spectro-aras.com

Hello all,

I have now a permanent setup for spectroscopy: C11 + L200 (1800l/mm) + Trius SX814
I made several spectra, but i noticed some inaccuracies and non-represenative observations.

I observe now some Be-stars like 28Tau, VVcep, zetTau, gamCas... with the L200 & 1800 l/mm grating.
I see some scattering in the results over several days comparing other observations .

If i check my spectra carefully, i notice some problems.

What is do now is:
I take a calibration spectra with the internal calibrationlamp of the L200, then i take some spectra of the particular star, after that, i take again a calibration spectrum with the same lamp. Time between the 2 calibrations is about 30 ... 45 minutes.

If i compare the spectra of the calibrationlamp: start and end of the session, i see a shift in the spectra of about 1 ... 3px. For high-resolution (R=10000) is this huge !. So i think the scattering in the results is the cause of shifting of the spectr over time, possible due drop of the ambient temperature and/or flexure.

I did now also a check by taking lamp spectra at different points in the sky and i see also a shift of the emission lines of the lamp. It looks that it is more a flexure problem inside the L200.

In addition i send you 2 graphs: 1 is comparing of the spectra (6x300s) and 2 different calibrations: at begin and the end. Other graph is 1 spectrum as first image with a calibration at start and 1 spectra with last image and calibration spectrum just after session. You see here that this is a good spectrum, both are more or less the same, only the SNR is not very well.
Now the question is:
- how can I best deal with this problem?
- can I solve this via software, ISIS ? and if so, where can i find a tutorial?
- other proposal how i can deal with this ? other procedure of taking spectra ?

Thanking in advance for your time and sharing of experience

best regards,

Hi Erik,

The solution for this common problem with Littrow type spectrographs is the "wavelength registration" flag in ISIS:
http://www.astrosurf.com/buil/isis/guid ... to1_en.htm

So you can afford only 1 calibration at the beginning.
Be aware it works only in given circumstances, within a limited accuracy. With many sharp features on the continuum, it used to work well (till SNR is good enough).
Alternatively, you can detect the shifts yourself, and do the shifts manually before the sum (manual way).

However, for shortterm variations (close EBs) or high accuracy (exoplanets), this technique fails - then need calibrations much more often, and the processing gets more complex (no ISIS feature for that - all to be done manually).

- Peter