Hello,
I'm a product design engineer and I'm here today because I'm stumped
by an optics problem, and I hope that someone with more experience can
help me out.
I'm currently trying to educate myself on what practical options are
available to project an image on a surface which is nearly paralell to
the projection direction, as opposed to the ideal, which would of
course be perpendicular.  I have about an inch and a half of space to
rise above the intended projection surface, and so I was thinking a
reflector would be the best option to help maximize my available
distance.  I believe the term that best describes my intent is 'acute
angle projection'.

I took a shot at creating an optics diagram to better illustrate my
goal, and though I'm sure that there are errors in the light path, I
hope that it will help to clarify my arrangement.
I realize that there is an inherent problem in that traditional optics
bend light to a focal point.  By definition, a system with a focal
point converges to a single point which can be adjusted to clear
focus.  Because of my angle, the problem I have is that this point is
stretched.  The result I have achieved is a clear image projected at
center, and blurring quickly out of focus from that point.  My hope is
that I can find an arrangement or cut a custom lens that will
essentially have a variable or 'stretched' focal point which is the
inverse of the projection blur.

I've spoken with a several optics manufacturers, but all seem to be
experienced in more traditional optical engineering tasks.

In considering this problem, I have thought of a few ways where it
seems that it should be possible.  One theoretical example is to
imagine the image broken into 1000 "pixels".  Each pixel of light
could then be directed through one of 1000 individual lenses and out
to the resulting end image, each maintaining its individual focal
point, and producing and end result that is in focus in spite of its
orientation.  Effectively, 1000 pixels = 1000 lenses = 1000 focal
points.  This thought makes me wonder if I averaged the geometry of
said lens 'group' if that would deliver a shape which would achieve my
goal.  (a parabola, perhaps?)

I've considered some Fresnel type layered solutions, or a lenticular
sort of optic, but still haven't made any solid conclusions.

I'm still educating myself in the realm of optical engineering, so I
would greatly appreciate any insights, experience, or other
suggestions to help resolve this problem.

Thanks in advance for any help!!
-Kevin Dahlquist
img src="http://i59.photobucket.com/albums/g309/salukikev/
lightdiag.jpg"

On May 14, 3:02 am, saluki... at (no spam) gmail.com wrote:
Hello,
I'm still educating myself in the realm of optical engineering, so I
would greatly appreciate any insights, experience, or other
suggestions to help resolve this problem.

Thanks in advance for any help!!
-Kevin Dahlquist
img src="http://i59.photobucket.com/albums/g309/salukikev/
lightdiag.jpg"

The solution is actually quite easy. Tilt the object plane with
respect to the projection lens and your object plane will also be
tilted.

Helpful person wrote:
On May 14, 3:02 am, saluki... at (no spam) gmail.com wrote:
Hello,
I'm still educating myself in the realm of optical engineering, so I
would greatly appreciate any insights, experience, or other
suggestions to help resolve this problem.

Thanks in advance for any help!!
-Kevin Dahlquist
img src="http://i59.photobucket.com/albums/g309/salukikev/
lightdiag.jpg"

The solution is actually quite easy. Tilt the object plane with
respect to the projection lens and your object plane will also be
tilted.

What I think HP is referring to is similar to the way a view camera is
adjusted to give an undistorted image when the object plane being
photographed is not parallel to the film plane. The trick is to have the
object plane, the lens plane, and the film plane intersect in a single
line. Think of the planes as three pages of a book, with the spine as the
common line.

After a quick search:
http://en.wikipedia.org/wiki/Scheimpflug_principle

Dave

The "Schiempflug principle" describes object and image planes that are
tilted so that the two planes are in (paraxial) focus everywhere. (The
two planes intersect at the principal plane.) The images still show
strong keystone distortion.

Bob Knowlden wrote:
The "Schiempflug principle" describes object and image planes that are
tilted so that the two planes are in (paraxial) focus everywhere. (The
two planes intersect at the principal plane.) The images still show
strong keystone distortion.

