Sunday, November 13, 2011

Ghosts in the Mirror

Professor Johnson explained atmospheric effects as decreasing the image resolution by increasing the full-width half maximum of the intensity plot. When we had first covered n-slit diffraction I would have thought that simply making the telescope lens bigger would counter this effect since resolution is proportional to lens diameter (resolution = \[Lambda]/diameter so a larger diameter means better resolution). 

Here's a simple explanation of how deformable mirrors work: In response to John's post Happy Halloween where he identified molecular cloud ghosts, here's a toast to making ghosts out of deformable mirrors.


Back to the problem of resolution - Naturally it is very expensive to make larger telescopes. The white board explanation showed each incoming wavefront as a line (assuming a very distance source,/k /circular wavefronts become lines), and atmospheric perturbation introduce phase differences so you had a bump readout instead of the flat line.

Another question I had but forgot to ask - how large are the disturbances in the wavefront and how much does deforming mirrors adjust this?

Another cool video about deformable mirrors and the private industry:

I liked this one more (barring the lack of ghosts) because the music was more motivating. I felt like scientists were superheroes.

3 comments:

  1. your diagram is really cute!

    gosh, I'm not sure how big the disturbances in the atmosphere are. i do know, though, that it is harder to correct for the disturbances at smaller wavelengths - I think that's partly because a small difference in path length leads to a larger phase difference at shorter wavelengths, but the atmosphere may also refract different wavelengths differently. how about you head back down to the AO lab and ask down there?

    ReplyDelete
  2. i mean, the disturbances in the wavefront due to the atmosphere...

    ReplyDelete
  3. hmm, one sense of scale I do know about: typical "seeing" limit (the size of the area that a star gets blurred over) is 1 to 2 arcseconds (optical wavelengths). if you were to take very very short exposures with your telescope, you would see almost the diffraction-limited pattern of the star (much smaller than the seeing limit), but jumping around between exposures within a 1 to 2 arcsecond area, because where the star gets refracted to changes from moment to moment. "first-order" AO corrects only for this phenomenon, using only a tip-tilt mirror that quickly changes its angle to keep the star in the same place. however, there are also more complicated wiggles in the wavefront as shown in your diagram, and you need a deformable mirror to correct for those to fully recover the diffraction-limited image of the star.

    ReplyDelete