Spectrograph Notes


SARA NORTH — Spectrograph Notes: 03/12/2014 (by Gary Henson)

Camera control is under Maxim DL. The spectrograph is Camera 1 (FLI) [Note Camera 2 is the guide camera] THUS, NO HEADER INFO IS PASSED FROM ACE TO THE CCD IMAGE, You’ll have to edit this. 

Temp is set at -50C, leave it there, cooling system should be stable since everything is in the control/warm room.

Camera should be set for 1×1 binning, I have set the image to “no rotation” so the image should diplay horizontally across to look like a typical echelle spectra. Adjust magnification to your preference.

Images are also saved under Maxim. ACE does not control this camera. It’s easiest to save using the “Autosave” Menu. I think most everyone probably has a copy of Maxim DL5 that you may use locally. If not, the menu options are fairly user friendly. Do note that the image save path, calibration options, special settings,etc. are available using the “Options” tab under both the “expose” menu and under the “autosave” menu. You can set bias, dark, flat, light exposures under autosave as well.

On the ACE side

Current Focus Settings under the Echelle Spectrograph GUI within ACE:

For the “fiber plate” try -100 (that is a negative)

For “Spectrum” i.e. the FLI ccd, try 650

Both of the above are the current settings. If you want to fine tune them, steps of 50 (maybe 25) should be sufficent.

NOTE: Since the star is focused on the fiber plate as seen on the video feed, it is important to First focus the fiber plate image for the video. This is most easily done using a very bright twilight sky, but can also be done with the Thor/Ar lamp if it’s light is put slightly off center.

The Fiber Plate image is the 3rd video screen (lower left) on the Dome video camera display.

With air/telescope temperature stabilized at 46F in the early morning, I found the following focus settings:

Main Telescope focus is ~ 203,500 {this is versus ~194,000 for the ARC ccd}.

Autoguider focus is ~18,500 (versus ~14,000 for the ARC ccd)

Note all autoguider fields should still be available, I just used the center field with no problems.

There is no tool to measure focus on the Video display, you just have to go by eye to set the smallest diameter image on the fiber plate.

ARC VERSUS FIBER POSITION OFFSET: There is only about a 1 arcminute offset between the fiber and ARC ccd center. Bright point objects should be easily within the fiber plate video view when pointing the telescope. I did establish the approximate fiber center location on the ARC ccd, however.

With the ARC ccd frame oriented to show N up and E to the left, a star at X = 453 , Y = 510 would then be at the Fiber center with the spectrograph module in the beam.

For bright objects, you can simply adjust telescope position with the software handpaddle to put the star in the center of the fiber. The seeing disk for very bright stars will spread out beyond the fiber edge, but you will then see an obvious “donut” when the disk center is over the fiber. Fainter stars should disappear completely when centered on the fiber (an ~ 2.5 arcsec diameter).

The Autoguider works fine in parallel with the spectrograph camera control, you’ll just need to keep all the windows straight (size, contrast settings, location on the screen, etc.)


Within ACE instrument control is a Tab for the Stella Cam. This should allow for “gain” and “exposure” adjustments to view very faint objects, including the fiber position with a red LED illuminating it. One should be able to alter these settings to give a better view of the star image on the plate.

{The functions weren’t responding last night, but are the standard control options available with a Stella Cam hand box. Once Peter gets that working, I can comment on recommended settings here}

Under the Echelle Spectrograph Tab you will see options for turning on the Flat lamp and Thor/Ar lamp on the right side. You will see a very bright spot show up on the fiber plate video. Exposure times of only 0.1 sec, or at most a few tenths of a second, are all that is needed to yield a full flux spectrum.

Since the ccd/spectrograph is table mounted in a constant thermal environment and these lamps are bright, one can go ahead with all calibration images during the day and save the night for actual spectra.

I took several 10 minute dark frames and did not notice a lot of thermal noise with the ccd cooling to near -50C now.

FINALLY: The only challenge so far is marking the fiber center reliably on the video diplay. I used a transparency over my laptop monitor and drew a small circle with a marker to use in guiding the star image to fall in the fiber. Peter does have a red LED that should illuminate the plate while you also see the star on the plate. You should be able to see the black spot of the fiber end as you move the star towards it then. Alternatively, Peter is looking into a software aperture that could be permanently added to the video display to mark the fiber position. The LED should be ready as soon as Peter checks it out again, the software aperture will come later.

Peter Mack replies:

Hi Gary,

Thanks for the report / notes.

The Stella Cam was not working because it did not have power on the camera
(only power on the controller).
I think I know why (I had accidentally removed a pair of conductors from a
connector and when I put them back I made a mistake counting from the wrong
end of the row).
So you were using the Stella Cam in non-intensified mode last night.  The
LED’s were working but could not be seen without intensification.

The cooling sits steady at almost -50C.  We should get the next sized cooler
for the CTIO setup as this one is running full bore.

– Peter.