Does Diffracted Light Need to Be Focused or Collimated on CCD?

[einstein1921]

The question could be stupid,but It really confused me.
when we capture spectroscopy using reflected grating and CCD, Does the diffracted light must be focused on the plane of CCD (using a plane grating),or it must be collimated to arrive the CCD.

 

[Simon Bridge]

Neither. But the CCD surface does need to be wide enough to collect all the fringes you are interested in.
Spectrometers would usually be built with some sort of extra optics though.

 

[Drakkith]

I believe the light is typically collimated before passing through/reflecting off of the grating, then focused down onto the CCD sensor. Otherwise you don't really have any fringes, just a big blurry mess.

 

[Simon Bridge]

It has to be collimated first - but the CCD is just acting as a screen.
Typically there are other optical elements for light-spectroscopy ... a spectroscope, for eg, usually has a pair of telescopes. What's needed depends on what you want to do with it.

 

[einstein1921]

It has to be collimated first - but the CCD is just acting as a screen.
Typically there are other optical elements for light-spectroscopy ... a spectroscope, for eg, usually has a pair of telescopes. What's needed depends on what you want to do with it.

If I want to design a spectrograph , I need plane grating ,colimator, focus mirror. what confused me is where should I place the CCD?

 

[einstein1921]

I believe the light is typically collimated before passing through/reflecting off of the grating, then focused down onto the CCD sensor. Otherwise you don't really have any fringes, just a big blurry mess.

If I want to design a spectrograph , I need plane grating ,colimator, focus mirror. what confused me is where should I place the CCD?

 

[einstein1921]

Does CCD have itself focus system?

 

[Drakkith]

If I want to design a spectrograph , I need plane grating ,colimator, focus mirror. what confused me is where should I place the CCD?

Does CCD have itself focus system?

Building a spectrograph isn't really something we can cover very well. Instead, you're going to have to get into the nitty gritty details of it. I suggest reading up on the subject. Here's a book I have that may help.

https://www.amazon.com/dp/1441972382/?tag=pfamazon01-20

And check this site out: http://www.astrosurf.com/buil/spectrographs.htm

 

[Khashishi]

The image should be focused on the plane of the CCD.

 

[Simon Bridge]

There is a common confusion about what "focussed" means - lots of people think it means the screen has to be at the focal length of the lens system.

Drakkith is correct - go into the nitty-gritty.
We can only really do general principles in this forum.
The general principle here is to design the internal optics around the basic components and what you want them to be able to do.

i.e. the spectrum you want to image is probably bigger than the CCD surface - so you'll need to do something to make it smaller. Moving the CCD surface closer will help but too close and the spectrum image be blurry... since the grating is not a point source for the fringes. So you'll need some sort of lens. You'll need to figure out what sort you need compared with the placement and type of the CCD. The manufacturer will probably have published specs to help.

 

[Khashishi]

Honestly, it's not that complicated, and you can get pretty far without the nitty gritty. In a typical system, you will have a slit, two collimating elements, and a grating, with grating between the collimating elements. The collimating elements might be curved mirrors or lenses--it doesn't matter (lenses will have some chromatic aberration, but you can use camera lenses which are quite excellent). As long as you aren't going for really small f/d, geometric optics should be good enough. If you know the focal length of your two collimating elements, it is simple to figure the distances.

Put the slit at the object plane and the CCD at the image plane. The object is to see the slit in focus at the CCD. The grating between them will change the angles, but won't really change the distances.

 

[Simon Bridge]

yep - a telescope into make sure that parallel light is sent to the grating.
This will need to be adjusted for the source distance and type ... i.e. what you want to look at.

the back end wants an aperture to single out a single order of the spectrum - i.e. not the central maximum.
you then want another optical system to put an image of the spectrum in the right place for the CCD whatsit to be able to see it. That depends on the CCD specs.

For visible wavelength spectroscopy this is quite easy to set up if you don't have other restraints - since you can usually just eyeball the setup to see if it is working. I've done something like this using cardboard tubes, crazy glue, and a plastic protractor.
You calibrate the device using well-known spectra (hydrogen, sodium etc).
Building scientific equipment used to be a core skill - now it seems to be a specialist field.
But it's fun.

 

[einstein1921]

The image should be focused on the plane of the CCD.

What is difference between the plane of CCD and the position of CCD chip ? where does the light should be attack? what I want to build is a EUV spectroscopy!

 

[einstein1921]

what I want to build is a EUV spectroscopy!

 

[Cthugha]

what I want to build is a EUV spectroscopy!

In that case you also need to worry about operating in really high vacuum, using components (optics, CCD) that work in the EUV, radiation damaging the components and radiation damaging the soft matter operating the spectrometer (that means you). This is something you do not just build at home. You should have an experienced advisor for this.

 

[Drakkith]

What is difference between the plane of CCD and the position of CCD chip ? where does the light should be attack?

The plane is the surface of the chip that you focus the light onto.

 

[Simon Bridge]

what I want to build is [for] EUV spectroscopy!

"!" indeed :)

EUV wavelengths: from 124 nm down to 10 nm ... which is quite a range.
Astronomy may only go as short as 100nm - what do you need the spectrometer for?

that would mean a grating with order 10nm spacing ... 1e-8 or 100,000 lines per mm.
Looks like you can source them that fine. There are also CCD cameras designed to operate in that range.

Reading around the topic:
http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=1242170
... usefulness of CCD imaging for EUV spectroscopy.

Considering the kinds of questions being asked, my gut reaction is to advise trying to build simpler, lower energy, spectrometers first ... get used to what's involved.

 

[Khashishi]

Lenses won't work in that range, and you'll have to use mirrors, I guess.

 

[Simon Bridge]

EUV lensing is done with mirrors yep:
Since there aren't any useful transparent materials for EUV light, these lenses rely on curved mirror surfaces with tailored resonant-reflective multilayer coatings formed from alternating layers of molybdenum and silicon, just a few nanometers thick...
http://spie.org/x8891.xml