Homebrew CD spectroscope?!? / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Homebrew CD spectroscope?!?

1 2 next

Move Thread

LAN_403

PSCG

Tue Aug 09 2011, 08:06AM

Registered Member #3792 Joined: Sun Mar 27 2011, 06:07PM
Location:
Posts: 136

After searching for "Homebrew Diffraction Grating" on Google, i came accross this : http://sci-toys.com/scitoys/scitoys/light/cd_spectroscope/spectroscope.html

And my question is : is it possible to create a (good?) diffraction grating spectroscope this way? Professional diffraction grating spectroscopes are not cheap and are hard to get (at least here).

Proud Mary

Tue Aug 09 2011, 08:19AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

The CD or DVD spectroscope is good fun to demonstrate the principle, but of course all the emission lines are curved and not well dispersed. I couldn't see the Fraunhofer lines at all when I tried it.

I bought some of these gratings on ebay

and found them better than I expected for the small price.

Conundrum

Tue Aug 09 2011, 10:19AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4059

Maybe you can adapt the CCD off a camcorder as a diffraction grating?

-A

Forty

Tue Aug 09 2011, 05:22PM

Registered Member #3888 Joined: Sun May 15 2011, 09:50PM
Location: Erie, PA
Posts: 649

what quality of diffraction grating is required for such a thing? lines/mm, material? there are a couple whole sheets of diffraction gratings on lcd screens.
of course with a key to the optics lab, i'm sure i can find better..

PSCG

Tue Aug 09 2011, 05:50PM

Registered Member #3792 Joined: Sun Mar 27 2011, 06:07PM
Location:
Posts: 136

I searched the Google for modifying CCD for diffraction grating spectroscope, but it didn't give me any results about it (maybe i'm too stupid?).

The spectoscope must be capable of analyzing the 600 - 650nm portion of the spectrum with high resolution, so the grating must be at least 1000 lines/mm, of the reflection type and the material (preferably) from glass. Not interested for wavelenghts outside that area.

Bjørn

Tue Aug 09 2011, 06:11PM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

Have a look here

, make sure to read the post about variable track pitch.

I got some of that plastic film grating that Proud Mary links to, it is better than CDs and DVDs. If you keep touching it it will stop working, it will start working again if you manage to clean it properly.

To get started the quality is not important, for example to measure the wavelength of a laser pointer you can use a knife to make two slits, then use the formulas in my link to calculate the wavelength. The farther apart the slits are the longer baseleline is required to be able to measure anything. Several metres in the case of making the grating without a microscope.

When you want to study weak light sources and split close wavelengths you need a high quality grating or it will take forever with a high quality camera to gather enough light to record it.

Conundrum

Tue Aug 09 2011, 06:37PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4059

PSCG wrote ...

I searched the Google for modifying CCD for diffraction grating spectroscope, but it didn't give me any results about it (maybe i'm too stupid?).

The spectoscope must be capable of analyzing the 600 - 650nm portion of the spectrum with high resolution, so the grating must be at least 1000 lines/mm, of the reflection type and the material (preferably) from glass. Not interested for wavelenghts outside that area.

Essentially the surface of the CCD acts like a diffraction grating because of the vertical lines.

I noticed this when dismantling a camera, never actually tried it but it makes sense.

-A

Shrad

Tue Aug 09 2011, 09:33PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

a nice idea would be to construct a homebrew scanning michelson interferometer, capture the resulting beat frequency with a soundcard and perform a (eventually realtime) FFT, and calibrate against known light sources
(helium neon 632.5nm, green 532nm and 1064nm leakage)

it would be easy to obtain a 0.1nm accuracy with a homebrew setup out of scanner/printer parts

PSCG

Wed Aug 10 2011, 09:59AM

Registered Member #3792 Joined: Sun Mar 27 2011, 06:07PM
Location:
Posts: 136

To provide more info, the project that will use the spectrometer is an Optical Frequency Standard based on the 640.2 (or 585.25) nm resonance line of Ne atom (like the cesium frequency standards, but in the optical range), must be able to scan a very narrow area below and above these lines for the computer to resolve the transmission line. I'm still on the designing part, so no testing for now

@Shrad: The original thought about a diffraction grating spectroscope was the lack of complicated construction. Michelson interferometer is a lot more complicated that diffraction gratings, and that's why gratings were my first choice.

@Bjorn: Fortunately, (most) strong neon lines are far apart each other, so splitting close wavelenghts is not my problem (it will also be under magnetic field, so the Zeeman effect will help me split them more). And the solution with the slits isn't bad at all: simple and sturdy (the points that the spectroscope must cover). But i must make it scanning and interface it with the computer (math will take care for the rest and the computer will draw the resonance line on the screen).

Shrad

Wed Aug 10 2011, 02:10PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

with a soundcard, a CaF2 window, several mirrors and a piezzo speaker, you have a michelson without any mechanical engineering needed

what do you mean by narrow? if really narrow, a scanning fabry-perrot interferometer would be the best, plus you could buy some dye laser cavity mirrors designed for 642nm (tunable, so broad reflectivity around 640nm)

have a look at sams laser faq under spectrometers and interferometers, you'll find your need ;)

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob