Skip to main content

Spectra calibrations?

27 posts / 0 new
Last post
stellakafka
stellakafka's picture
Spectra calibrations?

Colleagues,

 

I just got back from the BRITE collaboration meeting. BRITE is a group of toaster-size satellites that are dedicated in observing the brightest stars in the night sky for all kinds of interesting research! One of the reasons why I went to that conference, was to advocate for the AAVSO to participate in the relevant science by providing ground-based support to the satellite data. It is a pleasure to report that all participants were excited by the help our observers can provide, and they are very eager to start working with us! And guess what they are also interested in: spectra!!!

 

You know that the AAVSO has always been interested in archiving spectroscopic observations alongside light curves. This project is in its planning stage right now, and will be implemented soon. The biggest challenge is calibrations – for data to be scientifically useable, we need good wavelength (and sometimes flux) calibrations of the extracted 1D spectra. I know that some of you are using gratings and some have access to spectrographs, so I would like to have a sense of the means that people use for wavelength calibration. How do you calibrate your data? Also, what software do you use to extract 1D spectra?

 

Thank you in advance for your help with this project. Archiving spectra will be an exciting new chapter in the services the AAVSO’s provides for the scientific community, and I am convinced that it will lead to great new science. We'll keep you posted on that!

Best wishes – clear skies,

Stella.

Ken4optics
Stella, I'm pleased to hear

Stella,

I'm pleased to hear of a renewed interest in spectroscopy.

The topic of spectral processing and wavelength calibration has been discussed previously. Depending on the choice of grating/ spectrograph there are a few options available.

I would suggest freeware software like John's BASS Project

https://uk.groups.yahoo.com/neo/groups/astrobodger/info

or Valerie's VSpec

http://www.astrosurf.com/vdesnoux/download.html

 

andyjwilson
andyjwilson's picture
Hi Stella,I have an LHIRES

Hi Stella,

I have an LHIRES III and I use the internal neon lamp for basic calibration. I've recently started to use the telluric lines around H-Alpha when I need a more accurate wavelength calibration in high resolution mode. I occassionally use the hydrogen balmer lines to calibrate low resolution spectra where they are prominent, understanding of course that will calibrate to the reference frame of the star rather than the Earth.

I usually use ISIS to process my spectra. I then use BASS to present the spectra and for post-processing like redoing the wavelength calibration with telluric lines.

I'm starting to get involved in spectroscopy with the BAA, where I'm the VSS Database Secretary. We have a spectroscopy workshop next weekend which will be useful for gauging the level of interest within the UK. I've been actively carrying out spectroscopy for about a year, and the level of interest at least feels like it is growing. These are certainly very exciting times for spectroscopy!

Best wishes,

Andy

andyjwilson
andyjwilson's picture
Hi Stella,My first post from

Hi Stella,

My first post from yesterday never appeared to make it, so hopefully I won't end up with two posts.

I've been doing spectroscopy for about a year, currently using an LHIRES III. I use the inbuilt neon unit for basic wavelength calibration. I've recently started to use the telluric lines around H-Alpha when I need a more accurate high resolution calibration. I've occassionally used the hydrogen balmer series to refine wavelength calibration at low resolution, though of course this will be to the reference frame of the star rather than the Earth.

I use ISIS for the main processing as it is very good at bulk processing. For presentation and if I need to refine the calibration then I often use BASS as it is very user friendly and has some rather useful features. When applying a polynomial fit to a group of lines it not only tells me the RMS but the deviation of each line from the polynomial, which makes it easy to spot any outliers.

I've been getting involved with spectroscopy in the BAA, where I'm the VSS Database Secretary. We have a spectroscopy workshop next weekend. This will be very useful to help us judge the level of interest in spectroscopy in the UK. I would not be at all surprised if we have some related conversations about how we record our spectra. Certainly exciting times for spectroscopy!

Best wishes,

Andy

rmu
rmu's picture
I am using two spectrographs

Stella, these are very good news. From one year and a half I've just got introduced into spectroscopy. I've never used an star analyzer; instead of this, I operate two spectrograph models:

-Alpy600, low resolution, R~500, grism spectrograph. Very appropiate to obtain whole optical spectra for objects as dim as mag. 11 in a, 8' telescope.

