Question about collimated display systems.

[Snovadnb]

http://www.youtube.com/watch?v=Wqq4-XC5f8U - You see this!?

[Snovadnb]

http://www.youtube.com/watch?v=Wqq4-XC5f8U - You saw it?

[sacad]

hi everybody ..
i've been trying to keep up with new stuff , and found my self reading the whole thread .. this is the most interesting topic of all
i made a solidworks prototype based on some hypo dimentions ..
proj.JPG

[arnolde]

wledzian, geneb,
It's been a couple months now since your last post, how is the wide mirror doing?
Hope you are continuing with your fantastic success

regards,
Ethan

[geneb]

Sorry guys - take a peek at these videos:

http://www.youtube.com/watch?v=j3FvFx9z2R8
http://www.youtube.com/watch?v=Bx5eb5yTDQw
http://www.youtube.com/watch?v=_cyJIGII6JE

The display system is essentially completed. The vacuum controller has been rebuilt and is working perfectly. The last "construction" that needs to be done involves building a cardboard "blackout cap" to enclose the display.

g.

[arnolde]

Really amazing guys. I wish I'd been as resourceful 5 years ago when I was trying very hard to come up with a wide visual.

Could you post the mechanical dimensions of your mirror frame please?
Also, I wonder if you can see the front projection screen in night scenery? (since it's white, does it reflect ambient light?)

[wledzian]

It does reflect ambient light, but the entire assembly will eventually be capped. Ambient light shouldn't be an issue.

[arnolde]

I was thinking that even when it's fully enclosed, the few bright dots of light that are projected onto the screen at night might stray around enough to cause enough ambient light to make the screen body visible. A Sim engineer I talked to once told me thats one of the reasons that usually back projection is used. Not that that would stop me wanting one I light daylight scenes better anyway!

[wledzian]

rear-projection also has its problems - light from any point on the back side can spill light on the rest of the screen, reducing contrast. The problems you mention are just as much of an issue on first-surface projection as well.

[presse55]

Hello

I'm glad I found this thread. I find this topic fascinating and am working to improve my understanding of the optics involved.

Following wlzedian's earlier post (ex. #58 ), I developed an excel spreadsheet to calculate points on a surface screen for a hypothetical collimated display. Using the resultant points, I am able to develop a very close fitting ellipsoid (major axis is horizontal, center is same as mirror, rotation axis is vertical also through center of sphere).

What I get are parameters (a,b) for the ellipse as shown here

http://www.flickr.com/photos/alfredetlaura/6032314024/

What puzzles me now is:
From the (pilot's) eyepoint, how much vertical and horizontal movement still provides an image reasonably free of distortion? In other words, how can I assess what I will see as my head moves from sided to side, or up and down? Any distortion?


Mike.Powell writes (post #78 ) something I do not understand:
"Once you have a workable configuration, you can ray trace diverging paths from multiple points on the screen surface to check for collimation accuracy."

I'm not sure how to go about doing this 'in 3D'. I can hardly immagine doing these calculations manually. I am under the impression that to answer this question, 3D ray tracing is probably required, but I'm at a loss as to how to go about doing this. I am not familiar with TurboCad.

Any pointers are well appreciated!

Ron,


On a side note:
In astronomical telescope design, a similar problem arises with the so-called tilted component scopes, such as the Schiefspiegler; with these scopes, there are two spheric reflecting surfaces tilted with respect to one another. In telescope design in general, ray traces and spot diagrams are done for two perpendicular planes: the tangential and the sagittal planes. This is well explained in Telescope Optics - Evaluation & Design, Rutten & Venrooij, Wilmann-Bell, 1988.

I'm left wondering if methods used in evaluating tilted component telescope design can be used in collimated displays.