First steps with the LISA: question about the best focus.

[Simon de Visscher]

Dear all,

I have recently bought a second-hand LISA in pretty good state, the idea behind is that for faint targets the 35um slit is better adapted to the 2m F.L. of my instrument than the 23um slit of the ALPY, considering the large variance of seeing that I have on my observation site.

The first acquisitions have shown me the obvious: the LISA is a far more demanding tool than the ALPY. I had read it across discussions on the web, now I understand why the LISA is not a spectrometer for remote observations

I have of course started playing with the LISA on a table, installing the two cameras (ASI533 for science, ASI178 for guiding), and trying to get a coarse focusing. With the current ~25 degrees temperature difference between inside the house and outside at 2am, I didn't expect the focusing to be stable of course.

After a couple of trials (it started well, then unfortunately I degraded the quality of the spectra up to the point ISIS could not perform a wav calibration), I could understand a bit the interplay between the telescope and spectro focusing, in particular how sensitive is the entire system to the coincidence between the telescope and guiding camera focal planes, and of course the science camera focus w.r.t. the slit plane.

Here is what I get now (HD29722, A1Vn):



Reading especially http://www.astrosurf.com/buil/isis/lisa_ri/index.htm it seems my focus is a little too much optimized on the red part of the spectrum, but from my limited experience with the LISA it is hard to tell whether I can get really better. This is visible for instance looking at the raw spectrum from the nice Be star HD276886: we see that the Ha emission is small, but I wonder if this is not too small, meaning a degradation of the blue part of the spectrum.



So the question is: shall I try to focus more towards the green part of the spectrum? The difficulty is here that there is no strong and narrow absorption line between red and green on A/B stars spectra which could be used to do this. So I only have to look at the (vertical) width of the spectral line? Btw the resolution I get here is close to 700.


Thanks a lot !

Simon

[Rainer Ehlert]

Hi Simon,

I am not an expert but I would say the first spectrum is better then the second one.

I have a LISA too and I always aim for a focus like the first spectrum. Blue will be always hard to really get in focus.

IMHO

Rainer

[Simon de Visscher]

Dear Rainer,

Thanks for replying! Actually both spectra are obtained with the exact same focus. The difference is that the first one was obtained with a 15s exposure on a bright star , while the second one is obtained after 300s on a mv~9 Be star. The seeing was really poor and hence affected more the second one due to the mouvements of the star pattern along the slit, diluting then signal a bit. But on the direction of the diffraction both are exactly the same.

Best,

Simon

[Jacques Michelet]

Hi Simon,

Happy to see the Lisa works well.
I agree that achieving a reasonnably good focus is somewhat tedious. To do so, I used a isolated street lamp located in a mountain village that is ~3km away. And, by chance, its bulb is still a neon bulb. So I took spectra of this bulb, selected the 'yellow' line at 585,2 nm, and achieved focusing by reducing its FWHM to its minimum (Isis includes a FWHM measurement).
So, in a nutsheel, I tried to adjust the focus in the center of the Lisa spectrum range. 585 nm is not that far for this center, so I found this way as a pretty good and handy solution. However, the blue fish tail cannot be fully removed, it is more or less a 'design feature'.
Hope this helps,

Jacques

[Simon de Visscher]

Hello Jacques,


Oui le LISA a déjà bien pris du service, je ne suis amusé avec quelques Be, une NP extrêmement faible, la comète du moment,…je l’ai déjà équipé d’un module SPOX pour gérer les sessions à distance (50m ).

Le premier focus que j’ai fait était plutôt très bon (ISIS me donnait une résolution supérieure à 800 avec la fente de 35um). C’est ensuite lorsque j’ai essayé d’affiner la MAP que ça s’est dégradé.

Une chose que j’ai remarqué, c’est que la « pente » des raies de Balmer (cas le plus simple à observer) change légèrement en fonction du focus. Je parle bien ici du grand axe de l’image (raie d’absorption) de l’étoile au niveau du capteur, donc en quelque sorte la PSF de l’étoile à travers le spectro. Je ne connais pas les aberrations du LISA donc c’est peut être une mélange d’astig et d’AS. Par ailleurs les pentes des raies principales sont différentes entre elles.

Lors de la prochaine session je vais essayer d’utiliser ce marqueur par exemple la pente de la raie Ha) pour retrouver la première MAP qui était excellente.

Bonne soirée,

Simon

[Rainer Ehlert]

Dear Rainer,

Thanks for replying! Actually both spectra are obtained with the exact same focus. The difference is that the first one was obtained with a 15s exposure on a bright star , while the second one is obtained after 300s on a mv~9 Be star. The seeing was really poor and hence affected more the second one due to the mouvements of the star pattern along the slit, diluting then signal a bit. But on the direction of the diffraction both are exactly the same.

Best,

Simon

Hi,

I understand this in that way that you did not change the focuser position and if that was the case then the temperature change did make your second spectrum unfocused. Telescopes shrink when getting colder and so one has to focus new.

One more thing is you can not compare a 15s exposure image with a 300s exposure second as they look as you have after stretching the histogram.

[Hamish Barker]

I expect you are already familiar with the sequence of focusing all three (spectrograph,guide and telescope) focal planes, but in case not, here is my procedure:

For my own spectrograph, I focus the camera using either a distributed light source such as the daytime sky (when one can have exactly the same exposure if the instrument is always pointing to blue sky, and I can see how good the focus is by how well defined the absorbtion lines are), or the calibration lamp (again, how well defined the long emission lines are at diffferent points in the spectrum - but there is a complication since their brightness as well as how many are present varies with wavelength.

Once the spectrograph is focussed, I tape the focuser with aluminium tape so that it does not get bumped or drift.

Next, again with the telescope pointing at blue sky or a flat lamp, I focus the guide camera (forget about the telescope focuser). It is easy to see when the slit is at best focus. One can also see when one is focusing too far out, as one will start to focus on the dust particles which sit on the front surface of the glass slit plate (the actual chrome slit is on the rear surface of the slit plate.).

The last thing I do is always the telescope focus. Here again, I can use the daytime sky and look at how well focused the absorbtion lines are. But also then checking with a bright star, as you have. But my telescope is a purely reflecting optic (an SCT is not), so I know that any chromatic error is purely the spectrograph lens (mine is a littrow confuration where the same lens is used in double pass).

As you have an SCT, there might also be a little bit of chromatic error towards the violet end of the spectrum, although I am guessing this is probably much less than is the case with refractors and may be anyway hidden within a typical 3 arcsecond seeing disc. I defer to those with SCTs (Robin Leadbeater is probably here - I think his experience with chromatic issue in scts is more when using a focal reducer, any comments?) for their experience.