-L-200. Littrow design spectrograph. Fitted with a 600 lines/mm reflective grating, it can achieve R~4000, but operation is restricted to a narrower spectral magnitude 7 and brighter in an 8' telescope.

In Alpy, wavelength calibration is performed from Ne-Ar reference spectra and H Balmer lines from an A type star. In L-200, using Ne-Ar emission lines is enough. My tool for spectra processing, calibration and 1D spectra extraction is ISIS.

 

FJQ
FJQ's picture
Spectrographs

Nice comments Miguel!I'm working on getting data from my L-200 with a 600mm/ln grating. Its amazing how much moresensitive the Alpy600 is compared to the L-200 with its standard grating. I need to get a better grating orAlpy600 to get into the realm of variables I'm doingphotometry on (11th-14th V magnitude). with a 33
and 43 Cm. apeture scopesJames

andyjwilson
andyjwilson's picture
L200 and Alpy 600

Hi James,

Have you thought of getting a lower resolution grating for the L200? The 150 ln/mm grating gives a resolution R value of 470 which is similar to Alpy and would enable you to go fainter as the spectrum would have a lower dispersion.

Best wishes,

Andy

FJQ
FJQ's picture
L200 and Alpy 600

Andy, I ordered the L-200 delux package that was supposed to come with 3 gratings and a powered neon calibration source, but the company has only provided me with the single (600 mm/ln) grating with the spectroscope without a working calibration source. I orderd this back in April. Hope to get the 150 and 300 mm/ln grating soon but I'm not holding my breath. If I was doing this over again I would have bought either the Alpy600 or Lhires III frrom France; although the cost of the Lhires III is almost 2x that of the L-200.

James

andyjwilson
andyjwilson's picture
L200 and Alpy 600

Hi James,

That explains it and is a shame. I hope the other items arrive soon.

Cheers,

Andy

rmu
rmu's picture
L200 deluxe package

James, I'm waiting too for the rest of the items of L200 deluxe package. As you, I've received first my L200 unit and a 600L/mm grating plus a Ne calibration lamp.

Independently of using the attached lamp, it is not difficult to attach the Alpy calibration module by Shelyak, using a T2 male thread. This calibration module works fine for L200 too!, it provides me of calibration Ne-Ar light, and tungstene light for flats, and it can be operated remotely with some kind of relay operating interface.

amallama
Spectral calibration

Hi Stella,

I have used both BASS and RSpec successfully. RSpec seems to work better when the 32-bit processing option is selected.

 Best regards,

Tony Mallama

 

BHU
BHU's picture
ALPY spectrograph and ISIS reduction software

     Hi, Stella:

     Great news that AAVSO is working on the challenge of archiving spectra.  I have a Shelyak ALPY-600, modest resolution (R~ 500) that is great for giving context to higher-resoution spectra.  The wavelength coverage runs from about 3750 Angstroms to at least 7500 Angstroms (with my C-11 and ATIK 314L+ CCD).

     I use the "ISIS" software to process the images into spectrum profiles.  Wavelength calibration in ISIS is based on a mixed approach that uses a Ne-Ar lamp to calibrate the middle and red end of the spectrum, and a nearby A-type star's Balmer lines to calibrate the blue end.  This gives a total of 12 lines (ranging from 3835.39 to 7383.95 A) and a 4th order polynomical fit of wavelength vs. pixel position.  Typical formal error between the polynomial and the actual line wavelengths is less than 0.5 Angstrom.

     ISIS includes an option to correct the spectrum profile for the instrumental+atmosphreic wavelength response (based on the measured intensity vs wavelelngth of a nearby early-type star).  This provides a pretty good display of the continuum shape.

     Like VSpec, ISIS is free.  It's a bit of a challenge to learn, but it has the advantage of containing quite a few special features that are tailored for the ALPY (and, I think, other Shelyak spectrographs).  For example, it handles seamlessly the tilted and curved lines in ALPY's calibration lamp images ("tilt", "slant" and "smile" parameters).

     I sympathize with the complexity of defining criteria for acceptable spectra.  The BeSS data base requires several parameters that I wouldn't have thought of, but which are certainly important to ensuring the usefulness of the archived spectra.

