About atmo. dispersion and optical chromatism effects

Show your spectra, your results ...
Post Reply
Christian Buil
Posts: 1431
Joined: Mon Sep 26, 2011 6:59 pm
Contact:

About atmo. dispersion and optical chromatism effects

Post by Christian Buil »

La dispersion atmosphérique est toujours un problème en spectrographie
large bande et à fente étroite. Ceci est indépendant du type de spectrographe (eShel, Harps, LISA, Alpy 600, ...)
si un système ADC n'est pas utilisé (ce dernier est très complexe d'usage).

J'ai mis sur cette page quelques observations concernant ce problème. J'espère ne pas
trop vous effrayez ! C'est un peu brut de fonderie, dites si vous êtes perdus. C'est un sujet important et
il ne faut pas ce voiler la face, tout ceci existe. C'est ici :

url]http://www.astrosurf.com/buil/dispersion/atmo.htm[/url]

A présent que des spectrographes comme Alpy vont populariser définitivement
la spectrographie amateur (j'espère), il est bon de dire que la vie du spectroscopiste n'est pas toujours simple ;)
Mais ca fait partie du jeu et d'une forme de plaisir !
Le message est que le spectrographe est bien sur loin de faire toute la qualité
d'une observation. La stratégie, le télescope, la rigueur dans l'étalonnage... tout ceci est fondamental.

===================================================================================================

Atmospheric dispersion is always a problem for broadband spectroscopy and with narrow slit. This is independent of the type of
spectrograph (eShel, Harps, LISA Alpy 600, ...) if ADC is not used (Atmosphéric Dispersion Corrector - a complex setup).

I put here some comments on this problem. I hope I did not scare you too!
But the subject is important. Look the page:

http://www.astrosurf.com/buil/dispersion/atmo.htm

Now that spectrographs as Alpy will definitely popularize amateur spectrograph (hopefully), it's good time to say that the life
of the spectroscopist is not always simple. The spectrograph is far to give all the quality of an
observation. Observation strategy , telescope quality (or type), rigorous calibration ... All this is fundamental.


Christian B
Robin Leadbeater
Posts: 1928
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: About atmo. dispersion and optical chromatism effects

Post by Robin Leadbeater »

Scary stuff indeed :shock:
For me it makes taking reference stars for each observing session under as near as possible the same atmospheric, focus and guiding conditions even more important to have the best chance to cancel these effects.

Robin
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Martin Dubs
Posts: 141
Joined: Fri Jan 06, 2012 9:16 pm
Location: Maienfeld, Switzerland

Re: About atmo. dispersion and optical chromatism effects

Post by Martin Dubs »

Hello Christian,

you are quite right, it is a difficult subject. It gets even more complex at long focal length, such as the IAC80 F/11 and with a fiber coupled echelle, where you do not have the possibility to orient the slit. We thought about an ADC, but do not use it for several reasons:
- not enough backfocus
- too complex for people not used to it, there are plenty of other things you can do wrong
- too expensive
We hoped to reduce the problem with a focal reducer, which also reduces the length of the atmospheric spectrum on the fiber, but in exchange introduces chromatic aberration. Fortunately the elevation limits on the telescope are set so that spectra at very low elevation are not possible.
Since we are interested more in lineshapes than the continuum shape, the problem is manageable.

Interesting post, Christian
Francois Teyssier
Posts: 1521
Joined: Fri Sep 23, 2011 1:01 pm
Location: Rouen
Contact:

Re: About atmo. dispersion and optical chromatism effects

Post by Francois Teyssier »

Robin, I do agree
In order to find a reference star near the target I use an excel spreadsheet : http://www.astronomie-amateur.fr/Templa ... ne-xls.png
In addition, a good refrence star must have low luminosity in order to give exposure times of about 20 seconds or more ; statistically, you'll avoid the "edge effect"
Vega is definitively a very bad standart for low resolution specytroscopy

Finally, with standart procedure you describe, one can get very good results
See for instance the comparison of the some star (a Be, with great interstellar exctinction) ; Christian and I got these spectra regardless (i.e. wwithout having previously established a procedure)
HD344313_Comparison.png
HD344313_Comparison.png (3.74 KiB) Viewed 12821 times
François
Andrew Smith
Posts: 321
Joined: Fri Sep 30, 2011 12:23 pm

Re: About atmo. dispersion and optical chromatism effects

Post by Andrew Smith »

Christian - a very impressive piece of work and experimental results. While I understand the need not to confuse it is important we bring these issues out.

