Spectroscopy - another taker!

[TheThing]

Was just about to get an early night when SWMBO mentioned that the clouds had parted and it was clear out.

I then had a mad rush to get set up and aligned before clouds covered my target for the night - Rigel. Just as I finished aligning, the clouds did indeed cover Rigel. I spent 5 mins looking for another target, but then noticed Rigel had cleared again, so I set up.

Next came probably the most frustrating hour of my life. I could find Rigel fine in the finder, a 40mm EP and a 9.5mm EP. I could get it in the FOV of my webcam as well, but no matter what I tried, I could not find the blaze anywhere! According to my Star Analyser 100 guide, with the mark on the analyser to the right when viewing the camera from behind, the sprecta should be to the right of the star in the FOV of the webcam.

I tried every exposure setting I could, fiddled with the brightness, gamma etc, but nothing. Was just about to give up for the night when I knocked the scope by mistake and found the spectra to the left of the star and at an angle! D'oh! At least I had something now!

Took four videos which I have stacked in Registax as below. I think my focus needs working on? Comments appreciated!

I've tried to open them with vspec but it needs fits or 'quimps'(?) Any suggestions how I save in that format please as a Save As doesn't help. I've tried putting it back thru Registax but that only allows tiffs which are no good.

I've run the video thru Rspec ,but as I don't know quite how that works, I'm not sure if the result is any good (if I can uplaod it here).

So, from the depths of despair to great joy at getting something on video and being able to see dark lines. I'm sure with more practise and help from Kent, John et al, I should be geting some better results soon and some proper science will be done!

[Rob1961]

Hi, Gimp v2.6 will save in *.tiff format and if you want to use a different program there is FitsLiberator v3 from NASA/ESA

Rob

[Merlin66]

When you get the .bmp from Registax, open IRIS and crop and save as a .pic file - this is a 16bit fits format.

Vspec will open it and allow you to calibrate the spectrum.

Good begining!!

Yes, try to focus the spectrum so the green area is tight in focus - this will minimise the chromatic aberrations.

Onward and upward!!

(Rigel is a B8Iab star so you've picked up some hydrogen Balmer lines and atmospheric stuff....)

[Merlin66]

I see from the crop lines that at least one of the spectra was imaged at an angle to the CCD frame....this will loose you resolution!

Try and move the grating so the spectrum lies along the horizontal axis.....

[jsandse]

Nice specta,

Got the balmer jump as well...yippee

Excellent first attempt

cheers

John

[Demonperformer]

Congratulations on these, from another 'newbie'

[TheThing]

Thanks gents.

Yes, the spectra were diagonal across the webcam. I still haven figured quite how to line it up properly. A task for next time.

I assuming the spectra was off to the left as I was using an 8" newt? That may also explain why it is diagonal. I think I need to line the web cam to the light path to prevent this.

Also, the 'exposures' were only 30 secs each, about 400 frames, so I think I'll need longer too?

I'll search out IRIS via Google and give that a go.

[andrew s]

Well done great start. I think its part of the initiation to have lots of software and hardware problems. Keeps us on our toes! Andrew

[Merlin66]

Hugh,

What was the distance between the grating and the CCD chip?

You may have to reduce it to 50mm or so to starT with - this will maximise the spectral coverage and allow you to include the zero order image (makes initial calibrations easier;))

I've updated the inputs to the TransSpec spreadsheet to reflect what I think is your set-up. See attached.

TransSpecV2.0-THING.zip

[Demonperformer]

The Thing: link to IRIS: Iris software

[TheThing]

Thanks Mr the DP, just downloading it now.

Ken, the filter is about 55mm away from the sensor, very roughly. I was using a LX modded SPC900, but in the non-LX mode. I do have a standard SPC900 in which the filter will be approx 45mm away. Do you think that will be better?

Thanks for the Transpec file. Very kind of you to run that thru for me. I'm hoping to disseminate it later this evening.

[Merlin66]

IMHO the webcam with LX will be better!

55mm to too large a distance with a webcam sized chip to get both the zero order and the spectrum on the CCD. Max is around 50mm.

(Play around and change the distance in the spreadsheet and look for the size of spectrum - this will show the answer....)

[TheThing]

Got the spectra loaded into vspe and produced the following graph. Unfortunately, I have forgotten all of the things that John showed me, so I'm not entirely sure what the graph represents (other than that the 'dips' will correspond with the balmer lines and the largest dip will be, I believe, Hydrogen Alpha).

I've forgotten completely how to calibrate as well, so if anyone can point me in the direction of a primer (either on-line or paper) I'd be grateful.

Have to admit, this is a very interesting subject. I'm enjoying it immensly.

nrigel 3 pic[1].bmp

[jsandse]

try this tutorial for calibration in vspec

Vspec - Tutorial 1

to help you I uploaded a star analyser spectrum which was taken with a toucam pro of Vega in my album here

Stargazers Lounge - jsandse's Album: Spectra - Picture

You can see it is a similar shaped profile to yours.

So that should help you calibrate your spectrum correctly - similar instrument response

cheers

John

[TheThing]

Thanks John, I'll give that a go later.

[TheThing]

Finally got my new pc/netbook up and running and had another bash at calibrating the spectra following John's excellent tutorial, as below.

Much happier with this one which has come out with Angstroms 11.3406/pixel - pretty close to the more experieinced members calibration.

Rigelcalibration.doc

[Merlin66]

I'm assuming, looking at the response curve out in the red/IR that you were not using a UV-IR filter on the webcam?

(good move - you want to record as much of the spectrum as possible)

Next stage is to make the corrections for the camera response curve ( you can use the standard Rigel spectrum from the Vspec Library).

You're getting there......

[TheThing]

No, I left the UV/IR cut filter off just to be on the safe side.

Just got to learn how to correct for the camera response now. I've not had this much fun for a while!