Calibration Spectra

I use an older "Filly decorative lamp" (first suggested by Christian Buil on his website <link>) for line spectra.

The line wavelengths are taken from a calibrated high-resolution echelle spectrum published by Thierry Garrell <link>.

A self-written python program makes a gaussian fit of my line spectra to obtain line positions in pixels, which are then associated to line wavelengths from  the echelle spectrum. The association may be done following a line list, or interactively by clicking alternating on lines in my spectrum and lines in the echelle spectrum.

My camera unfortunately has strong etalonning, probably due micro-lenses. It appears that spectral features (wave patterns) are strongly dependent on beam collimation. The usual flat fields (from uncollimated light) yield wave patterns quite different from stellar spectra (where the incoming light is perfectly parallel). Therefore, the intensity calibration uses reference spectra from bright stars  (mostly spectral type A0) but that remains problematic around strong absorption lines. For the region around H-alpha I'm experimenting with B3 stars.

Since the spectrograph was built around the camera, a major modification would be required when using another camera.

My current observations of H-alpha emission in Miras, which is narrow compared to the etalon waves, are not strongly affected. As long as only narrow emission/absorption lines are studied.