Spectrum software anomaly?

[Gasman]

Hi Guys

Hoping someone could shed some light (forgive the pun!) on why I seem to be losing detail when opening fits or png files in spectro software?.

This is an Astroart5 unprocessed screen grab of Arcturus (used because it was bright!)

and this is the same raw file opened in BASS again unprocessed (does the same in Rspec)

A couple of strong absorption lines are clearly seen in the first image to the left but disappear in the second. The spectro software appears to be doing some stretching and losing detail, any ideas what to try please?.

cheers

Steve

[Merlin66]

Steve,

I think it may be that the image in AA5 has the visualisation (gain) setting higher - note the slider positions in the rh column.

I think you'll find that the normal binning (BASS etc) will pull out all the detail you have.

[Gasman]

Cheers Ken. The only thing I did in AA5 was an 0.8 degree rotation and then saved the AA5 image which I then opened it in BASS so you would think that the images would be the same before any processing is done?

best

Steve

[Merlin66]

Steve,

Nothing to do with the tilt correction (remember you also need to apply slant correction based on your reference lamp)

It, I think, comes down to how AA5 displays the image on the computer screen....play with the sliders at the rhs of the AA5 screen - you'll see what I mean.

This doesn't affect the RAW data, only how it displays on the screen.

[Gasman]

Hi again Ken

I`ve been out of the spectro loop quite a bit lately and just refreshing my memory reading your `Processing spectral images in BASS Project`.

Some  4 or 5 months ago I used a combined Neon/Argon bulb similar to that used in the Alpy 600 calibration module although it does require a 120 second exposure to get decent lines especially at the blue end where most of the Argon lines are. In your tut. you seem to use either a printout of Neon or printout of Argon lines, do you find just one or the other adequate for calibrating?. I always thought it best to get a broad image of lines but it would make life a bit easier just using say the Neon spectrum for calibration. I used this lamp at the time as I was mainly using ISIS spectro software which needed both sets of lines but I do like using John`s BASS software and trying out 1.7beta which is working well.

cheers

Steve

[Merlin66]

Steve,

It really depends on the target wavelengths and the resolution.

For Ha spectra then the neon is all you need.

I was observing down around Hb for the Nova last year and needed the FILLY lamp to give good calibration.

If you can get three or more reference lines across your spectrum you'll usually get a good calibration.

Hope this helps.

[robin_astro]

Hi Steve,

Yes with full range spectra at low resolution you ideally need some lines at the blue end to augment the neon lines to get a good calibration over the full wavelength range. (I prefer to only trust the calibration between lamp lines and avoid extrapolating beyond them wherever possible. You can see the magnitude of the errors at the blue end  just using neons here at the bottom of the page, using the LISA as an example.

http://www.astrosurf.com/buil/isis/guide_lisa/tuto_en.htm

In particular, extrapolations  if using a high order equation  (>2 or 3) can be very unreliable.

As you say the trick with ISIS  with the LISA  and ALPY for example is to combine Balmer and neon lines using the calibration assistant.  

I expect it would be possible to adapt the techniques for use with other programs and for other spectrographs, although it probably would not be as automated. (I am currently looking at the best way to calibrate my low resolution ALPY 200 where many of the Ne/Ar lamp lines normally used for the ALPY are unresolved.)

Robin

[Gasman]

Thanks Ken and Robin helps a lot!. So thinking about it it does make sense for an overall calibration spectrum to keep the Ne/Arg lamp. I have made an adaptor that fits over the end of my 120mm refractor which has a 3/8" hole drilled in the middle and the lamp is behind this hole which is maybe not the most efficient way as I say it needs at least a 120sec exposure to make out lines at the blue end but they are visible. I have plenty of these (starter  ) Ne/Arg bulbs and I did think about maybe having 3 or more bulbs fitted to the adaptor to increase the luminosity but then maybe you have the problem of the calibrated light on different axes from each bulb giving conflicting lines whereas the existing bulb is along the central optical axis??

regards

Steve

[robin_astro]

Hi Steve,

When I first got my ALPY (before the calibration module arrived) I used a starter Ne Ar lamp and injected it via a flip mirror mounted (the wrong way round) between the scope and spectrograph. (This was also useful for taking darks)  ISTR exposures were only a few seconds. If you use lamps ahead of the telescope, make sure they are arranged symmetrically about the line of the slit otherwise you will get small calibration offsets. Multiple sources will broaden the lines a bit but no more than the lines from the stars are. (The lines from a single lamp mounted this way are extremely sharp) 

Robin

[robin_astro]

Here is a photo of the ALPY setup with the flip mirror injection 

http://www.spectro-aras.com/forum/download/file.php?id=887

and the disposable camera powered NeAr lamp here

http://www.spectro-aras.com/forum/download/file.php?id=912

which is just tucked in behind the orange plug in the flip mirror port

I also tested it temporarily in the LHIRES (it had to be directly over the slit to be bright enough) and Shelyak have now adopted this arrangement in the LHIRES upgrade

http://www.spectro-aras.com/forum/viewtopic.php?f=8&t=606#p2380

Robin

[Gasman]

Hi Robin

Nice idea the wrong way round flip mirror, I have an unused one of those, same model as yours might give it a go!

cheers

Steve