Quite true, but depending on the application, keystone distortion may be
not an issue. For example, a projected keyboard - the image could easily
be shaped to produce a correctprojection.

Dave

The "Schiempflug principle" describes object and image planes that are
tilted so that the two planes are in (paraxial) focus everywhere. (The

sorry on the delay'ed response, just got back into town, and I need to
spend some time understanding the "Schiempflug principle" and the rest

saluki... at (no spam) gmail.com wrote:
So I'm still reading up on this.  Still working on this project today,
and trying to generate a setup which demonstrates the Scheimpflug
principle with regards to a projected image.  I understand how the
concept applies to film/camera setups, but in trying to apply the
concept to my projection setup, I'm running into some trouble.  Any
suggestions there?

These guys have solved a trickier proble; they do acute
angle projection carefully enough to use side projection
into a wedge to give a flat panel display ... once the
rays fail to undego total internal reflection!

http://www.camfpd.com/technology.htm

--
---------------------------------+---------------------------------
Dr. Paul Kinsler                
Blackett Laboratory (QOLS)        (ph) +44-20-759-47520 (fax) 47714
Imperial College London,          Dr.Paul.Kins... at (no spam) physics.org
SW7 2BW, United Kingdom.          http://www.qols.ph.ic.ac.uk/~kinsle/

So I'm still reading up on this. Still working on this project today,
and trying to generate a setup which demonstrates the Scheimpflug
principle with regards to a projected image. I understand how the
concept applies to film/camera setups, but in trying to apply the
concept to my projection setup, I'm running into some trouble. Any
suggestions there?

Helpful person <rrl... at (no spam) yahoo.com> wrote:
http://www.camfpd.com/technology.htm
Interesting.  I designed a system almost identical about 10 years ago
and it was not my original idea.  i wonder how valid their patents are?

I suppose that would depend on what they have patented, and when.  
It was 6 years ago I first saw a presentation (at a conference)
and thought it seemed a nice idea. I don't know how far back their
work goes -- since their wedges aren't all over the place by now I
assume the fabrication of optics or wedges is a bit trickier than
they hoped.

--
---------------------------------+---------------------------------
Dr. Paul Kinsler                
Blackett Laboratory (QOLS)        (ph) +44-20-759-47520 (fax) 47714
Imperial College London,          Dr.Paul.Kins... at (no spam) physics.org
SW7 2BW, United Kingdom.          http://www.qols.ph.ic.ac.uk/~kinsle/

http://www.camfpd.com/technology.htm

Interesting. I designed a system almost identical about 10 years ago
and it was not my original idea. i wonder how valid their patents are?

Hello,
I'm a product design engineer and I'm here today because I'm stumped
by an optics problem, and I hope that someone with more experience can
help me out.
I'm currently trying to educate myself on what practical options are
available to project an image on a surface which is nearly paralell to
the projection direction, as opposed to the ideal, which would of
course be perpendicular.

saluki... at (no spam) gmail.com wrote:
So I'm still reading up on this. Still working on this project today,
and trying to generate a setup which demonstrates the Scheimpflug
principle with regards to a projected image. I understand how the
concept applies to film/camera setups, but in trying to apply the
concept to myprojectionsetup, I'm running into some trouble. Any
suggestions there?

These guys have solved a trickier proble; they doacuteangleprojectioncarefully enough to use sideprojection
into a wedge to give a flat panel display ... once the
rays fail to undego total internal reflection!

http://www.camfpd.com/technology.htm

--
---------------------------------+---------------------------------
Dr. Paul Kinsler
Blackett Laboratory (QOLS) (ph) +44-20-759-47520 (fax) 47714
Imperial College London, Dr.Paul.Kins... at (no spam) physics.org
SW7 2BW, United Kingdom. http://www.qols.ph.ic.ac.uk/~kinsle/