robin_astro

If you have a deep sky capable camera you can take a spectrum of bright supernovae like  this just by putting a diffraction grating between the telescope and camera. Here it is with a Star Analyser 100, Atik 314 and CII  (18x20 seconds)

You can see the supernova (and other stars and galaxies) and the light spread out into the spectrum, blue to red

Cheers

Robin

13 hours ago, callisto said:

Ahh good, not just me then 😂

Questions welcome, just ask away !

Supernovae are a particular interest of mine.  They were once mostly discovered by  amateurs as this one was  but now most are found by professional survey telescopes. When you discover what is possibly a supernova you then have get it confirmed and determine what type it is by taking a spectrum.  (I started doing this in 2016 and have about 40 official classifications to my name now)  This one was confirmed officially from a spectrum by  Italian amateur Claudio Balcon (just 50 minutes after it was announced) so it is an "all amateur" supernova.

Supernovae are exploding stars of course but there are two main types.  Some are massive stars which suddenly run out of fuels and collapse under gravity causing the explosion (type II). This one though is a white dwarf in a binary star system which has been grabbing material from its companion until it reached critical mass and exploded like a nuclear fusion bomb. These are the type 1a which are used to measure distances and led to the discovery of dark energy for example

Cheers

Robin

I love the blue colour, confirmed by the spectrum which is brighter at the blue end

Here is the raw  spectrum image. (Using an ALPY 600 spectrograph)

The deep absorption line marked is from singly ionised Silicon formed by nuclear fusion in the thermonuclear explosion of  the white dwarf (this is a type 1a). 

(The other lines in the background are from natural airglow and light pollution)

and here is the calibrated spectrum 

We can measure the velocity of the  material thrown out by the explosion by how far the absorption line has been blue shifted due to the doppler effect (~15000 km/s.) The silicon is actually from the partially "nuclear burned" material near the surface of the white dwarf. Heavier elements up to Iron are produced  by further fusion deeper within the explosion. It is near maximum at about mag 12.5 currently which, taking into account the distance to the galaxy makes it several billion time more luminous than the sun

Cheers

Robin

For stars to test my equipment and processing though I use MILES stars which is a catalogue of stars measured using professional equipment. Here are examples where I have  used them with the Star Analyser and the ALPY600 

http://www.threehillsobservatory.co.uk/astro/spectroscopy_21.htm

(The spectra are as measured ie not corrected for interstellar extinction so what we want) They are included in the ISIS software database but I don't think they are in RSpec. They can be downloaded from the references there though

Cheers

Robin

17 minutes ago, robin_astro said:

Note though that these appear to be the raw spectrum images, not the calibrated spectra so you haver to process them your self which limits their usefulness

Looking more closely at these, I would forget it. They are not particularly good, for example they have made a fundamental mistake of not aligning the grating horizontally which produces artifacts. Also most of the stars are boring and interesting unusual stars for low resolution spectroscopy like Wolf Rayet stars which show up really well with the Star Analyser seem to be missing. A pity because it must have been a massive project

7 minutes ago, robin_astro said:

If you want  spectra of stars measured using the Star Analyser then this database might be useful. There are thousands of stars there but I have not checked it for quality

https://sdc.cab.inta-csic.es/sasdaba/

 

Note though that these appear to be the raw spectrum images, not the calibrated spectra so you haver to process them your self which limits their usefulness

If you want  spectra of stars measured using the Star Analyser then this database might be useful. There are thousands of stars there but I have not checked it for quality

https://sdc.cab.inta-csic.es/sasdaba/

Also the ELODIE archive which has professional spectra (but at much  higher resolution than the Star Analyser) for many stars

(A bit of Trivia:- ELODIE was the spectrograph used to detect the first exoplanet 51 Peg b in 1995)

http://atlas.obs-hp.fr/elodie/

Cheers

Robin

Hi Steve,

There are many professional catalogues of spectra but  probably the easiest way for "normal" stars is to use Brian skiff's magnum opus "Catalogue of stellar spectral classifications" which has the published spectral types for almost a million  stars.

http://vizier.u-strasbg.fr/viz-bin/VizieR?-source=B/mk

You can then use the classification there to call up the typical spectrum of that type from the Pickles catalogue which is included with several software packages like Visual Spec, RSpec ISIS etc.