Bob Buchheim

Ken4optics
James, The difference between

James,

The difference between the Littrow spectrographs (Spectra-L200, LhiresIII) with a 600 l/mm grating and the ALPY is the collimating lens used.

The Littrow has a 200mm lens, a resolution of 2.8A and R= 2317 about x5 the ALPY.

 

FJQ
FJQ's picture
James, The difference between

Thanks for the clariffication Ken!

James

Ed Wiley_WEY
Ed Wiley_WEY's picture
Spectral Calibrations

As I understand it from using my ALPY 600 and SA200 and talking with some active spectroscopists; I use the following method with ISIS. Get your darks, bas and flats sometime during thee run (I use the calibration module). Pick the target. Pick the "comp star," close to the target for the instrument response. Slew to the comp star, collect your neon calibration image, and then image the comp star. Next, slew to the target, again do a neon calibration image and then image the target. In ISIS you process the comp star and collect its instrument response. Then you process the target using the instrument response curve derived from the comp star. Having a close comp star allows for a "standard reduction" of the target, as I understand it. If everyone uses the same comp star then the results are, apparently, very comparable as shown by the group effort on the Nova Del 2013 campaign. Of course, if you have another target that night you have to pick another comp star that is close to that target, so each target requires its own comp star.

I took a Nova Sgr 2015b spectrum with an SA200. Basically, the resolution was not high enough for the data to do into the French database. (My Aply one was fine, plenty of resolution.) It also used a comp star (Zeta Sgr), but wavelength calibration was via the Balmer lines. Two problems: (1) as mentioned, not enough resolution and (2) the calibration was a bit off on both the blue and red ends due to lack of enough Balmer lines for a really critical calibration. This is shown in the attachment: My SA200 is in purple, Jim Edling's (Lisa spectrograph) in blue.

Ed

Sorry, don't know where the attachment ended up, but its not available.

 

 

 

 

TCB168
TCB168's picture
Spectral processing

Stella

I use a LISA spectrograph and process with ISIS. It is an almost automatic process now that I can do whilst still taking more images. My data has mostly been added to the BeSS database.

Calibration is with a neon/Argon/Xenon "Filly" light or the internal neon light combined with Balmer lines from an A calibration star.

This also provides an instrument response curve to correct the "shape" of the spectrum. This however is difficult to maintain consistency over many nights and some variation occurs. The wavelength calibration is very consistent from night to night.

I have explained how it is done on this page.

http://users.northnet.com.au/~bohlsen/LISA_instructions/Start.htm

Terry

 

andyjwilson
andyjwilson's picture
Previous amateur involvement in the BRITE

I have a message from Robin Leadbeater who is unable to login to the AAVSO website at this time.

Cheers Andy
 
I see there is some interest in taking spectra for the BRITE project. You might be interested in passing on some information about previous amateur involvement in the BRITE project last year through the Convento group. (P.I. Noel Richardson University of Montreal). More information here on the Convento Group forum
http://spektroskopieforum.vdsastro.de/viewtopic.php?t=4163
In fact I am using it as an example of high precision radial velocity measurement at the NLO workshop.
My advice to anyone considering submitting spectra would be to find out from whoever is asking for the spectra exactly what is required in terms of resolution, calibration precision,(both in wavelength and flux), frequency of observation, data reduction etc etc before embarking on taking spectra. Without this it is very easy to end up taking data which is of no value.
Cheers
Robin

DJD
Spectra

My experience in capturing spectra is as follows: I have been using the software RSpec developed by Tom Fields. So far I have used a Pentax k-x DSLR camera with a 50-200 mm Pentax lens to capture my spectra. By experimenting I have found that a Focal Length of 125 mm gives me the best results which yields a specrum with a definition of 2.20 Angstoms per pixel. I use a Rainbow Optics Star spectroscope (200 lines/mm). The spectroscope is attached to my camera using a lens cap. A hole is made in the lens cap that will accept the spectroscope.

