I also noticed what Francois shows - without emission -, but at least an emerging level:
- 2017_0226_0715.png (12.5 KiB) Viewed 18084 times
OFF: Christian, I've played last year lot with the ATIK 414 + LHires III.
The internal flat is highly reproducible (same values per wavelength - even after 1 year), just the spectrum not
(ripples shifted by like a +/- 0.3A, varying depending on the spectrum Y position but their amplitude also don't match)
External flat is better (ripples amplitudes much better match), but need a nearby good ref.star (Vega) to shift the ripples to a good position. Or just use that star to fix the ripples (taking out from flat, as flat ripples still shifted).
What is sure, don't use the flat in the usual way, need the ripples image decomposed and shifted (and maybe multiplied).
EDIT: I've even wrote a sine .fits generator in C (using libfitsio), the sine function perfectly matches, just the amplitudes modulated in some random way.
ON
- Peter
EDIT#2: before anybody asks, my spec. on 2017.07.15 got unusually good by SNR: min ADU (H-alpha) = 1670, max ADU (continuum) = 2941, ATIK 428 EXm, bin2x2, exptime=60 seconds. (FHWM by eye should be like a 4 pixel)
