Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Optical Smoothest yet not specular metal or plastic sheets?

  1. Sep 14, 2018 #1
    I'll get into the details why I need this for below but first I'll explain what I'm looking for.

    I basically need a flat sheet of aluminum or any plastic in grey or black color that doesn't provide specular reflections, yet is smooth enough to not have any noticeable texture when looking with the naked eye from the closest possible distance.

    I've checked some catalogues for banners, video projection screens, matte plexiglass and matte aluminum, sandblasted glass and acrylic and the off-the-shelf ones I've seen so far have a pretty noticeable texture from up close. I've also gotten some silver nano sized particle powder but haven't managed to succuessfully evenly adhere it to a surface. Haven't been able to test etching yet as well.

    The reason I need as smooth as possible is I am building a stereoscope ( https://en.wikipedia.org/wiki/Stereoscope ) but the images projected by a pico projector rather than printed on the screen. This poses a problem as any texture on the screen becomes much more noticeable, more than when looking from very close distance. However if I use a specular screen such as smooth plastic or metal then the issue is the lens of the projector becomes visible as well which even though is AR coated still has some of the projected light scattering inside it or projecting on it as well so a very visible glare is seen with the projected image. So I think a specular screen is out of the question.

    With diffuse screens though the off-the-shelf screen materials I've tested so far are too textury.

    Any ideas what I can use given this info?

    PS. I am aware of VR headsets, I'm working on something different for which HMDs aren't suitable.
     
  2. jcsd
  3. Sep 14, 2018 #2

    Tom.G

    User Avatar
    Science Advisor

  4. Sep 14, 2018 #3
    "9 times finer" isn't fine enough I believe and that's fine for video projection screens that are meant to display at most 4-8 times more pixels than 1080p ("future-proof" "up to 8k" as second result mentions) and be viewed from at least a meter away and need the ability to be rolled up and stored. For this though what I said still stands and when I say off-the-shelf screen materials I've tested so far I also mean the catalogues some of these projection screen companies sent me.
    You have to realize that even 8000 pixels over 100 inches is nothing compared to 1920 pixels over just few inches viewed through a magnifying lens like in a stereoscope.

    What I need basically doesn't have a market so it's fine there isn't an off the shelf product just for what I need, but I was hoping we could find an off the shelf item that I could repurpose or modify or build my own somehow.
     
  5. Sep 14, 2018 #4

    Tom.G

    User Avatar
    Science Advisor

  6. Sep 14, 2018 #5
    Thank you but I'm sorry, Tom, it seems like you're just Googling stuff without checking the results. I can use google too, but I need helpful links or advice, not just links. Sorry for being blunt but this is what I mean:

    quote from the patent:
    "Projection screens according to other exemplary embodiments of the present invention comprise a substrate including a reflective metal face having a textured surface, to provide reflective microstructures of about 0.5 μm to about 500 μm in size"

    "Projection screens may be fabricated, according to exemplary embodiments of the invention, by texturing a surface of a reflective metal substrate..."

    Well that's what I am basically asking, how to get a metal with small enough textures. The patent talks about more stuff to improve gain while retaining maximum possible FOV and I don't even need that because the eyebox in a stereoscope provides a very tiny FOV, I just need a metal (or plastic if that is possible) sheet that is smooth enough (small enough texture) yet not so smooth to be a completely specular mirror.
     
  7. Sep 14, 2018 #6

    Tom.G

    User Avatar
    Science Advisor

    OK, sorry.

    I was operating under the apparently false assumption that the granularity size was the important characteristic and other things, scattering angle, etc., were secondary as long as they didn't get in the way.
    I do have some vertical blinds in the bedroom with no specular reflection and no visible texture using a magnifying glass of 2in. focal length. Trouble is they are an off-white plastic and I have no idea what brand they are, but that is at least an existence proof that such characteristics are available.

    Hopefully others here can find better options for you.
     
  8. Sep 14, 2018 #7
    Tom, what is this in response to?

    That does sound like something I could use. Haven't seen such matte blinds myself.
     
  9. Sep 15, 2018 #8

    Tom.G

    User Avatar
    Science Advisor

    In response to a continuing theme in you posts:
    and the quote from the patent...
     
  10. Sep 15, 2018 #9
    I'm still not following, sorry. What do you mean by "I was operating under the apparently false assumption that the granularity size was the important characteristic and other things, scattering angle, etc., were secondary as long as they didn't get in the way."?

    I've said few times the textures (granularity) I currently achieve are too noticeable. So it is the important characteristic.

    So what do you mean?
     
