So I bought a Star Analyser 100...

[PhotoGav]

Yes, I took the plunge and bought a Star Analyser 100. My first experiments have been with a SkyWatcher 200 PDS on an HEQ5 mount with a ZWO ASI 120MC-S. I removed the IR filter from in front of the chip on the camera, fitted the SA100 and pointed the scope at Vega and this is what I got:

Well, in reality, I got a slightly slanted image, but I have rotated and cropped it in Photoshop for the purposes of display.

I shot a .ser video and exported about twenty individual frames from the video. I took them into BASS and stacked them. I then rotated the resulting image to make it horizontal, calibrated with the zero order and what I hope is the hydrogen beta dip. I then guessed a bit and annotated the graph with elements that match the wavelengths of the dips. Here is the output:

How far off the mark is this?! It seems a bit random and very much guesswork with regards to matching dips to elements.

Anyway, I'm delighted to have got something that looks like a spectrum.

My next step will be to use a mono camera and to get the SA100 aligned so that the spectrum is recorded in the horizontal position. I will then see what BASS makes of that.

Any advice or thoughts on what I have so far would be greatly appreciated.

My ultimate goal is to estimate spectral class of a range of stars. My method will be to compare with reference spectra of known star types and use the 'best fit'. Is that a valid approach?

Thanks,

Gav.

[Craney]

This is from the internet.   In fact the company that manufactures the  SA100.

https://www.rspec-astro.com/star-analyser/

I don't know if you have used the same analysis software.   This is real time data as opposed to a stack.

I really do not know enough about the subject area to make any helpful comments, apart from well done.

 At least it is recognisable as a spectrum.  Although not spectroscopy, my first 'normal' astrophoto looked like a Jackson Pollack !!!

 

[PhotoGav]

Thank you @Craney - yes, I've spent some time on the rspec-astro website, watching videos and I downloaded the trial version of the software. Gave it a try as well as the BASS software. I feel like I am just at the very foothills of the spectral analysis learning curve!

[andrew s]

Looks like  a very good first go. I would advise rotating the grating to align the spectrum with the camera rows as software rotation adds artifacts and reduces resolution.

You look to have got good focus which is often difficult.

The calibration looks fine. The O2 is teluric in origin I.e from or atmosphere. The Fe  and Mg may not be real. I would need to check.

Great start.

Regards Andrew 

[Merlin66]

Craney,

RSpec don't manufacture the SA100, they only resell it with their software. Its produced by Paton Hawksley here in the UK.

Gav, good start..needs a bit more focus on the spectrum, the "hump" at 5900 A is due to the Bayer matrix in the colour camera.

 

[PhotoGav]

Thank you Andrew. Yes, there do seem to be some artefacts introduced from the rotation. I will definitely be spending more time on the grating alignment before hitting the record button! I just need some clear sky time to have my next play...

[PhotoGav]

1 minute ago, Merlin66 said:

Gav, good start..needs a bit more focus on the spectrum, the "hump" at 5900 A is due to the Bayer matrix in the colour camera.

Thank you! My research let me to believe that the dip in the 5750 A area was due to the OSC camera - thank you for confirming and interesting that it is a hump not a dip! My mono camera will be given an outing next time - I just wanted some pretty colours on my first play!!

[robin_astro]

5 hours ago, Craney said:

This is from the internet.   In fact the company that manufactures the  SA100.

https://www.rspec-astro.com/star-analyser/

I don't know if you have used the same analysis software.   This is real time data as opposed to a stack.

I really do not know enough about the subject area to make any helpful comments, apart from well done.

 At least it is recognisable as a spectrum.  Although not spectroscopy, my first 'normal' astrophoto looked like a Jackson Pollack !!!

 

 

Note Tthe Star Analyser is not manufactured by this company. It is manufactured by Paton Hawksley in the UK.  It is sold in the US bundled with the RSpec software.  How do I know?  Well I developed it in 2005 🙂

[robin_astro]

Hi Gav,

A good start to identify the Balmer and Telluric lines (The only features which appear in the Vega spectrum).  You should be able to get the lines clearer though.  The spectrum looks flat topped so I suspect it is over exposed, very easy to do with such a bright star.  err on the underexposed side to stat with. Focusing is critical (use the RSpec real time display to tune the focus to maximise the depth of the H beta line, not the zero order focus)  also because the resolution depends on the size of the star image, the best sharpness of the spectrum is on nights with good seeing and targets at high altitude. 

Cheers

Robin

[robin_astro]

8 hours ago, PhotoGav said:

My ultimate goal is to estimate spectral class of a range of stars. My method will be to compare with reference spectra of known star types and use the 'best fit'. Is that a valid approach?

Yes, very approximately, within the limits I already outlined in this thread

Spectral classification depends on the lines not the overall shape so you have already made a start as the presence of only  Balmer lines signifies we have a hot star (likely class B or A , in fact A0v)

Cheers

Robin

[PhotoGav]

Excellent, thank you Robin. The journey has begun! I will play with exposure times and focus when I next get the chance. No doubt plenty more questions to come!

[Merlin66]

We're always here to help.

 

[robin_astro]

Gav,

Your mono camera will make life much easier as it will not have the confusing humps and bumps in the response that colour cameras have.  As well as Torsten Hansen's collection of colourised spectra I posted in the other thread, this report of mine will also give you an idea of typical Star Analyser spectra for a few stars of different spectral classes. Note that this particular setup was tuned for sensitivity rather resolution so the resolution is quite modest. A system optimised for resolution (by increasing the dispersion relative to the star image size) would show a bit more detail than this.

http://www.threehillsobservatory.co.uk/astro/SA100_miles_instrument_response.pdf

It is from this page where similar exercises using an ALPY spectrograph at approximately 5x higher resolution can be found giving an idea of the effect of resolution

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

 

Cheers

Robin

Edited November 6, 2019 by robin_astro
typo

[robin_astro]

19 hours ago, PhotoGav said:

I will definitely be spending more time on the grating alignment before hitting the record button!

