slus...@lw4u.com wrote:
On Feb 12, 2:02 pm, Bruno Barberi Gnecco
brunobgDELETET...@users.sourceforge.net> wrote:
I need to design an optical setup to project images coming from a
multimedia projector to certain surfaces, for instance a cylindrical
screen. I've been thinking about using mirrors for that instead of lenses,
since they seem to be cheaper and easier to work with (I can use polished
metal sheets, for instance).
I'm looking for books or other material that help me design these
optical systems. I'm particularly worried about keeping image focus across
the screen. Any recommendations? Any practical advice?
Thanks a lot for any help.
Do you already have some experience at fabricating sheet metal into
optical surfaces? It is *much* easier to talk about it than to do it.
Forming sheet metal into simple but efficient light collecting shapes
(solar furnaces, condensor systems, etc.) is already a challenge.
Imaging is at least an order of magnitude tougher.
I was expecting to either deform thin sheets previously polished
or coated with a reflexive material, or to coat them later. From earlier
calculations and tests it seems that the surfaces won't be too
complicated to shape, and I've been talking to people who have made
mirrors before. The overall opinion is that, for the quality I'm
aiming (not a telescope, but a simple projection system) I should
be able to achive the quality necessary.
Any tips, comments or "I've tried that and it doesn't work"
advice?
I don't want to discourage innovation and experimentation, but you
should know that for okay image reproduction, optical surfaces
typically must be accurately formed (figured) to within a small
handful of wavelengths of light. Thus, for yellow light (~0.59
microns), you might want to form your surface accurately to within
maybe 6 microns or less. That doesn't even take into account surface
finish requirements.
On the other hand, it sounds like you might have a fun project on your
hands.
I understand that, but I think that your figure of 6 microns
for surface accuracy is way too much for my needs. I agree that I need
a very good coating, plain and uniform, to get decent reflection without
too much aberration and light dispersion, but I think the results will
be satisfactory if the surface is off from the design for as much
as 200um as long as it's smooth. Remember that I'm using the mirror to
project onto a large screen far away from the user; imperfections of up
to a few % on the final image will not be noticeable.
It's fun but it's challenging
- Hide quoted text -
- Show quoted text -