I  then set up my camera on a tripod and orientate  my view finder so that the star and it's spectrum are horizontal in the view finder. The spectrum must then be aligned  (pointing to ) Polaris. I use 40 second exosures and let the Earth's rotation widen the spectrum.  The spectrum is then loaded into iPhoto on my iMac computer where I can enhance the spectrum . DropBox (it's free) is then used to transfer the image to my Windows laptop where RSpec is located. I usually use a one point  (the actual star) for calibration and then enter my 2.20 Angstrom per pixel value and there is my calibrated spectrum.

I have been doing this for some years and I check my results against  Richard Walker's " Spectroscopic Atlas for Amateur Astronomers" which is available online.

I feel that my results agree with those in the Atlas.I have been using this method as our winters here in Thunder Bay, Ontario, Canada are quite cold and working with my 10 inch Dobsonian  or my 5 inch refractor is quite challenging at times under these conditions. So on cold nights I simply set up my camera and spectroscope indoors after making a viewing plan and then go outside for about 15 minutes and capture my spectra.The disadvantage is that I can only capture the brighter stars using this method but it is quite rewarding. I plan to adapt my methods to my refractor telescope whenever the temperatures and skys will allow me to do so. Keep in mind that at 49 degrees North Latitude we do not get good dark skys in the warm summer . To compensate for this I use SSON Remote Telescopes to do my variable star work when the weather or light pollution becomes a hindrance. Respectfully  DJ Dowhos   DDJ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SBL
SBL's picture
  Hi Stella, the amateur

 

Hi Stella,

the amateur astronomy community of Europe has spectroscopy experience since more than then years. They have at their disposal spectrographs that can be controlled remotely and also free software to process spectra with scientific value. More and more there are Pro-Am collaboration projects. There are also extensive database for instance of Be stars.

SOFTWARE: all free

Buils ISIS software: a complete program for pre-processing and processing spectra recorded with Shelyak spectrographs but also with L200, homemade spectrographs etc.. It works semi-automatic and has excellent tutorials. A great majority of amateurs in Europe Australia and New Zealand use ISIS

 

Vspec:  easy to use, excellent documentation. Used in combination with Specaudace or ISIS for instance after calibration.

 

SpecAudace by B.Mauclaire:  very easy, fully automatic pre-processing and calibration.

 

For the LISA spectrograph in Las Crucas we (i.e. my astro-colleague and friend Dr. Valère Van Ruymbeke) and myself intend to use both ISIS and SpecAudace +Vspec

 

For the eShel :

The best choice is recording with Audela and processing with the dedicated software for Eshel in Audela The process is fully automatic.

 

Another good option: the Eshel module of Isis.

 

For processing with ISIS some previous rules must be observed:

 

  1. Isis does only accept files with extension.fit or.fits NOT .fts
  2. The number in the name of the recorded object comes just before the extension:
  3. Hr4963-002flat.fts; must be FLAT_HR4963-001.fit
  4. Isis don’t work with files named XXXYYY-001.fit but has a tool to convert the files in XXXYYY-1.fit
  5. For eShel there are strict rules for naming the files
  6.  

xxxx_tung-1      xxx=name object  ex delsco_tung-1

  •  

            Led     xxx_led-1       delsco_led-1

                       xxx_led-2

                       xxx_led-2

            Dark    dark_xxx_yyy-1    xxx=acquisition time in seconds yyy: temperature

                       Dark_300_20-1

 

            It’s a good idea to apply these rules also for the LISA

best regards and clear skies!

Bart

 

 

 

Ken4optics
Bart, For completeness you

Bart,

For completeness you should also have mentioned the freeware processing software BASS Project.

The latest V1.8 upgrade of BASS Project will handle all the processing needs for gratings and Littrow spectrographs.

https://uk.groups.yahoo.com/neo/groups/astrobodger/info

Ken

 

 

daj
daj's picture
LISA

Hi Stella,

I'm just getting started with a LISA that arrived yesterday.  Doubt I'd be of much use initially but it's great to see the experience represented in this forum and hopefully I can get up to speed.

Thank you.

 

 

 

gka
gka's picture
AAVSO Spectroscopy

I checked the spectroscopy forum after a hiatus, and I saw this interesting thread. It would appear that AAVSO is now going for spectroscopy results that can be very useful in research. I have been doing photometry and spectroscopy concurrently using two scopes, and this simultaneous spectroscopy/photometry appears to be a goal of AAVSO. Having a repository at AAVSO for both photometry and spectroscopy will be a very welcome addition.

