Sat, 02/21/2015 - 22:49
It's been very quiet over the past few weeks...
I assume by now everyone has received their gratings.
What's the feedback? To difficult to incorporate into the filter wheel and photometic programs?
What other issues have come up with new users?
Hello Ken
All is well with the effort. I know personally, we have had 80 inches of snow in the past month, and that has impacted my observing as well as the operation at HQ.
WGR
Gary
I've been working with
All,
I've been working with my SA200 for a few weeks now and feel I am beginning to get the hang of it a bit more. I've attached a few jpegs to show recent results. I missed out on the original group purchase, but Arne kept me in mind for any available extras, and I thank him for his efforts in providing me one. Being able to produce spectra with my modest equipment has opened a new world to me, My thanks to Arne and Tom Field for their guidance and support.
I mounted my SA200 in my SBIG CFW8 attached to my ST-8XME and found I needed to add some space between the filter wheel and the camera to get a resolution of 13.6 Angstroms/pixel. Because of the added length, I needed to remove my f/6.3 focal reducer, so I'm operating at my telescope's (30 cm SCT) native f/10 focal ratio, otherwise I would crash the camera into the mount at about 55 deg declination. Astrometrica is fine with f/10 for my asteroid work and VPhot is fine with f/10 for my long period variable work, so good enough for me then. A couple of thoughts I've had:
1. Regarding the alignment of the SA200 in the filter wheel. At first I thought it would be best to orient the SA200 so the spectra are dispersed in declination, since any periodic error in my mount (I'm not autoguiding) would not broaden the absorption features as may be the case for dispersion in RA.
But, I thought more about about it (over a homebrew dark chocolate porter and a Cohiba cigar) and decided to orient the SA200 so the spectra would disperse in RA. The reason is that in order to get a good exposure on the 1st order spectrum (about 45K ADU on my camera), I would necessarily over-expose the the 1st order star. On my camera, that would result in bleeding in the vertical, or declination, pixels. But that's okay, since it would result in the magnitude of that part of the spectrum being magnified but the position of the 1st order star would be retained (so okay for calibration). If the SA200 was oriented in the perpendicular direction, the bleeding may spill into my 1st order spectrum, which could corrupt my calibration and/or analysis.
2. Sum stack or average stack? I'm limited to 60 sec unguided exposures, so I will sum stack if I need to to get a good SNR and peak 1st order spectrum in the range 45K ADU (limit of linearity of my camera), otherwise if I can get the same result by average stack or no stacking at all, then I won't. I just take a series of 60 sec exposures and then decide afterward.
3. I found that a linear calibration in RSpec didn't line up the all Hydrogen Balemer lines and I needed to do a 2nd order fit. Okay, fine, but what then when I take a spectrum of, say, BL Ori? I haven't figured out how to apply the 2nd order calibration to other spectra (perhaps another homebrew and cigar would resolve that issue?)
In summary, I've found the SA200 to work well for me. As for the RSpec software, I found it fairly easy to learn and use, but I'll get with Tom Field offline to recommend some minor changes to his software to better align with operational procedure. He's a Premier Speaker at Rocky Mountain Star Stare (RMSS) this June in southern Colorado, so I'll talk with him there.
So, that is my report. At this point I can easily go from making photometric measurements in the V band to spectral imaging with the SA200 with very little effort and have been been getting, I think, good results with the SA200. I look forward to continuing my education in spectral imaging and analysis and discussion with members of this group to that end.
Scott Donnell
Vice President,
Colorado Springs Astronomical Society
csastro.org
rmss.org
Attachments:
1.Alhena_Labeled.jpg. Calibration on an A0iv star. H Balber absorption lines prominent
2. IC 2165.jpg - Planetary nebula.Just wanted to see how well the [OIII] 4959/5007 forbidden line would show up
3. BL Ori - A carbon star, and I like carbon stars. How did I do with line IDs?
.
To: Ken,
I purchased my SA200 at the Summer AASVO meeting held in Ontario, CA. I did a lot of fun stuff with it for 2 months (spectra of Vega, 5-10 other stars, redshift of 3C 273) then kinda languished. I recently took a spectrum of CN Ori when it was in outburst on 15Feb15. I had to re-watch all of Robin's videos again to reduce this to a calibrated spectrum; http://www.astroimage.info/images/CN-Ori%28SA200-Spectra%29.jpg
I've been agonizing on whether to further my spectroscopy by purchasing either an Alpy600 or the higher resolution Lhires III. Even though the Alpy600 has only 1/4th the wavelength resolution that the Lhires III has, the Alpy600 seems more versitile when acquiring SN or dim object spectra.
Also, the SA100/200 can be made more powerful using C. Buil's technique seen here: http://www.astrosurf.com/buil/staranalyser2/evaluation_en.htm
James Foster
How did you sum
Hey scott
How did you sum stack your individual images and then use rspec? My understanding from tom fields is that rspec only averages?
