Tutorial OHP 2009

OHP 2009 workshp - observation with Astrphysic 120ED refractor (5"), GP-DX mount, Lhires3 1200l/mm, Audine KAF 400 and KAF 1200, guiding camera Watec 120N- see the instrumentation and the results here

The first installation

The adjustments have to be done if you dismount/mount one of the cameras:

Finally, it is strongly recommended to take flat images at the beginning of the observation session. If the camera is having trouble, this can be detected and corrected before spending hours for nothing if you do not trust your camera behavior (freezing, artefacts)

When camera does not collaborate...

The Audine camera is a wonderful camera but depending onhow it was assembled can be sensitive to freezing. This is not easly noticeable. Some tricks: "effect of lens" around the neon lines or on a spectrum of ta bright star. Or, the flat image, spread dust on the side, growing with time.


Capella spectrum - abnormal light enforcement

Some cameras can also create "moire effect" with some spectrograph. It is still unknown the conditions of such artefact, and the reality is this is really annoying as it is very difficult to remove, even with Falt images.  As I had the possibility to change again camera, I decide to not investigate and simply try a new one. (As the designer of the Audine, Christian had some...)


Flat with non-homogenous lines
Zoom 1:1 on the interferences

Thus, after two nights of camera mount-dismount, it is finally an Audine KAF400 without shutter which was the best. The camera did not freeze and its flat was repeatable over nights. A tip here: it is recommended to take a neon image after the flats, in case some shift occured in wavelength. The camera not having shutter, the falt lamp was shut down after 10s of exposures, before the exposure end and the sensor reading. The exposure is adjusted to fill 2/3 of the dynamic. Special attention shall be paid to not have any saturated areas.


Flat night 3

Flat night 6

Night protocole

if the spectrograp is kept attached to the telescope between the two observations session, with the cameras, only some checks and adjustements has to be performed when starting:

Softwate"Suite "

Audine USB interface was run with Pisco, AudeLA was used for auto-guiding. On top of those you need to add a star map software to check the field in the finder. The only 2 times I did not recognize the field fully, I picked the wrong object...I used "Carte du ciel", freeware and sometimes Megastar.

Autoguiding

Autoguiding in spectrography is not only a comfort. If correctly adjusted it is a must to capture the maximum of flux from an abject. In only rares conditions the manual guiding will better perform that the automatic one. (wind, bad equilibration). In this campaign, all was acquired with autoguiding. From time to time, you still to check that the star has not gone away or check for the possibly oscillation and overshots at specific telescope position.

The audeLA autoguiding feature is easy to set-up. At the beginning, the 2 USB links shall first be recognized: the grabber for the video camera and the LX200 link to the mount. I did not run the learning function, but directly try-and-adjust the value of the RA and DEC speed. As they can be changed during the guiding, it was very easy. just click on the "configuration" button. To guide on the slit, click to place the red crux on the exact slit position. Move back and forth the star across the slit and detect which position the flux is minimum. The slit was vertical in this set-up. The slit guiding feature was not used here, only the "stay-here!" guiding. Keep the vertical position of the guiding the most repeatable you can, which means the spectrum will always be around the same Y-axis position. If you checked on the flat that there is no dust, better to stay here and if you acquire in window mode, better to make sure the spectrum is in...

Altair, on the right side of the slit

Star mag 4.5 in the slit

Star mag 4.5 left side of the slit

Star mag 4.5 right side of the slit

The weather was really great, with almost no wind, one night the temprature was still at 20°C at 4:00am. Only one night some gusts wind was quite strong and make the guiding much more difficult. On the images below the effect can be seen on the guiding star. The flux throuput was largely degraded, so the signal to noise ratio and the lost of star could happen. 


Wind effect

The two set-up: astrophysic and the C8 of Christian Buil.

Iris processings

All the pre-processing was made with Iris. the offset, dark and flat are first prepared, with the cosmic file. The spectra processing are then made through the menu Spectro>Traitement des spectres LHIRESIII 1200.

But some values have to be determined before. Here is the sequence:


neon spectrum

neon profile in blue, H-alpha position in green and Altair spectrum in purple.

