robin_astro

Also presumably the absorption of a photon imparts some small velocity to an atom to conserve momentum and again in recoil when a photon is emitted? Dont ask me to calculate the dominant mechanisms in a particular scenario though 😉 Robin

There are a number of ways photons can transfer momentum eg to electrons, atoms and molecules. For example Directly through scattering By exciting vibration modes in a molecule Through radiationless transitions where electron excitation levels generated by photons are translated into vibration modes within the molecule The atoms/molecules then interact with each other throughout the medium (solid, liquid or gas) raising its temperature

Also this paper briefly mentions flat field correction for a slitless spectrograph on a 2.2m telescope at La Silla https://www.aanda.org/articles/aa/abs/2008/32/aa10157-08/aa10157-08.html

Hi Andrew, A couple of references to how they do it for the grism spectrograph on the HST https://ui.adsabs.harvard.edu/abs/2005acs..rept...10W/abstract https://ui.adsabs.harvard.edu/abs/2009PASP..121...59K/abstract Cheers Robin

It would be great if you come up with a solution. This so far intractable problem has been bugging me for 15 years ! It is potentially particularly troublesome when attempting spectrophotometry with the Star Analyser where the target and reference are by necessity in different positions Robin

A while back I wondered if it might be possible to use a conventional flat (without the Star Analyser) specifically to get rid of flat defects produced at or close to the camera sensor ie dust and pixel to pixel variations. It failed with dust (probably as the geometry of the beam in the dispersed spectrum is different). It should work with genuine PRNU pattern defects though if they are an issue I think Robin

If you are seeing a fixed pattern response from the camera I think this would best be corrected separately using a PRNU (pixel response non uniformity) file generated from a conventional flat which just contains this short spatial variation information. Cheers Robin EDIT as Andrew has just suggested !

Flat correction is useful for slit spectrographs and I used to flat correct Star Analyser spectra though no longer recommend it. The problem is spectroscopic flats are a mixture of position and wavelength dependent effects. With a slit spectrograph the position of a particular wavelength in the flat is fixed and defined by the position of the slit so a conventional flat taken through the spectrograph can be used but in a slitless system any particular wavelength ends up at any location across the flat image. Professional slitless systems also have this problem and the solution for them is to build up a 3D flat ie a separate "flat" for each location in the image. This is obviously impractical so the advice I generally give now is:- Keep the sensor as clean as possible to minimise dust donuts and place your reference star and target at the same location in the field. The instrument response will then take care of any vignetting type issues If you want to explore the possible errors due to not taking a flat or you are forced to measure spectra at different positions in the field then I can suggest taking spectra at different locations and seeing how much the spectrum changes for your particular setup Cheers Robin

Note it is very important to accurately remove the sky background before attempting to measure the instrument response otherwise it will not work on other spectra. Other tips:- Turn the image gain up to make sure you select the whole height of the spectrum when binning and that the sky background does not contain any of the measured spectrum or any other background stars or spectra. Smooth the published spectrum first to approximately match the resolution of the measured spectrum (ie make the Balmer lines look similarly wide so they divide out more accurately.) Remove any remaining artifacts from dividing the spectrum lines but not other features that are actually in the instrument response Do not over smooth the result. In particular, make sure it fits well at the blue end. If it does not your spectrum will rapidly shoot off either high or low. When you have the instrument response,use it to correct the spectrum you produced it from. The result should of course closely match the published spectrum. If it does not, investigate why it is wrong

And here is an example specifically for a colour camera using a Star Analyser http://www.threehillsobservatory.co.uk/astro/spectroscopy_11a.htm Cheers Robin

I think it might be worthwhile taking a step back here. There are several sources for the difference between the shape of the spectrum as measured and how it should look. The camera response is one of them and mono cameras certainly have a smoother response than colour cameras. There are other effects though including the response of the grating, the telescope optics and the earth's atmosphere so knowing the camera response is not enough. Fortunately a simple way to account for all these effects is by measuring a star that we know what the spectrum should look like and dividing our spectrum by the published spectrum. (We normally choose a hot eg A star as these have a simpler spectrum. If using the Star Analyser, this is typically the same one which is used to initially focus and wavelength calibrate the spectrum). This gives us the response of the whole system including the camera. We can then use this to correct the spectrum of any star. The process is the fundamentally the same whatever software you use. There are examples of doing this for both the Star Analyser and ALPY in my tutorials here http://www.threehillsobservatory.co.uk/astro/spectroscopy_10.htm and documents here http://www.threehillsobservatory.co.uk/astro/spectroscopy_21.htm Cheers Robin

The observation that really made me go "yes!" out loud though was, after weeks of planning, watching the ISS streak across in front of Jupiter in 2004.

