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.

DLP230GP: Light Engine Design

Part Number: DLP230GP
Other Parts Discussed in Thread: DLPC3432, DLPA2005

Hello, 

I am trying to design a wearable augmented reality HMD setup using the DLP230GP (with DLPC3432 and DLPA2005). I am currently working on a custom light engine design, and am trying to figure out how the rays from the light source would diverge after they reflect from the DMD and pass back through the TIR prism to the open air as an image. If I wanted to slightly expand the image size prior to reflecting the image off of several non-planar mirrors, do I need to have a full projection lens assembly included after the DMD and TIR prism, or could I just use a single convex or concave lens (aspheric or otherwise) to expand the image?  I understand that the point of complicated projector lens assemblies is to prevent various types of optical aberration, but the TI whitepapers and other non-TI literature that I read were unclear as to how much of a problem optical aberration(s) are for wearable HMD's.

Thanks,

A.

  • Hello A.,

    Welcome to the E2E forum and thank you for your interest in DLP technology.

    We are discussing your questions with engineers from our optics team and will return with an answer shortly.

    Regards,

    Austin

  • A,

    can you provide more details about the physical layout of your light engine. specifically the prism and lens you describe? from what I can understand you wish to place a lens in front of the wearers eye to adjust the size of the image seen through the prism?

    In wearable applications typically the geometry of the prism itself would size the image correctly to focus the image on the pupil of the eye. These curved surfaces are typically defined as freeform curved surfaces.  They are created from high powered optical design tools such as Code-V and Zemax.  The two complex surfaces combine the function of 4 lenses in a standard eyepiece.

    Optical aberrations need to be controlled for the eyepiece design in much a similar way to the full projection lens

    Best Regards,

    Arthur Brown

  • Arthur,

    My current design for the light engine uses an RGB LED, a collimating lens, a Fly's Eye Array, and a couple of relay lenses before the TIR prism. My current design for the TIR prism consists of two triangular prisms that will be laminated together. I do not have access to Code-V, Zemax, or similar software, so it would be very difficult for me to adjust the prisms to incorporate free-form surfaces.

    Neither the light engine nor the lens will be located in front of the user's eye, as my application requires that the user be able to see what is in front of them. I want to place the lens just in front of the TIR prism, after the light reflects off of the DMD and back through the prism, such that the resulting image would appear to expand in size from the lens focal point.  Placed a short distance in front of this lens will be a small non-planar non-spherical mirror. This mirror and several others will reflect the image to a partially transparent window / mirror in front of the user's eye such that the reflected image would overlay onto what the user can see in front of them, as in an HUD.  

    Thanks,

    A.

  • A.

    Thank you for the clarification. I believe I understand. between the DMD/prism assembly, and the series of mirrors that direct the image onto the HUD you wonder if a single lens is acceptable as a means to enlarge the image.

    I again reached out to our optics team with this clarification, and I will post back as soon as I hear from them.

    our team that supports E2E doesn't have a resident optics expert, so I appreciate your patience as we wait to hear back.

    Another resource you may be able to use is to contact the optical module manufacturer for this DMD. they may be able to help. here is a list of manufactures TI uses https://www.ti.com/dlp-chip/display-and-projection/pico-chipsets/buy-optical-engine/optical-module.html

    Best Regards,

    Arthur Brown

  • A,

    I consulted with one of our optics experts and in there opinion a single lens for sizing in an optical engine is not ideal. there exact words "won’t be anywhere close to acceptable nowadays". 

    they went on to suggest you to look at some microscope(like would be used in a lab or found online) eyepiece in the market since those are very similar to AR eyepiece. But AR eyepiece is harder to design due to the spacing available. The aspheric mirror(s) you use can help the image quality a little bit better, but one lens is still way too few for AR. The current AR eyepiece are already aspheric, and at least 4-5pc.

    When searching online, there might not be a lot of AR eyepiece design data, so microscope eyepiece design with similar field of view and F/# would be a good starting point to look at.

    https://patentimages.storage.googleapis.com/ee/b1/22/b1d034ea1a0952/US4525035.pdf

    Here’s a patent found online on eyepiece.

    Regards,

    Arthur Brown

  • Arthur,

    It looks like I'll have to use an already-existing lens design or find another way to expand the image then. Thank you for your help!

    Best,

    A.