instrument respons LHiResIII-2400l/mm grating

[Erik Bryssinck]

I had a question regarding the correct star for instrument response with LHiResIII and 2400 l/mm grating for H-alpha and HeI @6678 Å.

Generally speaking, I have applied the following rule until today: I use a set of reference stars depending on the period of the year: Altair for summer, Castor for winter and Regulus for spring. I may sometimes also use Vega. I don't take a spectrum from the star every night to determine the instr. response. Usually I use the same correction for a few months. If this is not correct I would like to hear it

So far this has worked quite well, but for some time I have had problems with the correct continuum of the obtained spectra.

With the spectra of gam Cas for the HeI, I notice that the difference is now very clearly visible. The continuum changes as function of the height of the observed star, I suspect that this is the cause, I am not sure.
Example below is the spectrum with instr. response of 15Cas and Altair. 15 Cas is more close to gam Cas and Altair not at all. (blue=Altair for instr. respons - green=15Cas for instr. respons)


Until now I usually only used spectra with the LHiResIII for the BeSS database in high resolution and H-alpha in the centre of the image and instr. response as described above.

The question now is which star is best used to determine the instr. response for LHiResIII and 2400 l/mm grating and is there a difference depended in which part of the spectrum we work - to make the correct choice of the star used for instr. response ?

thanks !
Erik

[Robin Leadbeater]

Hi Erik,

The change with air mass is only small over such a small range in the red

A thought. Are you using ISIS and have you updated the version you are using ? The way ISIS uses the flat changed after 5.9.3 (The Planck correction has been removed.) This means that responses measured on the old version do not work correctly on spectra reduced using later versions

Robin

[Robin Leadbeater]

Erik,

I have just noticed the scale. The differences are quite small ~3%. What source are you using for the reference star spectra? perhaps one of them is not so well flux calibrated

Cheers
Robin

[Erik Bryssinck]

Hi Robin,

Thanks for your reply!

Hi Erik,
A thought. Are you using ISIS and have you updated the version you are using ? The way ISIS uses the flat changed after 5.9.3 (The Planck correction has been removed.) This means that responses measured on the old version do not work correctly on spectra reduced using later versions
Robin

I use AudeLA with SPcAudace for reduction, but i can try with ISIS and see what i get there... i think i need some time to use ISIS again, i used ISIS first a long time agoo.

I have just noticed the scale. The differences are quite small ~3%. What source are you using for the reference star spectra? perhaps one of them is not so well flux calibrated

I use the ref. spectra that are available in SPcAudACE. don't think there is a problem otherwise other user should experience the same issue, i suppose.

cheers
Erik

[Robin Leadbeater]

Hi Erik,

OK we can rule out the changes in ISIS as the cause then. (no need to change to ISIS) Have you used 15 Cas before as a reference? If your spectra reduced using Altair, Castor and Regulus have been agreeing before then I would say the 15 Cas library spectrum is the most likely cause. (Have you used it successfully before?) You could test this by taking a spectrum of 15 Cas using Altair (which is still high in the early evening) to flux calibrate and see if the result agrees with the library version. The difference in air mass between Altair and Gamma Cas/15 Cas in the early evening is only ~0.2 at the moment (UK) which will make very little difference at this wavelength over this narrow wavelength range.

Cheers
Robin

[Robin Leadbeater]

I could not find 15 Cas in my usual sources for high resolution reference spectra (UVES POP and ELODIE 3.1) Do you have any information on the source of the 15 Cas library spectrum eg from the fits header?

Robin

[Robin Leadbeater]

Hi Erik,

The change with air mass is only small over such a small range in the red

I made a quick check using Francois Teyssier's extinction calculator spreadsheet.
http://www.astronomie-amateur.fr/ProjetsSpectro0.html
The error in relative flux over this wavelength range between a reference star at 30 deg and a target at 80 deg would be ~ 0.5%

Robin