Cheers

Robin

A big chunk of the cost of the ALPY is the guider module (essential) but  the 3D printed module used in the Uvex can be used instead with the critical core module part  from Shelyak

https://spectro-uvex.tech/?p=1820

In the same 3D printed vein, Christian Buil's Starex which can also double as a spectroheliograph in Solex mode (lots of builds of this going on currently supported through forums)

http://www.astrosurf.com/solex/sol-ex-stars-en.html

Also his Uvex which is a superb wide spectral range all reflective design (now also offered commercially by Shelyak), though it can be tricky to align.

https://spectro-uvex.tech/?p=1362&lang=en

http://www.astrosurf.com/buil/UVEX_project_us/

Kits of the special optical parts for these are available from Shelyak

3D printed designs do tend to suffer from poorer stability though compared with the very stable ALPY

On 31/03/2022 at 21:48, vlaiv said:

What you are talking here about is interstellar reddening.

Correct. the effect of the interstellar medium on the spectrum of distant objects is well defined and different to that seen from cosmological expansion. See Lyman forest for example 

https://en.wikipedia.org/wiki/Lyman-alpha_forest

The animation there is particularly nice

Here are some measurements made by me of high redshift distant quasar spectra clearly showing the Lyman Alpha line shifted into the red from the UV and the Lyman forest from absorption by intervening material. The photons forming the rdshifted Lyman alpha emission line are the same ones that left the quasar that were not absorbed/scattered by intervening interactions

https://britastro.org/observations/observation.php?id=20210411_134753_85f4b3ebf4faaefe

For fun I thought I would try to take a spectrum of the JWST with my  modified ALPY200 faint object spectrograph. (Keeping a mag 14-16 target on the 3 arcsec wide spectrograph slit for 90 mins presents some interesting challenges, also applicable to comet spectroscopy)  After correcting for the solar spectrum it turns out that the reflectance spectrum is reddish with a tinge of blue and the green wavelengths significantly attenuated which matches NASA's description of the coatings used on the sun shield.  The spectrum and more information are on my BAA webpage here

https://britastro.org/observations/observation.php?id=20220327_125654_228ed4b0a22ce097

Cheers

Robin

5 hours ago, SteveBz said:

The width at half intensity (6.0E5 on this plot) goes from 655.0 nm to 657.7 nm giving a width of 3.7 nm.  I should probably call that 4 nm.  656/4=164 or 656/3.7 = 177, so it's a bit over 160, say in the range of 160-170.  Is that right?

Yes.  With slit based systems the resolution is fixed by the slit width but with slitless systems it will depend on the star image size (seeing etc) 

Cheers

Robin

16 hours ago, SteveBz said:

1) How do you calculate the resolution of those spectra?

2) What is a 'Star - Be' or 'Star - Be Candidate'?

1) To measure the resolution you need a spectrum line which is narrow compared with the resolution of the spectrograph.  With slit spectrographs this is usually measured from the width of one of the calibration lamp lines. with a slitless spectrograph you can measure it from a line in the star spectrum you know to be narrow. You can see an example here where I used the lines in P Cygni. 

http://www.threehillsobservatory.co.uk/astro/spectroscopy_16.htm#SA200_filter_wheel

You will often see the resolution quoted as resolving power R = wavelength/resolution so for example an R of say 120 typical of a simple Star Analyser setup equates to a resolution of 50A at 6000A while my LHIRES at maximum resolution has a resolving power of ~15000 or 0.4A

2) Be stars are main sequence B stars which intermittently show H alpha in emission, from a circumstellar disc.  They are a popular amateur target as they are part of a big pro am project to track them and better understand how the disc forms and disappears. There is a database dedicated to them

http://basebe.obspm.fr/basebe/

and a website which shows which stars need spectra currently

http://arasbeam.free.fr/?lang=en

Be Candidates are stars which have been found with Be like emission but are not currently catalogued as such. (Because discs form and disappear a normal B star can turn into a Be star) There was an amateur  systematic survey to discover some of these a few years back

http://www.astrosurf.com/aras/be_candidate/auto-be-candidate.html

and candidates pop up from time to time for example in the  Gaia transient survey where they can be followed up by amateurs eg

https://www.spectro-aras.com/forum/viewtopic.php?f=40&t=2940

Another popular area of pro-am work is Cataclysmic variables.  (novae, dwarf novae, symbiotic stars etc) There are more examples in the ARAS database

https://aras-database.github.io/database/index.html

Cheers

Robin

You could perhaps take look in the BAA database to see what others have been looking at for a particular time of year. (Most observers there are northern hemisphere, though accessibility will still be latitude dependent of course ).  They are generally at higher resolution than the typical Star Analyser spectra (you can limit the resolution in the search if you like) but you should be able to see the main features shown there provided they are bright enough for your setup

https://britastro.org/specdb/data.php

(enter the date range of interest) 