On this spectrum I took of 3C 273 last year I stacked 5x5-minute exposures in MaximDL, then imported the stacked/summed 25-minute fits image into RSpec. Here is a calibrated spectra of 3C 273: http://www.astroimage.info/images/3C-273%28Mt.Pinos%293.jpg
James
Mr. Theophilusmonk,
Sorry for the delay in responding. I sum stacked not in RSpec but rather in AIP4Win. For the alignment stars I chose two zero order images. Seemed to work well.
Scott
Hi Scott,
Because resolution is dependent on star image size, you may find that using a slightly smaller spacing at f6.3 gives better resolution than f10 at a larger spacing. The calculator can be used to help you decide. (Check the calculated star image size it gives in pixels and multiply this by the A/pixel disperion figure. This will give you an idea of the maximum potential spectrum resolution. (Choose the configuration with the lower the figure) The smaller star image also means you dont need as long a spectrum to get the same resolution so you can go fainter. If you are not sure, can you post screenshots of the calculator with the two configurations and I can advise.
1. If you have any RA drift and the dispersion is in the RA direction then the spectrum will be blurred and resolution will be reduced. With cameras without ABG, bleeding is vertical (ie along a column) so is not a problem if the grating is orientated with the dispersion direction along the row). A solution could therefore be to orientate camera so the RA direction is in the column direction. Alternatively, you could expose so the zero order is no over exposed and combine several exposures. (There will be a marginal worsening of SNR using this method because of the extra read noise doses)
2. For best snr, fewer longer exposures is marginally better but for slitless spectrographs spectral resolution depends highly on the star image size so for bright targets, aligning and stacking short exposures can help resolution. (aligning the sharpest exposures is the key here - like in planetary imaging)
3. provided you have included the zero order as a calibration point, you should be able to use a 2nd order (or higher) calibration in the same way as a linear fit using the zero order in the target spectrum as a reference point to define the constant in the equation for the target you are transfering your calbration to. (This works in ISIS for example and I am sure it will work in RSpec too though I dont know the exact procedure)
Robin
No more comments from first time users???
I'd be very surprised if the grating worked first time for everyone!
Typically what are the gratings be used for???
Hi Ken,
For my ST-9 the gratings just didn't have the spacing in the filter wheel for satisfactory dispersion for my C-8. I've looked at different ways to mount either grating on a nose-piece, but converting between the two requires time that I have little of these days, and I would not like to lose time from the EB's I enjoy imaging.
The weather hasn't dumped the snow on us as it has on New England, but in Milwaukee for the month of February there were 22 days with light snow, 6 days with snow, 1 day with rain, 1 with freezing rain. (Yes I know that adds up to more than 28. Some days had a mix). The bitter cold when we were clear siezed the drive gears lubricated with what I am sure is not rated below zero degrees F. In normal winters I usually sit out two or three nights a winter because of this, but this winter has not been normal.
I AM intrigued by what I could get even at a 40 angstroms/pixel and would like to do better. I am considering purchasing a second telescope - or perhaps a new mount using an old optical setup - or perhaps I will mount a grating on a DSLR objective.
I really cannot wait to retire!
Neil Simmons
Here is a recent calibrated spectrum of GK per in outburst made from 5 combined 3-minute exposures with an SA200 on with an ST-10XME and 33Cm apeture classical cassegrain; no lens element reflector.
http://www.astroimage.info/images/GK%20Per%285x180sec%2913Mar15.jpg
There was a faint star superimposed on the spectral line I could not align-out so I deleted it from the spectrum where it says "NO DATA." I can't say much about this spectrum other than the obvious, very strong-wide absorption feature centered at 7200A and a identified telluric line for H2O at about 7400A. I could only faintly identify any balmer H lines. Since GK Per is fairly low in the western horizon, dispersion effects probably caused some error from the A-star standard (Lambda Gem) I used as a reference 2-weeks before, near zenith.
James
Hi Neil,
Sorry to hear you are not getting so good results with your setup. I ran it through the calculator and it confirms what I suspected that with the relatively large pixel 20um size of the CCD in the ST9 and a spacing of 25mm with the SA200 mounted in the filter wheel, the resolution is limited to 80A which is well below what the Star analyser is capable of. To get a good result with this setup we would need to increase the spacing somehow to around 50mm. You could perhaps do this by adding a spacer between the camera and filter wheel, though I understand this might not be straightforward if it is close coupled to the camera.
A potentially better alternative could be to add a second filter wheel (or filter drawer) in front of the existing wheel and mount the Star Analyser in this. For photometry this second wheel would be set to a blank postion and for spectroscopy the wheel with the photometric filters would be set to a blank position. (You could even perhaps add an order blocking filter in the photometry wheel if there is spare room for doing spectroscopy in the IR) The C8 should be able to cope with the additional back focus and the photometric setup would remain essentially unchanged. I have attached copies of the calculator printout for your existing and the proposed setup.
Cheers
Robin