I'll second that, particularly after missing out on totality in clear skies by a couple of minutes after driving across Europe in 1999. Are you in any of these photos I took in Kizilot by any chance? http://www.threehillsobservatory.co.uk/astro/spectra_27b.htm Robin

Hi Louise, Christian Buil has a nice page which goes through the design of a classical spectrograph here http://www.astrosurf.com/buil/us/stage/calcul/design_us.htm The Shelyak instrument probably most similar in concept to the lowspec is the LISA, which is also a classical spectrograph design, though the better LISA optics give better performance at the UV/IR ends and allow it to be used on larger aperture faster telescopes. The lowspec is more versatile as you can change the grating to alter the resolution. The LHIRES is a Littrow design in which the light passes through the same lens on the way to and from the grating. This was originally an open source design by Christian Buil and others and was sold as a not for profit kit by the design team. (Mine is one of these) The Shelyak company was then set up to offer it commercially. You can still find all the drawings etc here but it is not really practical to build one yourself as a one off, though some people have. http://www.astrosurf.com/thizy/lhires3/index-en.html Cheers Robin

Hi Louise, You can see Paul Gerlach's complete thread describing the development of the instrument from the original idea to the current design here on the VdS forum. It evolved significantly over 3 years https://forum.vdsastro.de/viewtopic.php?f=28&t=4432 Cheers Robin

I can now reveal Star B is 2MASS J19095965+0437597, a previously unknown variable, co-discovered with Adam Nowak https://www.aavso.org/vsx/index.php?view=detail.top&oid=1498835 Note the brightness of this star at ~mag 19 g which would make it by far the faintest object recorded spectroscopically using amateur equipment. This is a bit of a cheat though as my spectrum mainly covers the R and I bands where the star is significantly brighter (This is an M6 star, further reddened significantly by the interstellar medium, which explains its presence in the 2MASS catalogue of objects detectable in the IR at 2 microns. Cheers Robin

That has been tried eg Ken's original L200 design. The problem is not so much the transmission, (You can use ordinary glass and just reflect 10% to the guider or use a hot mirror and guide on the unused IR) The main problem is keeping it all aligned so you know where the slit is and that you are in focus at the slit. The Starlight Xpress spectrograph uses one but I have never seen any results from this spectrograph. I dont think the OVIO slit is horrendously expensive. It is used in the lowspec for example

This shows the star I recorded a spectrum of and the guide star I used. (The target is focused and moved onto the slit (black line), where it disappears as almost all the light passes though the slit into the spectrograph. It is then kept on the slit using the guide star (PHD guiding program)

The mirror slit guider uses a mirror with the spectrograph slit in it where the reflective surface has been removed. The field is viewed with a separate guide camera. This allows the star field to be viewed, the star of interest brought to focus and moved onto the slit and kept there. Without it you are working completely blind. This is an example of a typical optical configuration (The tragos spectrogaph) though all commerical amateur slit spectrographs have something similar. and what the image typically looks like for my ALPY spectrograph

With a slit spectrogaph is is really tough and frustrating even on bright targets. There are techniques which make it just possible but to be honest the mirror slit guider revolutionised astronomical spectroscopy as it allows you not only to guide but also find the target and put it on the slit in the first place and focus it when it is there Cheers Robin

The problem with Czerny-Turner and similar designs for stellar spectroscopy is the inherent astigmatism which they produce. (OK for a lab bench instrument but a problem for stars as you dilute the spectrum against the background.) Christian's clever precisely tuned optical configuration tames this to some extent and what remains is largely eliminated by the cylindrical lens. (Note any slit spectrogrph is effectively unuseable without a guiding module, the 3D printed design of which has not yet been released forthe UVEX so you would need the ALPY guide unit.) Classical designs like the Lowspec are much more forgiving to build to give good performance and the lowspec design is complete including guiding module. It does not give as wide coverage without chromatism problems which defocuses the spectrum at the extreme ends but from what I have seen, the lowspec works fine though with reasonably slow telescope optics over the usual range ~3800-7000A. (There is an example currently on ken's astronomical spectroscopy yahoo group) With a slit design, atmospheric turbulance does not affect the spectrum though it does reduce throughput. Chromatism in the telescope optics with slitless systems degrades resolution, but with slit spectrographs there is no loss of resolution. Good instrument response correction can be more difficult though due the selective sampling of the wavelength range at the slit. This is generally reasonably manageable though provided the focus does not shift between reference and target measurements. Cheers Robin

Hi Ken, Shelyak have their own software (Demetra) that they are developing for their instruments (currently only for the ALPY, I believe) Concerning ISIS, this predates the formation of Shelyak as a company. It is no secret that as arguably the world leading amateur spectroscopist, Christian Buil develops ISIS primarily for his own use so it true that it has features that make it extremely efficient when processing spectra from the spectrographs that were designed by him but it has all the tools to reduce the spectra from any spectrograph. As an example here are the various options for wavelength calibration ranging from fully manual line selection through to a one click automated system based on either a built in or a user generated custom file of calibration lines for any specific spectrograph and camera combination http://www.astrosurf.com/buil/isis/tutorial/calibration_us.htm

Yes it is very important that the calibration passes through zero at the zero order otherwise you cannot use it to calibrate other stars where the wavelength of the lines will not generally be known. This may be at the expense of precise alignment of other lines but this is less important Cheers Robin

Hi Nigella, The calibration of the Star Analyser is very close to linear so to start with you can use just the zero order and one other line eg H beta which are easy to identify. If you first calibrate using these two lines and then overlay the Hydrogen lines it becomes obvious which features approximately line up with the other lines and you can then add these to the calibration if you like. (In practise I dont usually bother recalibrating using a non linear calibration and just use a linear calibration as wavelength calibration of slitless spectra is only approximate anyway) Once you have this calibration, you can use this and the zero order to wavelength calibrate any spectra using just the zero order. Cheers Robin

I use ISIS for all my Pro-Am work as I would say do the majority of amateurs doing research grade spectroscopy but it is not so easy for a beginner, particularly with the Star Analyser. Another freeware alternative is Visual Spec which I "grew up with" though it is prone to crash.