Neon shit

Along the night, a shift in X (wavelength) can occurs in regard to the CCD sensor. It has be checked as well that this shift is not visible at the beginning and the end of the sequence with no telescope position change if you decide to only take one neon. In the example below, the shift is few pixels. But even if you enter a wrong value, as the software detects the line around the entered position, this error of few pixels does not lead to a bad calibration.


Example of a shift of the neon spectrum between the beginning and the end of a night

Visual Spec processings

Once the calibrated profile is created with Iris, check it with Visual Spec. To check the wavelength calibration, use the function of atmospheric correction to generate the synthetic spectrum of the erath atmosphere. Check the absorption lines of the synthetic spectrum match the star atmopsheric lines. Some shift can be seen. If the shift is constant over the profile, you can correct it by entering a shifting value in the dialog box. 

We then have to create the instrumental response curve. It is done using the Altair spectrum. The tip here consists in eliminating the atmospheric lines of Altair spectrum before you extract the response. 


Atmospheric lines correction by iterative adjsutment of the sigma and intensity paramters (depth and width of the lines of the syntheticprofile)

Altair profile corrected . the profile Division is moved to replace the intensity serie in the visual spec document

Load the "real" spectrum of Altair, here it comes from the UVES database. (You have it in the sub-directory "Training" of Visual Spec)

Dived the Altair profile by the UVES profile. The result is moved in the intensity serie if the checkbox is checked. 

Result of the division of the Altair profile by its UVES profile

Once the division performed, the next step consists in smoothing and saving it as the instrumental response profile, for the given night. 


Menu Radiométry>Extraction of continuum - as the atmospheric lines have been removed and as no other star lines alters the continuum you cn immeditely run the smoothing function. Move the cursor to increase or decrease the strenght of the smoothing to match the overall profile.

The resulting profiles being in a temporary serie, it has to be moved in the intensity serie.

Finale response curve.  This profile is saved under the name reponse.spc, for example

the instrumental response being computed, each spectrum profiles of the objects shall now be divided by this response. You can do it one by one, or use the function "script-box" of Visual Spec. 


Script assistant in Visual Spec - the files to be processed are loaded in the left list zone. The script is either written in the righ blue zone or loaded if already existing.

Vspec opens each file ;dat and generates a new file file for each. These new files conatins the profile divided by the response and save with the ending "_1c"

To work later only on the spc files, you can filter the extension in the open profile dialogbox.

The last step is the formatting to load them in the BeSS database. The file format shall be fits file, 1D profile. The fits header shall contain the keywords and the proper values to meet the BeSS standard. In Visual Spec, this is done throught the command of the menu file>export to fits (BeSS) shortcut Ctrl+W. Enter the dialog box required fields


BeSS fits formatting dialogbox in Vspec to fill the fits header

The name of the object shall be compatible with the Simbad base. You can test if the name is really recognized if you have an active internet connexion. The following rules is recommended: 3 first greek letters (Bayer convention) with no upper case, a space, followed by 3 fist letters of the constellation, starting with a upper case: iot Lyr. for "Flamsteed" format: number, space, then the 3 letters of the constellation starting with a upper case: 6 Cep. For HD format, just HDxxxx like HD22780.

The date of the observation and the time of the beginning of the exposure sequence can be entered manually, but the most convenient is to use the "Seq" function. Vspec will open the current directory, filtered on image extension. Just select the first and the last image of the sequence. Visual Spec will extrac the date, the time of the exposure start and the total duration of the sequence as specified in the BeSS specifications. This totale duration is not the addition of the indivudual exposures but really the time between the first photon and the last one in the sequence).


List of the images of the sequence contributing to the final spectrum profile.

Visual spec detects thefirst and the last image of the sequence. by looking at each image header. It extracts the date and time to compute the totale duration of the sequence.

Simultaneously, Vspec updates the corresponding fields in the BeSS dialog box

The specific parameters to BeSS are the isntrument and the site. Text chains are used by BeSS to uniquely indentify them. They have to refer to an existing instrument or site, or a new isntrument or site will be created in BeSS database. Visual Spec can keep a list of your most used isntrument or sites in two text files: list_instru.txt et list_site.txt - you can edit them to add new ones (using notepad or whatever). This eliminates typo errors which would drive to the creation of a false new instrument (or site). 

Once all the fields are filled, click on oK. The fits files is created and you can select the directory where you want to save is. 

Your file is ready to be load in the BeSS database !