A good tip to get the alignment right before you go to the telescope is to look at the camera sensor through the grating. if you get the light right  you can see a series of images of the sensor from the various spectrum orders. Just carefully align these so they lie in a straight line parallel to the sensor edge and your spectrum will be horizontal.  A laser pointer will give a line of red dots which can be similarly aligned.

Cheers

Robin

[PhotoGav]

Thank you Robin, good tips. Should save valuable sky time!

[Peje]

@robin_astro I've just fitted the SA100 into my system and grabbed a few images of Capella last night, that's as far as I've got unfortunately! A couple of queries / problems I have:
 

- I use an IDAS P2 LP filter, I presume this will create holes in my spectrum?

- The SA100 is running at a slope and I cannot rotate my camera without screwing up other alignment or hitting the filter-wheel off the shed. Is this going to be a major problem?

[Helen]

41 minutes ago, Peje said:

@robin_astro I've just fitted the SA100 into my system and grabbed a few images of Capella last night, that's as far as I've got unfortunately! A couple of queries / problems I have:
 

- I use an IDAS P2 LP filter, I presume this will create holes in my spectrum?

- The SA100 is running at a slope and I cannot rotate my camera without screwing up other alignment or hitting the filter-wheel off the shed. Is this going to be a major problem?

On the question of alignment, I added a little plumbers tape to the filter thread which allowed me to vary the rotation of the filter in the filter holder while still being secure.

Helen

[Peje]

22 minutes ago, Helen said:

On the question of alignment, I added a little plumbers tape to the filter thread which allowed me to vary the rotation of the filter in the filter holder while still being secure.

Helen

That's a good idea, I had been thinking of something like that. My main problem is that there is no spare height in my FW so if I unscrew the filter even slightly I will risk it scraping the inside of the casing.

[andrew s]

12 minutes ago, Peje said:

That's a good idea, I had been thinking of something like that. My main problem is that there is no spare height in my FW so if I unscrew the filter even slightly I will risk it scraping the inside of the casing.

In the past I have increased the spacing on filter wheels with thin spacers and once even 10mm. I found it quite stable. I used black masking tape to light seal the gap. I seems to recal the SA200 is thinner than the 100.

I needed to align a SW100 with a 200 in a filter wheel. I used a small laser to define a plane by projecting if throught the 200. Them selected the 100 which had glue on the threads and rotated it to match and then left it for the glue to dry. My wife has still not spotted the small marks on the wall paper marking the plane!

Regards Andrew 

PS these days you could 3d print a thin gasket. 

Edited January 7, 2020 by andrew s
PS added

[Peje]

I discussed using the SA200 low profile filter with Robin a few months back, he recommended using the SA100 in my system.

I'm currently trying to think if I could rotate my entire camera train by say 15deg and then move the camera by -15deg. My main constraint is that I need the camera sensor to be parallel to my scope axis to stop image rotation during long imaging sessions but also that if the FW get's too high it might clash with the roof of my roll off shed.

[andrew s]

2 minutes ago, Peje said:

I discussed using the SA200 low profile filter with Robin a few months back, he recommended using the SA100 in my system.

I'm currently trying to think if I could rotate my entire camera train by say 15deg and then move the camera by -15deg. My main constraint is that I need the camera sensor to be parallel to my scope axis to stop image rotation during long imaging sessions but also that if the FW get's too high it might clash with the roof of my roll off shed.

Life is full of annoying constraints. However, I don't see any way camara angle can impact field rotation.  Framing yes rotation no. I rotate my camera/spectrograph to avoid overlapping spectra with other stars. The orientation stays fixed all night with an eq mount but for the 180 deg turn on the meridian flip which I take out with the rotator.

Regards Andrew 

[Peje]

1 hour ago, andrew s said:

Life is full of annoying constraints. However, I don't see any way camara angle can impact field rotation.  Framing yes rotation no. I rotate my camera/spectrograph to avoid overlapping spectra with other stars. The orientation stays fixed all night with an eq mount but for the 180 deg turn on the meridian flip which I take out with the rotator.

Regards Andrew 

I can offer no explanation as to why but the further my camera is away from 0/180deg orientation on the OTA, the more rotation I seem to get over a nights image.

This is fairly anecdotal evidence as it was within the first couple of years of my AP so I was adjusting numerous things in unison.

[robin_astro]

ISTR this is a fast f4 telescope hence the recommendation for the SA100 provided it would fit. Having to rotate the spectrum to horizontal in software is not ideal, particularly for colour cameras but as you have a mono camera you can get away with this. (Most of my work with the Star Analyser has been done with the spectrum at odd angles!)  If you do see rotation artifacts make sure you get even coverage of pixels by taking several subs and allow some drift or dither a bit between them, then align and stack. 

Yes the LP filter will leave holes ! eg

http://www.threehillsobservatory.co.uk/astro/filter_spectra_3.htm

Cheers

Robin

[tooth_dr]

@Peje I have one of these, but cant remember if the glass can be easily removed.  If it can, just use some sandpaper, and take 1-2 mm of the height of it, then follow the advise above.

[Nigella Bryant]

I use the star analyser 100 with my 12inch f4 Newtonian. Not done much due to regime change (well, house move). Attached is my attempt at Vega with bass project software. I have found the software easy to use for a beginner.