Detailed startup and usage instructions
Courtesy of Rubab Khan and Rebecca Stoll
At the beginning of your run
OSMOS will typically be ready for operation when observers arrive for their
first night. The MDM staff will perform the following tasks
before a run, but they will be necessary for observers to follow in the event
of a power failure or lightning shutdown:
The next step is to launch various control clients and the User Interface on
the observing workstation. From the "Applications"..."Data Acquisition" menu
on the upper left-hand corner of the desktop, select the following options in
the order they are listed:
- Turn on the IC (Computer Room) and type 'O' for OSMOS
- Turn on the Instrument Electronics Box
- Turn on the Head Electronics Box
- MDM TCS Agent
- OSMOS IE Agent
Then type startup in Prospero to load the current instrument
The last step is to setup the mechanisms, including resetting the
the collimator and camera lens barrels to their nominal positions.
pr> call ossetup
This will initialize the filter, slit, and disperser wheels (verify
they are in valid positions), and reset the camera and collimator lenses
and set them to the nominal 15°C values. See the
section on the Collimator and Camera Focus
in the OSMOS Manual for details of the seasonal
settings for these optics.
Finally, make sure that the OSMOS instrument hatch is open.
OSMOS has its own instrument hatch that is separate from the MIS hatch.
The lever to open and close this hatch is on the opposite side of the
instrument from the access doors for the slit and filter wheels.
Remember to set the Collimator and Camera Focus
before you begin observing on the first night of your run. If the power is
ever cycled, both will need to be reset. On subsequent nights, change the camera
position if the temperature significantly changes.
Starting up John Thorstensen's Acquisition Scripts
Do not open ANY other ds9 session on agung. Check the machine itself to ensure
no stale sessions or the RETROCam control system are running.
- Open a terminal, ssh to agung
- $ cd /lhome/obs24m/Scripts
- $ pyraf
- $ pyexecute('osctrtask.py') (Centering task, certainly needed)
- $ pyexecute('osqlsp.py') (For quick look spectral reduction)
- $ pyexecute('osbias.py') (For focusing)
- $ !ds9 &
Starting up John Thorstensen's JSkyCalc24mGS for Guidestar Selection
- On a different workspace use the "Applications"..."Telescope Control" menu to
start the JSkyCalc24mGS program
- load your object list.
Starting up the Guider Camera
- Turn on the new guider PC called 'mdmikon1'. It is labeled, and is the
small Dell computer is on your immediate left as you enter the computer room.
Log-in info is on the white board.
- Start Maxim DL, find the camera control button and click on it.
- Click on the Setup tab. Click connect. Turn on cooler.
- Go to Expose tab, click start.
- The guider camera is now in acquisition mode.
Choosing a Guide Star
JSkyCalc24mGS, bring up your object either by selecting it in your
catalog or by typing in the coordinates.
- In the guide star window, if you are setting up to acquire on the red
(inner) slit, click the radio button marked 'red slit' at the bottom
(wait for the stars to offset). If not, make sure the field center radio
button is selected.
- Choose a likely-looking star, click it, and wait until the guider
coordinates update. Then either enter them in the xmis by hand or click
'Move Guide Probe.'
Guiding with the Guider Camera
- If you are in acquisition mode (exposing on the expose tab, image
updating) then click stop and go to the guide tab.
- In the guide tab, click on expose and click start. This will take a
single image. The program will choose the brightest star in this image to
guide on. This image also defines the position that the guider will work to
keep this star locked to.
- (This intermediate step should be done perhaps once per night, or perhaps
when pointing at wildly different portions of the sky) Click calibrate and
click start. Wait for it to finish (about 30 seconds) before beginning to
- To guide, click the track radio button and click start. The full field
image will shrink to a region around the selected star, and it will update
every few seconds.
- If you are taking long-slit spectra, choosing a guide star on the left
side of the display in JSkyCalc (high y rather than high x) will vignette an
unused part of the field, a part of the chip that is not even read out with
the 4x1k ROI.