The lessons for me are that if you want an accurate spectra over a significant wavelength range you need to ensure the slit is allowing 99%+ of the star light into the spectrograph i.e. the slit has to be wide compared to the linear dimensions of the chromatic star image on the slit or fibre. This may limit the resolution R but what good is a higher R if the spectra is not accurate?

It would be good if we could all match our choice of telescope & spectrograph to our science / hobby objectives but money, available equipment and skill if DIYing prevail.

I was interested in the comment by Martin on echelle fibre input I was hoping (having only done limited calculations) that a 50micron fibre with my 300mm F5.4 Newtonian would accept a reasonable degree of chromatism - time will tell as the build is going very slowly as I accumulate the necessary bit.

Thanks again Christian.

Regards Andrew
Robin Leadbeater
Posts: 1928
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: About atmo. dispersion and optical chromatism effects

Post by Robin Leadbeater »

Hi Andrew,
Andrew Smith wrote:
The lessons for me are that if you want an accurate spectra over a significant wavelength range you need to ensure the slit is allowing 99%+ of the star light into the spectrograph i.e. the slit has to be wide compared to the linear dimensions of the chromatic star image on the slit or fibre. This may limit the resolution R but what good is a higher R if the spectra is not accurate?
Note that Christian's measurement variations are absolute. Using a reference star under similar conditions will reduce the magnitude of these effects considerably.
Also I think we have to be a bit careful not to generalise here. For many (probably most) applications an absolute flux measurement or even an accurate continuum shape is not needed so I would reverse your statement and say what good is an accurate continuum if the resolution is not high enough to show the phenomenon you are interested in? For example you can still measure the EW etc of a line relative to the continuum just as well, even in a completely uncorrected spectrum. Even the UVES database does not guarantee the continuum calibration better than 10% ;-)

Cheers
Robin
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Robin Leadbeater
Posts: 1928
Joined: Mon Sep 26, 2011 4:41 pm
Contact:

Re: About atmo. dispersion and optical chromatism effects

Post by Robin Leadbeater »

Hi Francois,
Francois Teyssier wrote: In order to find a reference star near the target I use an excel spreadsheet : http://www.astronomie-amateur.fr/Templa ... ne-xls.png
Your link failed unfortunately. A significant problem seems to be finding enough reliable actual spectra of reference stars.(Using Miles spectra seems to work well but there are not so many suitable hot stars in Miles and generic spectra like Pickles etc gives many problems.) The photometrists have people like the AAVSO staff continuously measuring potential comparison stars for them but we are not so well off for good standard spectra.

Cheers
Robin
LHIRES III #29 ATIK314 ALPY 600/200 ATIK428 Star Analyser 100/200 C11 EQ6
http://www.threehillsobservatory.co.uk
Andrew Smith
Posts: 321
Joined: Fri Sep 30, 2011 12:23 pm

Re: About atmo. dispersion and optical chromatism effects

Post by Andrew Smith »

Robin Leadbeater wrote:Hi Andrew,
Andrew Smith wrote:
The lessons for me are that if you want an accurate spectra over a significant wavelength range you need to ensure the slit is allowing 99%+ of the star light into the spectrograph i.e. the slit has to be wide compared to the linear dimensions of the chromatic star image on the slit or fibre. This may limit the resolution R but what good is a higher R if the spectra is not accurate?
Note that Christian's measurement variations are absolute. Using a reference star under similar conditions will reduce the magnitude of these effects considerably.
Also I think we have to be a bit careful not to generalise here. For many (probably most) applications an absolute flux measurement or even an accurate continuum shape is not needed so I would reverse your statement and say what good is an accurate continuum if the resolution is not high enough to show the phenomenon you are interested in? For example you can still measure the EW etc of a line relative to the continuum just as well, even in a completely uncorrected spectrum. Even the UVES database does not guarantee the continuum calibration better than 10% ;-)

Cheers
Robin
Fair enough Richard, but as I said, I was commenting on "an accurate spectra over a significant wavelength range " - I should have said "as accurate as possible". If you want to measure the EW of a few lines then I fully accept what you say.

Regards Andrew
Andrew Smith
Posts: 321
Joined: Fri Sep 30, 2011 12:23 pm

Re: About atmo. dispersion and optical chromatism effects

Post by Andrew Smith »

It should have read "Fiar enough Robin" - sorry!
Francois Teyssier
Posts: 1521
Joined: Fri Sep 23, 2011 1:01 pm
Location: Rouen
Contact:

Re: About atmo. dispersion and optical chromatism effects

Post by Francois Teyssier »

Sorry Robin

Here's the good link :
http://www.astronomie-amateur.fr/Docume ... inder.xlsm

You've raised the important question of reference database repeatedly ; I propose we open a specific topic about this important subject

François
Post Reply