• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 27 subscribers
  • Views 881 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DMD Spectroscopy Optics

ThomasMs
Prodigy 30 points
Other Parts Discussed in Thread: DLP5500

In a current research project I´m trying to project a light spectrum which was dispersed by a diffraction grating to a XGA DMD surface. However, I´ve some problems in realizing the 45 degree light incidence angle without distorting the image of the spectrum.
Is there any well-proven mirror combination (or lens/TIR prism) which can fulfill these requirements in a relatively easy way? Which optical configuration was used in the demonstration shown in the TI Spectroscopy online video? I´ve seen some optical configurations which are commonly used in DLP projectors. Unfortunately, they usually used customized prism optics combined with relay and other lens systems.

Thanks a lot,
Thomas

  • 0
    TI__Genius 9935 points

    Thomas,

    The orthogonal pixel array DMDs do require the illumination to be directed at the DMD from a 45 degree angle with respect to the sides of the DMD. (This is in addition to the 24 degree from normal angle of incidence.) The light path does get a bit hard to visualize. Since this only has to do with the angle of the illumination light, this isn't usually a orientation problem with respect to an image. However, with a dispersed spectrum, you have the additional constraint of the orientation of the image of the spectroscopic slit (which determines the direction in which the spectrum is spread out).

    Usually, in a projector, the illumination is done through a specially designed TIR prism. These TIR prisms are difficult to find or purchase, but can possibly be "liberated" from a dead or surplus DLP projector. You can certainly use a TIR prism, but it still does not address the issue of the direction of the spread of the spectrum.

    Maybe what I did can help as an example. I did a demo of spectroscopy principles using a DLP5500 DMD, which is also an orthogonal matrix DMD. I simply rotated the DMD 45 degrees with respect to the plane of the dispersed spectrum. This results in the spectrum being spread across the diagonal of the DMD. This means that the wavelength selection pixels, rather than being vertical stripes of "on" pixels, instead need to be diagonal stripes across the DMD (at a 45 degree angle). This makes things harder with respect to creating the pattern data for each wavelength. But, it works. The only places that you get a bit pinched for space are toward the diagonal edges of the DMD, where the wavelength selection stripes get short.

    Perhaps this will help.

  • 0 in reply to Pascal DLP
    Prodigy 30 points

    Dear Pascal,
    Thanks for your answer. The image orientation constraint is actually the biggest problem. I´ve already tested the more or less obvious approach of simply turning the DMD 45 degrees. It works fine but the spectral bandwidth is due to the relatively low fill factor insufficient for our application. However, I will try to solve the problem by using several mirrors folding the image of the spectrum into the mirror plane.


    Nevertheless, on your TI webpage are some examples shown how to use the DMD in spectrograph applications. The appropriate pictures and drawings all are showing an ideal fill factor. Therefore, is there any TI engineer whom I can ask about the implementation of optics in the shown concept/prototype systems?

    Best regards,

    Thomas

  • 0 in reply to ThomasMs
    TI__Genius 9935 points

    Thomas,

    I would like to dig deeper into your questions. To do so, I need to know more about your project. Are you doing research, or developing a product? What is the goal of your project? If you would like to take the answers private, I can do that - just let me know.