Long Slit Acquisition
We strongly recommend that you use John Thorstensen's osctrtask in
pyraf. Although it is possible to line up the object in the slit by hand
using telescope offsets such as 'south 15' (arcseconds) in Prospero, the
telescope's small offsets are not very precise. The osctrtask
calculates a guider stage offset instead, which is much more precise, and then
you move the telescope by hand in order to realign the guide star with its
original position and resume guiding without re-exposing in the guide tab
- The steps are:
- While guiding, take an acquisition image of the field and a slit image. An
easy way to do this is with one of the following Prospero scripts.
Warning: please check that the scripts have correctly identified the
slits and filters you plan to use as the filters and slits may change
Choose the exposure time NN so that you can identify your object. Exposure
times of ~30 seconds will often yield enough objects for the
osctrtask to find a WCS solution and identify your object in the
- $ call fieldctr NN (for center slit, uses 1k ROI)
- $ call fieldctr2 NN (for red or inner slit, uses 4x1k ROI)
- from the pyraf window (which you have opened on agung),
- $ epar osctrtask
- modify the file numbers for the field image and the slit image, and make
sure the prefix is correct and ends in '.'
- If you wish to use the WCS capability, either enter the object coordinates
or your catalog file and the name of your object.
- Click execute
- If an iraf window with a WCS solution pops up, hit 'q'. Now look at ds9.
- If a WCS solution was found:
- verify that the circle on the field image is centered on your
- If not, your coordinates are not accurate or precise enough. Run
osctrtask again without WCS. If so, continue and use the resulting dy.
- hit 'x' on the center of the slit, then hit 'q'.
- If no WCS solution was found, hit 'x' on the center of the slit, use tab or
the ds9 button to switch frames, hit 'a' on the center of your object, and
- Write down the dy shift. Check to make sure the dx shift is less than
about 700. If so, then you may ignore it.
- To put the object in the slit, first draw a box around the guide star's
current position (click and drag a rectangle). Now you have two options:
- Option 1.
- Stop guiding (on the Maxim guide tab click stop).
- Go to the expose tab and start exposing. Make sure the rectangle is
around the guide star's current position.
- Enter the calculated dy in the xmis guider dy box, then move the
telescope by hand with the paddle until the guide star is back in the box.
- Stop exposing, go back to the guide tab, make sure track is still
selected, and start tracking again. Do NOT take an initial exposure in
the guide tab. This would set a new expected location of the guide star,
which you do not want.
- Option 2.
- Do not stop tracking. Enter the calculated dy offset in small
increments (~50 at a time) and wait for the guide star to be pulled to
the center again before entering the next increment.
- When you have entered the complete dy, are once again guiding,
and the guide star has settled into the center, you are ready to take your
- It is useful to take another image through the slit after completion of
these steps to confirm that the object is indeed in the slit. For bright
objects this will be obvious, while for faint objects it may be necessary to
run a second iteration. In either instance, an image of the object in the
slit may be useful for future reference.
Taking a Spectrum
An easy way to do the above is to write a Prospero script to do everything,
taking as arguments the object name and the exposure time. John
Thorstensen has written one called 'gotospec' for setting up to take
spectra with the VPH and the center 1.2" slit:
- Make sure that:
- The correct slit is in
- The correct disperser is in
- Filter wheel 1 is in an empty position
- Filter wheel 2 is in an empty position
- Your object name has been updated
- Your exposure time has been updated.
- The region of interest is correct (4x1k for VPH spectra, 2x1k for prism)
If you use one of the above scripts, your setup is now done. Type 'go'
to begin the exposure.
- $ call gotospec [objectname] [exptime] (for VPH and center 1.2" slit)
- $ call gotospec2 [objectname] [expt] (for VPH and inner (red) 1.2" slit)
- $ call gotospec3 [objectname] [expt] (for prism and center 1.2" slit)
Return to: [
Paul's Home Page |
MDM Observatory |
OSU Astronomy Home Page
Updated: 2012 August 21 [pm]