 

I use both the LHires and the Alpy600. I use the LHires with a 2400 grating for high res spectroscopy and the Alpy for low res.  I have been using ISIS for data reduction, and I use the internal neon lamp on each spectroscope for initial calibration. In addition, I tweak the calibration using the telluric line algorithm in ISIS when processing high res specs.  Since most of my spectroscopy is with Be stars, I use the calibration checker in Vspec to be certain my calibration accuracy <.1A.

 

When using the Alpy, I use pretty much an identical procedure to that of Ed Wiley (see his message for October 10). I do have the calibration and the guide modules on my Alpy as these devices do facilitate the acquisition procedure even though they are not absolutely necessary.

 

I will say that the learning curve for spectroscopy is a bit more involved than with photometry. One must be patient in learning the acquisition procedure and then the spectrum processing routine. But once you get the hang of it, everything falls into place and the results are very rewarding.  . I have found ISIS to be an excellent processing program. ISIS does have a learning curve, but there is an excellent procedure manual available. Also, there are forums out there where spectroscopists are very willing to lend a hand to help level out that learning curve And a big bonus is that it is free. Vspec is also free, but I do find ISIS easier to master. I do use both programs depending on what information I want from the spectrum.

 

I have to say that I am so glad that AAVSO is moving into the spectroscopic realm. The SA was a great way to introduce interested members to spectroscopy. But to take spectra that are most useful for research, a slit spectrograph is necessary. It is true that slit spectrographs do not come cheap, but for those truly interested in quality spectra, they are well worth the price. The spectra I take are those of Be stars and symbiotic stars. There are two repositories in France to which I send my spectra. But there is a need for a repository where spectra of CVs and other variable star types (like HMXB) can be sent. This is a role that AAVSO can fulfil.    

 

Keith Graham

Robin Leadbeater
BAA VSS Spectroscopy Workshop

Material presented at the recent BAA VSS Spectroscopy workshop is now on line

http://www.britastro.org/vss/

and  includes some run throughs of  acquisition and data reduction (mainly using ISIS) for a range of instruments and applications including slitless, low resolution and high resolution slit spectrographs.

Robin

Ernst Pollmann
Ernst Pollmann's picture
Calibration accuracy

Hi Stella,
one thing seems to be important in the context of wavelength calibration: a reference of the calibration accuracy of stored spectra.

Ernst Pollmann

 

Ken4optics
Stella, Where to from

Stella,

Where to from here?????

Are there any formal spectroscopy obsrvations being conducted, if so, where do they get recorded at the moment??

 

stellakafka
stellakafka's picture
Spectra

Hello all,

Thank you for your responses and suggestions. From the discussions in this group, I see that BASS, RSpec and ISIS are popular to get 1d spectra for all types of gratings/spectrographs and they perform reliable calibrations (if a relevant source is available). I haven’t used any of them (although I’ve seen RSpec in action); I assume they all work with fits files? Do they allow editing fits headers?

From here, we are thinking of what requirements of an AAVSO spectroscopy database would be. Spectra for the BRITE targets could be marked differently in the database, and everyone would have access to the data (similar with the light curves that are on the AID). Considering that even the most “regular” variable stars exhibit unusual behaviors when observed for a long time, a long-term spectroscopic database would serve both short and long-term projects. Ideally, it would be nice to find a way to connect spectra with light curves, but maybe this is something to consider after getting a spectral database going. Thoughts?

Best wishes – clear skies,

Stella.

PS: we do have one request for photometry and spectra right now, which is listed in the latest special notice:

https://www.aavso.org/aavso-special-notice-409

The target is between 11 and 12.6 in visual mag. For instructions on how to submit spectroscopic data, please contact Elizabeth Waagen (eowaagen@aavso.org).

 

 

Ken4optics
Stella, I think you will find

Stella,

I think you will find BASS Project, VSpec and ISIS all work with .fits files and allow editting of fits headers.

 

Log in to post comments
AAVSO 49 Bay State Rd. Cambridge, MA 02138 aavso@aavso.org 617-354-0484