Neon lamp to set spectroscope

[p1taylor]

hi all. hope someone can help, I remember seeing a article on how to get a neon setup lamp for spectroscope from a fluorescent starter.

peter

[robin_astro]

Hi Peter,

This is the one 

http://www.ursusmajor.ch/downloads/sques-relco-sc480-calibration-lines-2.0.pdf

I used a disposable camera flash circuit to power one

http://www.marcellocucchi.altervista.org/html/neonlamp.html

but note also the caveats here

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

(In French to start with but the important comments are in English down the thread)

HTH

Robin

[p1taylor]

Thanks Robin that is exactly the one I needed.

peter

[athornett]

I have started to make my RELCO starter based calibration light - my first version is driven by 240V through inverter in field, although I would like to have a go at making the disposable camera circuitary based version in the future. At our Astro club, earlier in the year one of our members (Ed) made some fantastic leisure battery based battery packs for us and hence we have access to ready supply of low amp 250V in field.

click on link below to view photos of this first stage of RELCO fluorescent starter calibration light:

https://roslistonastronomy.uk/modifying-relco-fluorescent-light-starter-to-create-neon-calibration-light-for-spectrometry

Next step is to house it all in something safe to stop getting electrocuted when used outside!

I also intend to use a circuit breaker in line with the plug.

NB I am not a qualified electrician so if anyone follows in my footsteps please ensure you are qualified to do the job or ask someone who is to do it for you.

Andy

[Merlin66]

Many of us are making use of the RELCO bulbs as reference lamps.

I have a set-up which makes use of a 12v to 220v PCB built into the spectrograph.

This restricts the higher voltage to the enclosed housing. Very safe, very effective.

 

[athornett]

That sounds a great idea. I have just finished my 240V RELCO calibration light:

https://roslistonastronomy.uk/first-calibration-spectra-from-relco-neon-starter-bulb-reclo-sc-480-s53993-taken-with-science-surplus-diy-spectrometer-9-9-2018

I have used external boxes and sealed everything with a glue gun and will also add in RCD in the field.

Andy

[athornett]

My first calibration spectrum using my new RELCO starter neon/argon light source - this is from Science Surplus DIY Spectrometer. As this Spectrometer is set up, it is missing most of red end spectrum.

Andy

 

[Merlin66]

Andy,

What you have is a reference profile...when you can identify the various neon/ argon lines and allocate a wavelength to each, you can then use this info to calibrate in wavelength (nm or A) and determine a valid A/pixel relationship.

When beginning it's sometimes useful to take a solar spectrum and use this to compare with the lamp (a bright sky should give you a good result) - may make finding and identifying the lines easier - you can access a detailed solar spectrum at http://fermi.jhuapl.edu/liege/

 

 

 

[athornett]

The following links to the start of my first attempt at calibrating above RELCO starter spectrum against solar spectrum.

https://roslistonastronomy.uk/category/spectroscopy/calibration-spectra

Andy

[robin_astro]

Hi Andy,

The solar spectrum is tough to use at this resolution because all lines you are trying to measure are blends of many lines so you don't know their precise wavelength.   If you are interested in an astronomical calibration source then hot, A type stars are good as they have strong unblended H Balmer lines. If  you produce say a 2nd order fit using these, your calibration should be close enough to  identify the RELCO  lines.  At this resolution you still need to chose the RELCO lines with care though, watching out for blends.  Once you have a good set of reliable lines you can use these to calibrate all your spectra. 

You can see half way down this page how ISIS uses 13 lines to calibrate  the similar resolution ALPY using a 4th order polynomial to better than 0.2A precision

http://www.astrosurf.com/buil/isis/guide_alpy/resume_calibration.htm

(Method 2)

Robin

[robin_astro]

Looking at your RELCO spectrum on your blog here

https://roslistonastronomy.uk/wp-content/uploads/2018/09/CCDSPEC-RELCO-Spectrum-130918e-Image.bmp

It looks like you have some problem with ghost spectra at the blue end which could make it difficult to indentify lines.  Your spectrum should be cleaner than this, more like this spectrum taken with my ALPY

 

[athornett]

Thanks folks for your comments - really appreciated!

Just goes to show how difficult it to identify the lines on the spectra......any ideas on the best way to do that?

I am coming around to the idea that the compact fluorescent bulb is the easiest to use method for low resolution spectroscopy - readily identifiable lines and in my experiments they are highly reproduceable between bulbs.

I have started a separate conversation on Stargazers Lounge for Compact Fluorescent Bulbs as I could not find one:

I am still hoping to sort out the use of the RELCO bulbs though, as I have put a lot of effort into them and they are a neat solution to the problem of calibration in the field.

Andy

 

[athornett]

Hi Robin,

I have had a go at calibrating the RELCO spectrum against CFL in order to determine the approx. wavelengths of lines on my RELCO spectrum - I am hoping I can then cross-check that the atlas of RELCO lines to get accurate wavelengths against http://www.ursusmajor.ch/downloads/sques-relco-sc480-calibration-lines-5.0.pdf

More information on my calibration process of my CCDSPEC Spectrometer RELCO Glow Starter Spectrum today at link below:

https://roslistonastronomy.uk/calibrating-relco-starter-against-12v-compact-fluorescent-lamp

In order to take the spectra today, I double bagged the RELCO/CCDSPEC - so hopefully, I will have avoided the ghost/contamination spectra Robin noted above.

