "Serier Truss" by typing "settemp" on the IRAF command line. If you set "setfocus" to "auto" in "instrpars", then Arcon will look at the difference between "temperature" (which is updated every time you type settemp) and "reftemp" and adjusted the focus for change in temperature. Just make sure that you have entered the temperature at which you focussed the telescope in the "reftemp" parameter and the correct focus in the "basefocus" parameter. If you'd rather enter the focus value yourself for every exposure, set "setfocus" to "yes". Setting it to "no" will cause Arcon just to use the focus value in the "telfocus" parameter.
5. To modify the focus as the temperture changes by hand, note that the focus changes as df/dT = -766.2mu/deg C.

Typical values:
12.8 degrees 156200

TELLURIC OR SPECTROPHOTOMETRIC FLUX STANDARD:

If you are using echelle more or observing longward of 6000A and need high S/N spectra, you probably need a telluric standard. It is best to do it through a few fibers. You can do it one fiber and exposure at a time. Alternatively you can bring in a number of fibers sequentially. Put the fibers at the large circle or park position. Bring one fiber into the center. Start a long exposure on the CCD (3600s). Expose for say 100s and then "pause". Then:

unwarp
gripin
park 000
move 001 0 0 <== some other fiber here.
warp
gripout

and expose for 100s starting with "resume" on CCD window. Etc. (Check to see if light is leaking in from gripper light.) You can do this on about 4-6 fibers with the telescope in open loop tracking.

Alternatively, you can put a FOPS down on one guide star (and have the object in the center). After moving the a new fiber to the center and warping the plate, recenter the guide star on the FOPS. You can turn on the FOPS guider to guide, but I recommend just guiding by hand on these short exposures.

You can get the *.hydra fibers of the tertiary spectrophotometric standards of Baldwin and Stone (1984, MNRAS, 206, 241; 1983, MNRAS, 204, 347) as calibrated at CTIO by Hamuy et al. (1992, PASP, 104, 533; 1994, PASP, 106, 566). These are 11-14th mag stars. I also have *.hydra fields for the secondary spectrophotometric standards (also called Hayes standards) as give in the Hamuy reference. These have fluxes every 16A. The stars are 4-6th mag.

The coords are taken from USNO-A2.0 and have the standard in the center with FOPS stars and skies scattered around in the field. See the directory:

/uw52/nick/hydra/text/specphot/*.coo

A selected list of bright, hot stars culled from the Yale Bright Star Catalog can be found in:

/uw52/nick/hydra/text/bstar.dat

Basic CCD reductions. The idea here is to process the ALL THE SPECTRAL DATA, including the object, pflat, dflat, and sflat data, to [OTZF] before extracting.

Find the biassec and trimsize. You must trim out as much of the image as you can to make the flat fielding as easy as possible. Process the data through [OTZ].

ccdr:

ccdpr: (last half of n2 and n3)

To process milk flats, first process through [OTZ]. The idea here is to bring the milk flat to 1.0 everywhere with the pixel to pixel variation left in. If a few wiggles are left in the image, that's okay. You will be dividing by the dflats later anyway. To bring the milky flat to 1.0, you need to remove the faint ripples of the fiber images. These run along the x-axis. The trick to remove these is to create a median filtered image using a long skinny boxcar and divide this into the data.

Do something like:

imdel temp*
# combine the milky flats into a single image
comb @in1 temp1
# remove the spectral shape in the x-direction
blkavg temp1 temp2 1 2000
fit1d temp2 temp3 fit ax=1 low=2 high=2 or=20 niter=10
blkrep temp3 temp4 1 1004
imar temp4 / 2000 temp4
imar temp1 / temp4 temp5
# remove the spectral shape in the y-direction
blkavg temp5 temp6 4000 1
fit1d temp6 temp7 fit ax=2 low=2 high=2 or=3 niter=5
blkrep temp7 temp8 3021 1
imar temp8 / 2000 temp8
imar temp5 / temp8 temp9
# filter the data with a long skinny boxcar
fmed temp9 temp10 xwin=51 ywin=1
imar temp9 / temp10 flat
(you may want to process with fmed in y direction also)

Now process through [OTZF]. MAKE SURE THAT YOU PROCESS THE PFLATS, SFLATS, DFLATS, AND COMPARISONS. YOU MUST CHANGE THE "IMAGETYP" TO "OBJECT" FOR THE PFLATS AND THE SFLATS TO GET THEM TO FLATTEN.

WAVELENGTH CALIBRATIONS

We have four types of lamps:

For low dispersion (RC) work:

Helium-Neon-Argon (a single lamp with all 3 gases)
"Penray" which is actually 4 lamps of helium, argon, neon, and xenon.

The individual lamps in the penray "lamp" can be turned on or off on the hydra instrument (but not from the Hydra GUI). All lamps should be on. We have been having problems with the neon lamp burning out or being weak.