  11. Sep 15, 2018 #10

    Tom.G

    User Avatar
    Science Advisor

    I mean that:
    • The referred-to patent described feature sizes at or well below the resolution of the human eye, which I believe met your requirement in post #1 of "not have any noticeable texture when looking with the naked eye from the closest possible distance."
    • And that the existence of the patent may give you a lead to a product as you requested in post #3 for "I was hoping we could find an off the shelf item that I could repurpose..."
    • And your post #5 "The patent talks about more stuff to improve gain while retaining maximum possible FOV and I don't even need that... led me to believe that scattering angle and FOV were detrimental to your requirements rather than don't-cares.
    • Also in your post #5 you stated "I need helpful links". OK, I was mistaken, the patent link turned out to be not helpful, for which I apologized.

    It's getting late here, we can continue this another day if need be. Goodnight.
     
  12. Sep 15, 2018 #11
    I wasn't saying the patent was not helpful because of what it achieved and I said it achieved more than what I needed. I was just saying it doesn't mention how the tiny granularity size/texture size on a reflective surface is produced because that wasn't the point of the patent.

    As for product lead, sadly first time I hear of such a projection screen and I've researched them all so this probably never became a product. Probably too expensive to produce and can't be folded up so has to stay in the size range of OLED/LCD screens at which point those are better objectively as don't require a projector .

    Still, thanks for the time and ideas.
    Those vertical blinds sound very interesting and I'd never expect to find a smooth but matte material there (the ones I've had in offices have had decorative textures on them). Definitely could work. Any chance you could figure out the brand? You might be able to find some branding on the slats (the plastic weights at the bottom).

    Some kind of rubbery material might also work.
     
    Last edited: Sep 15, 2018
  13. Sep 15, 2018 #12

    verty

    User Avatar
    Homework Helper

    What about a gel and a mirror? Or car tints on a mirror? You don't need to mass produce it, right?
     
  14. Sep 15, 2018 #13
    Not sure how that can be made to work. Please tell me more.
     
  15. Sep 15, 2018 #14

    verty

    User Avatar
    Homework Helper

    Well it's up to you to make it work. It's nothing to do with me, I just made a suggestion and it's for you to take it forward now. Bye.
     
  16. Sep 15, 2018 #15

    berkeman

    User Avatar

    Staff: Mentor

    Apolgies if this has been covered already in this thread (I just started reading it), but it seems like etching may be a good approach, as used in the semiconductor fabrication industry. What size do you need for this sheet? It sounds like it will be pretty small, right?

    Using the semiconductor etching techniques of typical fab lines, you can put down all sorts of patterns of different colors at very small geometries. Do you have a university near you that has a small clean room and fab line as part of their EE program? If so, you may be able to ask for help in making the surface that you want. They are often looking for interesting projects for their students in the EE program. You could look at what sorts of materials you would use and what etching patterns and layers to use to make the surface that you want. Just an idea. :smile:
     
  17. Sep 15, 2018 #16
    No. I'm trying to be polite here but this is not an advice. That's not how you give suggestions by throwing words and leaving. You can't give a hypothetical solution to a problem by only mentioning the "what" without the "how". You mentioned two materials and gave zero explanation how they would be used. That's not a suggestion. Both gels and tint films are specular. Unless you mean something else than what people understand by "gels" and "films", you should specify and I asked you to. Otherwise saying you gave a suggestion and saying bye after being asked to clarify is just plain rude and pollutes the thread with pointless posts.
     
    Last edited: Sep 15, 2018
  18. Sep 15, 2018 #17

    berkeman

    User Avatar

    Staff: Mentor

    I think there are frustrations brimming over from your other threads and posts. Let's all stay calm here and try to help you figure this out. Thanks.
     
  19. Sep 15, 2018 #18
    Berkeman, my supposed frustration comes from the two replies here, not internally or the single other thread I've made. I was given a suggestion with the poster refusing to clarify what he meant at all. I have zero interest in starting an online argument but the last response from the poster was uncalled for.
    I wanted to clarify that, now I can move on and respond to your suggestion, where you actually explain what you have in mind (etching).
     
  20. Sep 15, 2018 #19
    Definitely considering etching as a possible option at this point. My size requirement is only 50x50mm square space to be etched or otherwise textured.
    Any ideas what's a typical etching resolution for a PCB? Technically I think with either laser cutter or a demagnified DLP projector the resolution can be as low as the wavelength of light itself so below my ~5 micron texture requirement but how well the physics and chemistry behind etching will allow such low res mask textures to remained adhered to the sheet during the etching process and prevent the reaction to dig under the mask is another story I have no idea about.
     
  21. Sep 15, 2018 #20

    berkeman

    User Avatar

    Staff: Mentor

    I wasn't suggesting PCB copper etching -- that's probably around 0.1mm for typical PCB processes, I would guess. I was suggesting using silicon IC fabrication etching and deposition technologies. But I'm still not clear how big of a surface you need to have in your application. If it's bigger than about a 4" wafer, it may not be practical or low cost to use IC fab technology.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?