Wavelength Off With Demetra Profile

[brownrb1 0]

Hi,

I've just started learning Demetra to reduce my data and get spectra profiles. I had previously posted a profile of Vega and, attached here, a profile of Beta Lyrae. In both cases, the wavelength of absorption and emission lines were off. What could I do to correct this?

Thanks

Dick

[Nicolas Durand 3]

Hello Dick

 

According to your screen, the automatic calibration fail to find all the calibration lines it is looking for.

-> The spectrum may not be "clear" enough (too short exposure, bad focus, ......)

-> The spectrum may not contains the calibration lines that the process is looking for (do you uses the Alpy calibration module ?)

 

To understand what is going wrong, go to the "Calibration" step and click on "Calibrate..." then "run" then check the "Detection" radio button (under the "run" button on the right side)

-> You will see all detected lines in your spectrum (vertical lines on the spectrum), and you will see the lines list that the process is looking for. Each of theses last lines must be found by the process.

-> Can  you post a picture of this screen ?

 

Nicolas

[Nicolas Durand 3]

I forgot to post the link about calibration that may help you :

 

 

Modifié 8 août 2020 par Nicolas Durand

[brownrb1 0]

Hello Nicolas,

Thanks for you input. I've just started using Demetra and really like it. I was our last night and used Demetra to get some spectra using it to get correct exposure and focusing work well. I use a 10" Meade SCT with f/6.3 reducer, Atik Infinity monochrome camera and the Alpy Calibration module. I'm not comfortable using the guiding module yet , so I used the small slit to get my bias frames, darks, flats, and calibration and put them in a file. I use the 3mm hole to get my spectra. When I do the data reduction and get to "Calibration" I click "Geometry", then 'Tilt'-"Calculate"-"Detect" and that works fine. I the do "Smile"-"Calculate"-and when I click"Detect" a screen shows 'Warnin-Invalid Search' I then  enclose one of the bright vertical line and then"Preview" and nothing happens. I run them anyway and get yellow warnings. I checked "Calibration"->"Run" and it said "Image wavelength correct" and the Calibration srceen is attached

Thanks Again

Dick

Calibration.pdf

[brownrb1 0]

Here's thw Calibration screen

Calibration.pdf

[brownrb1 0]

Hi Again,

Here's a profile of a spectrum I got last night and when I did the data reduction the same thing happened with 'Geometry' but the wavelength seems to be fairly close.

Dick

Vega_4.pdf

[OlivierG 782]

I think there is an issue with the raw data. Is it possible to see them somewhere?

[brownrb1 0]

Hi Nicolas,

Attached is 1 of 6 images of P Cyg and the 1 image of Alpy calibration images. I also got 7 bias frames, darks, and flats. 

Dick

pcyg_A_Object_60s_20200809_051952-2.fit

vega 2_Calibration_120s_20200729_162339-1.fit

[brownrb1 0]

Hi Again,

While I like my Atik Infinity for images , the resolution is not that great. I was thinking that I use my ZWO 178 mono for calibration and the Infinity for images and see how that works.

Dick

[OlivierG 782]

Your images are all over exposed.

 

Many lines of the Ar/Ne lamp are saturated, so the calibration couldn't be done in good conditions

 

And the same with your P Cyg profil : H alpha is saturated and H beta close to saturation.

 

You have also to better align your spectrum on your camera.

 

[Nicolas Durand 3]

Hi,

 

In addition:

1)I tried the calibration image (vega 2_Calibration_120s_20200729_162339-1.fit) with the calibration tool and the result is OK.

 

 

This differs from the "calibration.pdf" screenshot you sent.

I only replaced the calibration image from the Vega observation from the setup with your image (after removing the first line and column to make the image size identical to the other image of the Vega observation)

I didn't changed any process parameter (-> geometry correction is not perfect)

 

-> The screenshot that may provide other useful informations is this same screen but after clicking on "Detection" radio button near Identification : it allow to see which lines has been detected and first theoretical line that are look.

(or better : share me a wetransfer link with the observation export archive)

 

2)

Le 09/08/2020 à 20:48, brownrb1 a dit :

Hi Again,

Here's a profile of a spectrum I got last night and when I did the data reduction the same thing happened with 'Geometry' but the wavelength seems to be fairly close.

Dick

Vega_4.pdf

I don't think you can check the calibration quality using the target star (Vega) while you change the spectroscope configuration between calibration images acquisition and target star acquisition:

Le 09/08/2020 à 20:24, brownrb1 a dit :

I'm not comfortable using the guiding module yet , so I used the small slit to get my bias frames, darks, flats, and calibration and put them in a file. I use the 3mm hole to get my spectra.

-> Applying the detected calibration polynomial function onto the target (star) spectrum image in these condition is not valid. You have to don't change anything on the physical device for it can work. What you are doing here is something "coarse" that can help you for a first try and to help you to understand, learn and go on your learning progress path. But you really need to make the guiding module works to be able to check calibration using the star spectrum.

 

Nicolas

 

Modifié 11 août 2020 par Nicolas Durand

[brownrb1 0]

Hi,

I think you're correct-I spent today setting up and focusing my guiding module.

Dick