Cheers

Robin

What a bizarre and alarming incident.  Strangely not the first  telescope to be attacked though. 

https://astroanecdotes.com/2015/03/26/the-mcdonald-gun-shooting-incident/

4 minutes ago, robin_astro said:

Notice how one of the spectra of Jupiter is much brighter than the others. this is the blazed first order and the one you want to get in the frame, together with the zero order

The region I have marked as the white rectangle.  Based on your setup, it should neatly fit in your field of view with the zero order near the left edge

The series of absorption lines at the left end look like the Balmer series which suggests you are getting something at the blue end now but we really need the zero order in the same frame to calibrate it properly without guessing. Were you able to find the various spectra I talked about and chose the right one?  This is what a wide field view (of Jupiter) through a grating looks like (From Christian Buil's page)

http://www.astrosurf.com/aras/staranalyser/flat.jpg

Notice how one of the spectra of Jupiter is much brighter than the others. this is the blazed first order and the one you want to get in the frame, together with the zero order

Cheers

Robin

The uploaded tiff image seems to be 1.5x larger than the camera sensor (4644x3120 compared with  3096x2080)  has it been resampled somehow?

Cheers

Robin

Perhaps, but I dont think so. The Balmer lines and Telluric bands line up too well the original way round and the dispersion is about right for the second order (higher than calculated for the first order as expected.) The very deep band at ~7600A from O2 in the atmosphere is also good indicator and a common feature seen in all spectra taken from the ground. It could also explain the bias in the spectrum towards the IR (The blazed grating is optimised to preferentially select the wavelengths in the visible in the first order)

OK I think I know what has happened. It looks like you have captured the second order (the next spectrum out) This will be ~2x longer and much weaker than the first order, particularly in the blue. I think You will be surprised by the much shorter exposure you will need when you find the right spectrum !

Cheers

Robin

11 minutes ago, Lucas Barclay said:

Here is what the results come out as on the calculator

OK it looks like you should just be able to fit the zero order and spectrum in the frame. Check the grating is rotated so the brighter blazed spectrum is to the right of the zero order (There is a mark on the Star Analyser filter cell to show  approximately the correct orientation) and place the zero order near the left edge of the frame. To get you started, the blamer lines should then be roughly in the right place. For example H beta at 4861A should be ~4861/2.7 = 1800 pixels along from the zero order. 

It still does not explain the apparent low sensitivity in the blue (Do you have any other filters in the imaging train?) but we can worry about that once we have the full spectrum and zero order in the image

Cheers

Robin

12 minutes ago, robin_astro said:

there is a calculator there to help optimise your setup (based on my calculations)

https://www.rspec-astro.com/calculator/

Can you post a screen shot of the calculator with details of your current setup entered please

Robin

2 hours ago, Lucas Barclay said:

Here is an image of the spectra produced without taking a cut from it : 

OK I recommend rotating the grating so the spectrum is horizontal and moving the spectrum so the zero order (the star image) is visible as this is the 0 reference point we can use for the wavelength scale. You might need to reduce  the distance of the grating to fit  the spectrum and zero order in the frame. Use the RSpec calculator as your guide

Cheers

Robin

Also have a look at the tutorial videos on the RSpec website (It uses RSpec software but that was loosely based on Visuals spec and the steps are the same.)

https://www.rspec-astro.com/more-videos/

there is a calculator there to help optimise your setup (based on my calculations)

https://www.rspec-astro.com/calculator/

The slides from my BAA tutorial "Low Resolution Slitless Spectroscopy" might also be of use, downloadable from the bottom of this page

https://www.britastro.org/downloads/15701

and Christian Buil's tips for using the Star Analyser

http://www.astrosurf.com/aras/staranalyser/userguide.htm