For high dispersion (echelle) work:

Thorium-Argon
etalon

The wavelengths are kept in the IRAF directory linelists$. Some additional notes if you want to go to the original sources:

For the Xenon lines, use the list in Striganov & Sventitski, "Tables of Spectral Lines of Neutral and Ionized Atoms", 1968, (IFI/Plenum: New York). This is a good source for all wavelenghts in the visible region and is accurate enought for all the RC setups. It is a nice table because it gives intensities also.

The best source for argon lines is: Norlen, 1973, Phys. Scrip. 8, 249.

The thorium lamp has rather weak thorium with respect to argon, and the lamp seems to have a quite different ionization than typical atlases for th-ar, such as the ESO atlas. Very accurate wavelengths are given in: Palmer, B.A., and Engleman, R. Jr. Los Alamos Sci Lab Pubs, LA-9615. In the blu, the ThAr lamp is basically all argon, with only the very brightest Th lines appearing weakly in the spectra.

In the blue, the He lines dominate in the penray lamps.

As far as I can tell, there is no neon in the he-ne-ar lamp.

SPECTRAL EXTRACTIONS

See Knut's reduction notes [20].

TAPING:

We have DDS3 and exabyte. We also have DLT 7000, but you will not be generating that much data to need a DLT. Most people write data to tape using the IRAF "wfits" command. This is not a particularly efficient way to write data, but it is tried and true. I would NOT recommend writing a large tar file. A tape error in the middle of a tar read can ruin your whole day because it is hard to force the tar past the bad data point (if this happens, turn off the parity checking in the tar command). If you do write tar, I recommend breaking the data up into tar files = say one per night of no more than 1-2 Gbyte.

Write two tapes just in case. Give one to your collaborator or sister or whoever, just in case. Data can be (painfully) recovered from the SAVE THE BITS backup tapes, so if you can't even read your sister's tape, you can ask us to recover it from the backups.

DON'T PANIC

These are common problems with simple solutions.

LOST FIBERS

You get an error while configuring where an annoying little grey box pops up and says: "Error:configure: OrderMoves Error: phys move. MoveButt couldn't pick up button xxx" where xxx is some fiber name.

What has happened is that the gripper has gone to the position of xxx and could not pick up the button. Much of the time this is because the button is slightly off from where it should be, possibly because it was placed down on a small piece of dust. Sometimes too much stress builds up in the hypodermic needle which houses the fiber and the fiber simply slides out of the gripper.

However a fiber might get lost, to recover use the gripper TV to place the gripper on top of the fiber. Then, in the command window, type "thisis". This tells the software that the gripper is on top of whichever fiber is currently lost. Note the use of small letters in this command. (The related command "ThisIs xxx" tells the software explicitly that fiber xxx is under the gripper.)

If the fiber is not visible in the gripper TV, you can use the GUI to cruise near where it should be and see if you can find it. If you find it, tell the gripper where it is using the "thisis" command. Don't go very far cruising or you may find another button and get Hydra really mixed up by identifying it incorrectly.

If you can't find the lost fiber near the nominal position, return the gripper to the "parked" position of the lost fiber. There is a good chance that the fiber will be there and was never picked up in the first place.

If the fiber is not in the parked position, you will see the hypodermic tubing enclosing the fiber crossing the field going to wherever the button actually is. Use the gripper motion controls to follow the tube out into the field until you find the button, put the gripper on top of it and then use the "ThisIs" command to tell Hydra where it is.

If the button appears tilted there is sometimes a tiny piece of ferrous material stuck to the bottom. There may be damage to the button. In either case if this happens, call Observer Support and have them inspect and retire the fiber manually.

Once the fiber is returned to the park position, you can use it again. If it keeps on getting lost, you should lock it and not use it for the rest of the run.

If the fiber is really totally lost, or this procedure doesn't work, you will have to have Observer Support come and retire the fiber manually.

CAN'T CONFIGURE THE FIBERS WITH THE HYDRA PROGRAM

You were able to configure the program with hydrasim, but you get an error about a transformation problem when doing the real configuration. This error happens just after you start the configuration.

What is happening here is that the hydra program is doing one extra step beyond the hydrasim program. It is taking the coords and assignments, and appplying a transformation to the appropriate zenith distance for the observations. For some reason, the transformation is producing a collision.

Possible causes:

1. you have some objects with the same names or the same postions in your file. Fix the file.
    
2. two fibers are really very close so that the atmospheric dispersion correction is causing them to collide. This happens most often if you have successfully configured a field, observed for a few hours, and you reconfigure. The new atmospheric dispersion correction is different enough that the fibers collide. Look for fibers which are very close and retire one.
    
3. one fiber is really quite bent, and the dispersion correction is trying to bend it beyond its software limits. Look for a really bent fiber on the GUI.