I couldn't find the Star Analyser on the Bintel website - but Don assures me it's there (somewhere!) and that Bintel do sell it.
He doesn't stock introductory spectroscopy books....
Hmmmm
That section of "Accessories" wasn't there at 10am this morning!
If you neeed a good introductory book to amateur spectroscopy - just let me know. I have a few "signed by author" copies available ;-)
Maybe your enquiry prompted a change! Already have your "Astronomical Spectroscopy for Amateurs" book Ken (valuable part of my library), maybe I should have waited for the offer of a signature!
damn, wish I knew that was there a few weeks ago! I bought one from the US for more $$ than BINTEL are selling them for Would have liked to support BINTEL.
Rob,
Next time you're in Melbourne, let me know - we can have a "signing"
I got a star analyser this week, and duly captured my first spectra last night. Very exciting. Struggling a bit with the learning curve of Visual Spec though.
Ken, your books are excellent, just what I needed. My problem is ... you can't sign my iPad ;-)
Jon,
Welcome to the Rainbow side.....
I'm sure with a bit of practise that you'll come to terms with VSpec....
If you get stuck or need more help just drop me an email - only to glad to help.
Onwards and Upwards.
Thanks, Ken. Much appreciated. As you said, VSpec is starting to fall into place. Managed to preprocess, load, calibrate some spectra (Altair and Vega), and correct for camera response. All good.
My main issue now I think is focusing. The lines are more like curves, which I'm assuming is focus.
Here's the raw image of Altair. See what I mean about the curved lines?
I say "raw" but I should be precise. I opened the Canon CR2 file in photosgop via an importer, and saved it as a jpeg. Lord knows what's happening to the levels. I can do the same in MaximDL but then I have to manually adjust the screen stretch to bring out the detail.
I suspect that if it's not focus, the issue may be over-exposure. This is an ISO200 1 sec exposure. And/or the fact that the spectrum was at a 45 degree angle to the CCD, off centre, and taken with a focal reducer.
Thanks Ken. What should I be looking for in focusing the spectrum? Obviously the height as tight as possible. But are there any clues in the spectrum itself? I certainly couldn't see any lines wen I was focusing; should I be able to?
How's your collimation on that scope? I can see what you mean. The lines are there but instead of being a straight line across the spectrum, they appear curved, and the zero order image looks a bit of a mess. I'm thinking both are perhaps collimation related.
Assuming the collimation is right, you should be able to focus the zero order image to a nice round star, then adjust focus on the spectrum for minimum width. You should see the lines become more definite as well in the process.
It may take a bit of practice to get the focus right because there will be a difference in focus between the red and blue ends, so it's a compromise unless you are only interested in a specific line (say Ha) but that's probably unusual for a SA.
FYI Not all stars have lines that are visually distinct in an SA, but you should nearly always be able to pick the Telluric O2 band in the deep red end once you know what you're looking for. Often you can't pick the lines until its binned in VSpec.
Al.
Last edited by sheeny; 04-08-2013 at 08:41 AM.
Reason: Afterthought
Jon,
With a A type star you should be able to "lock in" to the Balmer series..
As you get closer to focus these will become more defined...
If you can adjust the distance from the grating to the camera sensor to achieve close to 8-10 A/ pixel dispersion - this would give a width of around a 1000 pixel from the zero order to the end of the red spectrum - an easy fit across the 2500 (approx) pixel width of the DSLR.
Try to set the grating to give a horizontal image.
Your initial exposure looks OK...
Al - I *think* the collimation is OK, as I'm able to get nice round stars for regular astrophotography. But I'll check it.
Ken, I've got a DSLR that's 5184x3456 pixels. The grating is maybe 7-8 cm from the sensor. I'm using s 10" SCT with a 2540mm focal length, but a focal reducer bringing that down to 1600mm ... the upshot is that I'm getting about 1,400 pixels from zero order to the end of the red; 5.2 A/pixel.
I've been able to get *something* out of the spectra images I took on Friday by eliminating the curves by taking a 2-3 pixel horizontal slice out of the spectrum, spreading that, and feeding it through VSpec. Attached is my profile of Fomalhaut, compared with the AIIIV library spectrum. So I'm getting *somewhere*
Thanks for the tips, guys. I'll look at focus, collimation, and gettign horizontal to the sensor next clear night.
Jon,
With the grating in the converging beam the distance to the sensor is a critical one. Sounds like your DSLR has pretty small pixels (and you can't binned them...) Reduce the distance to around 50mm to get 8-10A/pixel.
Anything with a higher dispersion is just "empty magnification" - no more detail or resultion.....
The faster the f ratio the smaller the star image ( down to f5 ) but at the expense of added chromatic aberrations....
I think you're getting there - just more practise.....
Definitely getting there. Getting the spectrum horizontal and centred on the chip helped hugely, as did being meticulous with focus.
A quick question - are gratings bidirectional? That is, is there a "front" and a "back" to the optics, or doesn't it matter? If it doesn't meter, the I've found an easy way to mount the grating closer to the chip.
This spectral stuff is quite addictive. It's wonderful to be able to peer *into* the stars, even at the low resolution of the SA. But even after a week I'm thinking about an Alpy 600, and ogling the LhiresIII
Jon,
The grating can be mounted either way....
Softly softly....
Get some practise with the SA100, then the pre-processing and then on to the calibration and instrument correction(s).
The grating is normally good for "spectroscopy 101" and then after a few months consider your "upgrade" options.
I think Terry would probably say the LISA is a good option.....I think the new Spectra-L200.....
Would have liked to support BINTEL.
