Calibration ALPY600 with ISIS software

[Erik Bryssinck]

Hi all,

I have an ALPY600 spectroscope with guiding module, since a two weeks installed on my 0.4-m newton telescope. After some experimentation, I decided to purchased the calibration unit as well and installed a few days back. I took a calibration lamp spectrum and a flat and I try to apply it in the ISIS software

I come here on a problem because the calibration fails. evenso after adjusting the pixel size remains the deviation> 10 ang. I see mostly situated in the blue area (Ar). The pixel size of the SBIG ST10XME is 6.8 microns. Here the results from ISIS calibration of the calibrationlamp spectrum.

Fitting deviation (wavelength)
point #1 x = 676.291 lambda = 3934.747 dlambda = -45.697
point #2 x = 677.202 lambda = 3938.363 dlambda = 31.717
point #3 x = 714.093 lambda = 4083.896 dlambda = 17.854
point #4 x = 779.267 lambda = 4337.145 dlambda = 3.335
point #5 x = 920.475 lambda = 4872.490 dlambda = -11.150
point #6 x = 1187.031 lambda = 5851.034 dlambda = 1.456
point #7 x = 1162.755 lambda = 5763.166 dlambda = 1.254
point #8 x = 1212.562 lambda = 5943.202 dlambda = 1.628
point #9 x = 1302.365 lambda = 6265.392 dlambda = 1.098
point #10 x = 1370.083 lambda = 6506.113 dlambda = 0.417
point #11 x = 1554.571 lambda = 7148.943 dlambda = -1.903
point #12 x = 1624.401 lambda = 7385.452 dlambda = -1.502
point #13 x = 1726.673 lambda = 7722.268 dlambda = 1.492
--------------------------------------------------------------------------
RMS : 21.107494 (in angstroms)

I followed the tutorial C.Buil for ISIS and ALPY600.
as attachment, check the spectrum of the calibration lamp (50% resized) Ar_ne_calibratie.jpg

The attachment Ar_ne_calibratie.jpg is no longer available

below the flat image alpy ST10XME with 600 (50% resized)

50% resized image of flat with calibration unit of ALPY600 and ST10XME camera in 1x1 binning

flat.jpg (11.53 KiB) Viewed 5155 times

Anyone know how I can get the calibration right?

thanking in advance

Kind regards
Erik.

[Olivier GARDE]

Hi Erik,

I have take a look of your Ne/Ar Spectrum and It seems that the spectrum is saturated especially in the red part (see arrow indicating one of the saturated line).

[Robin Leadbeater]

Hi Erik,

You are not the first to have this sort of problem. I find the ISIS fitting routine is very sensitive to the pixel setting even to 2dp but once you find the setting that works, it is repeatable. (You will know when it is correct as the fit error will drop to a fraction of an Angstrom.) I have even got it to work successfully with 2x binning 2x 6.45um pixels where the lines are not so well resolved at the blue end.

It is important to have a well focused lamp spectrum, particularly at the blue end and a well exposed low noise lamp image. Use a sum of exposures to get good SNR on the faint lines without overexposing the stronger ones (Note the fit does not use the strongest lines, which helps) One hint is to look at the pixel positions the program chooses on the image. You should be able to see which line(s) it is getting wrong. You can then see what happens as you change the pixel value and try to find a value which correctly locks onto all the lines. If that does not work, try using a different line as the reference point and if all else fails, you can always use file mode with a list of lines with the offending line removed.

HTH
Robin

[Robin Leadbeater]

Erik,

Have a look at this thread too
http://www.spectro-aras.com/forum/viewt ... f=8&t=1246

Robin

[Erik Bryssinck]

Hi Robin & Olivier

Thanks for the quick reply!
I'll check this all out . First take my calibration images again.
I'll keep you informed.

best regards
Erik