spectro-aras.com

Bonjour a tous,

Je voudrais vous demander s'il est possible de calculer la vitesse radiale minimale atteignable par un spectroscope à partir de ses paramètres: lignes/mm du réseau, focale du collimateur, taille de la fente... Ou à partir des données fournies par la feuille de calcul "SimSpec" de Christian Buil.

Merci de votre attention.

Not direct from the spreadsheet. Radial velocity precision depends on many factors (spectrograph resolution, brightness of the target, how many lines are measured etc) but one of the most important is the stability of the spectrograph. It is also important to distinguish between precision (how small a change you can measure) and absolute accuracy, which is much harder to achieve

As a very rough guide, in my experience it is possible to measure individual line shifts due to radial velocity changes to a precision of ~1/20 of the spectrograph resolution in a spectrum with high SNR. For example with R~1000 you could measure to say 15km/s precision. You can then improve this by measuring many lines, for example using cross correlation but this needs a spectrograph with very good stability.

Robin

Fibre fed spectrographs have a big advantage here because they are more stable. Also they do not have the problem of shifts caused by small changes in the position of the star on the slit.

Robin

Very thanks for Your reply.
Amicalement

It's possible with an eShel spectrograph to measure low radial velocity using a process with a CCF.

See this article written by Christian Buil on exoplanet detection :
http://www.astrosurf.com/buil/extrasolar/obs.htm

Because eShel is very stable (no mechanical flexion because it's separate from the telescope and you can put this spectrograph in a thermal regulation enclosure).

Olivier GARDE wrote:It's possible with an eShel spectrograph to measure low radial velocity using a process with a CCF.

It is even possible to get this level of precision using CCF with a LHIRES but only on one particular star
http://www.spectro-aras.com/forum/viewt ... f=6&t=2800

David Boyd's presentation "Observing with a LISA spectrograph" (slides 44-52) is a nice example of measuring binary star RV to 6km/s 1 sigma precision using a telescope mounted spectrograph at R~1000 and cross correlation
https://www.britastro.org/downloads/15701


Cheers
Robin

Robin Leadbeater wrote:
It is even possible to get this level of precision using CCF with a LHIRES but only on one particular star
http://www.spectro-aras.com/forum/viewt ... f=6&t=2800

A more realistic example with the LHIRES might be my measurement of the RV of the single K 7699 line during the eclipse of eps Aur to ~0.8 km/s 1 sigma precision at R~20000 in fig 6 of this paper
https://arxiv.org/ftp/arxiv/papers/1206/1206.6754.pdf
but that needed a special calibration technique using telluric lines as the LHIRES is not that stable.

Robin

Hello at all,

I also want to relate to you the NOU_T thermal comportament, and I  would like to show that in a series of 10 integrations x 400 seconds to RS Oph  made on 27.08.2021.

You know that the NOU_T works directly to the telescope, so without any watertight compartment and  without any optical fiber, that observation was made with an ASI2600MM and the SC telescope from Piera.

Day 27th August 2021 day temperatures.
UTT  14:00  Temp. 33.9º C,    UTT  16:00  Temp. 29.5º C,  UTT 18:00 Temp. 26.2º C.

Position "x" and "y" of the same point in the #34 Order,  ASI2600MM binning 1x1, sizepixel 3.76 x 3.76 microns.

INT. NUMBER            X and Y

     1                       1706,2083     UTT 21:32:19  Temp. 24.3º C   
 2                       1708,2083   
 3                       1706,2083
     4                       1707,2083
     5                       1707,2083
     6                       1708,2083
     7                       1708,2083
     8                       1708,2083
     9                       1708,2083
    10                      1709,2083    End UTT 22:36:26   Temp 23.2º C

So it is possible with those results the NOU_T is very useful for all almost type of observations, that is not a supposition, that is the reality.

You can catch these 10 integrations in the file,  20210827_NR85.zip on, https://drive.google.com/file/d/1zImx8p ... sp=sharing, and you will see and play with them if you have some doubt about that I have said.

Best whises, J.Guarro.

Hi Joan,

That looks good stability and with an echelle you have the advantage of many lines to measure with CCF. (With radial velocity measurements though even <0.1 pixel error is significant.)

I think there is also a fundamental limit to how precise you can measurement the centroid of a line using a slit spectrograph. A small change in the position of the star on the slit changes the shape of the line slightly which is significant when you are trying to measure the position to a small fraction of the width of the slit. The fibre feed has an advantage there as the profile of the light on the slit/fibre is redistributed by the fibre.

Cheers
Robin

Hi Robin,

Many thanks for your precise explanations, and because I wrote "all almost". On the other hand, the NOU_T doesn't have a slit, it has a 50 microns diameter air pinhole.

Yesterday when I wrote the first text - you can believe me - I didn't remember that I made some pieces that allow passing the NOU_T from the "direct injection" at the telescope, to an optical fiber system, you can change it in only 30 minutes. 

I didn't text that in the NOU_T, but I have experience with the optical fiber on my MUSSOL and LINX echelles and sure this system works well.

You can see it in this picture with an ASI2600MM, but without optics, of course, that model can work with 85 mm. and 135 mm. objectives, the camera cmos size permites it.

But in my humble opinion it is more easy and and practical to observe on "direct injection", so it can take very well the "normal" observations that we make.

BTW, I don't know if for one better light distribution, the fiber patch cables that we use,they would have an achromatic lenses at the same way as the Caos Group show there, http://www.eso.org/~cguirao/caos/presen ... 4.pdf,what is your opinion ?

Cheers, Joan.