PDA

View Full Version here: : Newbie having some fun with spectroscopy


julianh72
03-03-2016, 01:10 PM
The following really belongs in a sub-forum for "Spectroscopy Beginners Start Here", but I thought I would post here to share my beginner's excitement, and as an encouragement to anybody who is thinking of dabbling in spectroscopy. (If you are, my advice is "Do It!" - you won't be disappointed!)

There are no great scientific revelations here, but it is really quite exciting to be able to re-discover for yourself some of the fundamental physics of the universe, using only very affordable hobby-grade equipment. I am using a Meade 8" LX-90 (although a more modest telescope can be used just as effectively - I've grabbed some nice spectra with my 90 mm Mak), a mono QHY5 camera, and a Star Analyser SA-100 eyepiece filter grating.

I took a range of spectra last night on some favourably positioned stars (and one emission nebula – Eta Carinae), and then stacked and scaled the spectra to compare them with each other. Some trends become very apparent:

• The position of the peak of each spectrum correlates with the surface temperature of the star.
o Acrux and Becrux are super-hot (> 20000 K surface temperature)
o Sirius has a surface temperature of ~ 9900 K
o Procyon has a surface temperature of ~ 6500 K
o Gacrux is a red giant with a surface temperature of “only” about 3400 K.

• The strong dips in the near-infra-red are visible in all of the plots
o This is atmospheric absorption, rather than being an intrinsic feature of the stars’ spectra. (I guess that's why all the best infra-red observatories are in space, above the Earth's atmosphere!)
o This will limit the amount of useful “fine detail” I will be able to collect in the infra-red, although I can still see the fact that red giants such as Gacrux emit well into the infra-red compared with the hotter stars.

• Hydrogen absorption is visible in all of the stars - but look at the strong hydrogen emission lines from the Eta Carinae nebula!!!
o There is not as much relative hydrogen absorption visible in Gacrux, but lots of other strong absorption lines around 6000 to 7000 Angstroms. I believe these are associated with absorption of other metals such as Calcium and Titanium, reflecting the fact that as a red giant, it has burnt most of its hydrogen, and is now burning helium and even higher elements.

Not bad! Real Science, all done with a decent hobby-class telescope, a cheap astro-camera (~$200), and a diffraction grating filter (~$150)!

Next step: Chase down some interesting spectroscopy targets - maybe a red-shifted Quasar or two, and a couple of Wolf-Rayet stars. Any suggestions for some interesting (but not TOO challenging!) targets?

Merlin66
03-03-2016, 02:45 PM
Julian,
Well done!
Your excitement comes through from your report....
You've taken a small step on the long journey that is spectroscopy.
A couple of technical questions:
What's the distance from your SA100 to the CCD Chip? (This obviously impacts on the dispersion and final resolution.

What software are you using to prepare the spectral profiles?
I recommend BASS Project for processing https://uk.groups.yahoo.com/neo/groups/astrobodger/info

BASS will allow you to make corrections to your profiles to reflect the response curve of your camera, it also has a significant collection of stellar reference spectra to assist in the calibration (wavelength/ pixel) and the identification of various features etc.
Download a copy of Richard Walker's "Spectral Atlas)
http://www.ursusmajor.ch/astrospektroskopie/richard-walkers-page/index.html
(The English version is listed down the bottom right hand side)

Regards southern hemisphere objects - I would get more practise on A type stars to refine the calibration using the Balmer lines etc. then move on to the more difficult objects.
WR stars are exciting targets but many of then are rather faint.
Gamma2 Vela is among the brightest and visible early morning
RA 8h 9m 32
Dec -47d 20.1m
Then there are also the emission line Be stars!

julianh72
03-03-2016, 04:13 PM
Thanks for the encouraging feedback, Merlin. (Following your posts on IIS and elsewhere was part of what first got me encouraged to have a try with amateur spectroscopy.)

I'm using the grating in a 1 1/4" nose-piece which screws into the front of the QHY5 camera body. I haven't measured the distance from the grating to the sensor exactly, but it would be about 40 mm or so. I'm getting about 12.2 Angstroms per pixel, when I calibrate between the zero-order and the 4861 Angstrom H-Beta.


I've been using RSpec. It has a suite of tools for sensor response curve correction, optimising exposure and focus, superimposing reference lines and curves, etc. I've only just this week started moving beyond the "taking pretty rainbow pictures" phase (it's ironic that I'm getting better spectral data capture with my mono camera than I do with my one-shot colour camera!) to starting to analyse the images I'm capturing, so I haven't started correcting and optimising my data yet.

I'll take a look at BASS as well - but there's a lot to be said for getting to know one software tool well rather than fumbling around with lots of them!


I've already downloaded a copy - what a fantastic resource for the amateur backyard spectroscoper!


Thanks for the tips. Yes, I was going to concentrate on A-type stars for a while to hone my skills, but capturing Gacrux and Eta Carinae last night were a bit of a "Wow!" moment for me, because their spectral curves were immediately and dramatically different to all of the other stars I had seen so far. (Eta Carinae in particular was immediately obviously different when just looking at the monochrome spectrum image, even before I processed it to extract the curve.) It will be nice if I can make a point to look at something "interesting" on most nights!

julianh72
10-03-2016, 10:21 AM
Last night was nice and clear (but very dewy!), so I made a point of shooting some spectra of stars that I chose to provide a broad spread of spectral classes.

I then combined all of the calibrated data into a single spreadsheet, then sorted them into a series from hottest to coolest, then scaled and plotted against an arbitrary intensity scale, giving me the attached diagram, which I am pretty happy with. "That's going straight to the pool room!" :lol:

(The vertical scale is arbitrary; the individual curves have all been scaled to show the same indicative range, and I have not factored in my "Instrument Response" curve yet - that's the next step in my journey.)

Equipment / software used:
Meade 200 mm LX-90 ACF
Star Analyser SA-100 Grating
QHY5 Mono Camera
Captured as video using FireCapture v2.4.12
Stacked and sharpened using RegiStax v6.1.0.8
Images processed using RSpec 1.8.0 (Build 14)
Processed data sorted and plotted in Excel

Merlin66
10-03-2016, 12:31 PM
Julian,
Looks like your still enjoying yourself!

Much as I like Tom and RSpec, I would suggest you check out BASS project.
It can prepare calibrated multi spectrum graphics and annotations directly from your spectral images.

As you say the next step is to prepare and apply an instrument response..
Onwards and Upwards