Andy

 

 

 

 

[athornett]

Hi Robin,

Thanks for posting your RELCO spectrum - below I have graphed it in RSPEC - is there any chance you could post a calibrated version (so that I can extract wavelengths of the peaks) + DAT file so I can load it in RSPEC and get accurate wavelength data?

Andy

 

[athornett]

I have just had a look at the 200 lines/mm chart in the atlas at www.ursusmajor.ch/downloads/sques-relco-sc480-calibration-lines-5.0.pdf and sadly I can't see any obvious match on the lines with my chart - ahhhh! Help please!

Andy

[robin_astro]

Hi Andy

Attached is a zip file using my ALPY 600 and RELCO containing 

1.  spectrum image (2D fits file)

2. wavelength calibrated spectrum (dat file)

3.  text file with the 4th order dispersion equation, the 13 lines that ISIS used to generated it and the residual errors (RMS  error 0.14 Angstrom, ~1/30th pixel)

Cheers

Robin

RELCO_ALPY_Calibration.zip

[robin_astro]

The image I posted earlier in the thread was gamma stretched and saturated to show the lines better so the spectrum you have produced from it will not have the correct relative line intensities 

To get you started the strong line you have labelled 4367.6 is actually an Ne line at 5852.49. From that you should be able to work your way up and down to identify the other lines in Richard Walker's annotated spectrum

Robin

[athornett]

That's brilliant. Thanks Robin!

Andy

[athornett]

Hi Robin

In image below I have tried to match your spectrum to mine. It is not obvious to me where I should stretch the image/i.e. a second data point compared to your data - can you help?

I have enclosed my combined image + data as ZIP file.

Andy

 

RECLO Calibration from Stargazers Lounge 021018.zip

[robin_astro]

Hi Andy,

Here is a rough 2nd order calibration of your 30 sec RELCO exposure, using 3 lines, overlaid on my ALPY spectrum. You can see the good agreement, though  I noticed your lamp spectrum is much noisier with a high background compared with my 10 sec ALPY spectrum.   (Note the extra line in your spectrum at 6563A. This is Hydrogen Alpha. It has leaked through the glass in my older lamp).  The sensitivity in your spectrum drops off at the UV and IR ends compared with the ALPY spectrum.  Are you using a UV/IR cut filter by any chance ?

Also attached is the dat file of your roughly calibrated spectrum. 

Robin

relco_ccdspec_300918c.dat

[robin_astro]

I forgot to add which lines I used for calibration

4545.05 (Ar)    5852.48  (Ne)   7067.22  (Ar)

The dispersion is approximately 6.5A/pixel , though this varies along the spectrum of course as it is non linear

It would be better to add a few more lines,making sure they are well spaced and cover the full spectrum (particularly the blue end, but they were rather indistinct in your spectrum)  and do a 3rd order fit

Robin

[athornett]

Thanks very much Robin - for future reference I have loaded up your calibrated version of my RELCO spectrum here. The spectrum here was taken using CCDSPEC spectrometer - I attempted to exclude extraneous light by wrapping both CCDSPEC and RELCO light in 2 layers but some light probably got in so spectrum won't be perfect. This is spectrum using RELCO and CCDSPEC - not using a telescope.

Robin used following lines for calibration, and commented that it would be better to add a few more lines, making sure they are well spaced and cover the full spectrum (particularly the blue end, but they were rather indistinct in your spectrum)  and do a 3rd order fit. The dispersion is approximately 6.5A/pixel , though this varies along the spectrum of course as it is non linear.

4545.05 (Ar)    5852.48  (Ne)   7067.22  (Ar)

Andy

 

[athornett]

I have now had a go at identifying the lines on the above CCDSPEC Spectrometer RELCO Starter Spectrum using http://www.ursusmajor.ch/downloads/sques-relco-sc480-calibration-lines-5.0.pdf

Once I am happy with the identifications, I will tidy up the diagram!

Robin - do you think I have correctly identified the lines?

Also included below are pictures of the CCDSPEC Spectrometer and my homemade RELCO starter calibration light and a dismantled RELCO Starter Unit showing the bulb as it comes out of the starter.

Andy

 

[robin_astro]

Hi Andy,

I would try to extend your coverage at the blue end a bit further if possible.  Your spectrograph should be able to cover down to at least 3800A.  As I mentioned though it looks like something is limiting your sensitivity at the blue and red ends. Are you using a UV/IR block filter somewhere, either on the camera or built into the spectrograph?

Cheers

Robin

[robin_astro]

On 05/10/2018 at 22:52, robin_astro said:

As I mentioned though it looks like something is limiting your sensitivity at the blue and red ends. Are you using a UV/IR block filter somewhere

Actually this latest spectrum looks a lot better with strong lines in the IR.  Have you removed a filter somewhere ?  (EDIT:  I see that spectrum was mine, not yours which explains the difference) You should nowbe able to identify some lines below 4100A. (the line at 3946 would be a good one to add if you can see it. Watch out for blends at this resolution though which is quite a bit lower than Richard Walker's spectra)