wledzian
04-26-2010, 04:48 PM
... and salutations!
I originally got into flight simming back in the days of Falcon 3.0 on a 486DX2-66, and grew along with the MSFS series. As I was never able to afford the latest-n-greatest PC technology, I usually flew head-down IFR-type flights in cessnas and the like, with the graphics cranked to minimum to keep the frame rate up.
When I finally entered the working world and started bringing home a respectable paycheck in mid-2000, I got my private pilot's license. On the intro flight, the instructor handled the takeoff, communication and navigation, but I had the controls for the rest of the flight, including a few touch-and-go's and the final landing (the instructor was ready with his hands by the yoke for the landings, but claims he never touched it). He seemed surprised when I told him that my only experience was a few too many hours with MSFS.
When life got busy and things got more expensive after 9/11, my license started gathering dust, and is now at least 6 years out of current. I've started simming again recently to get my procedures back in line before I make any serious attempt to get back to real-life flying.
One of my main concerns is pattern activity, something I've always felt a bit lacking in the sim with just a forward view making the turn timing a bit on the iffy side. And so... on to why I'm here.
I've tried looking around with the POV hat, but I feel it detracts too much from the sense of immersion.
I've tried the multi-monitor approach, but it never feels quite right.
I've tried FreeTrack, but it makes me motion-sick.
Which brings me to true wide field-of-vision displays.
I feel like if I'm going to do it, I want to do it right.
I want a full 180-by-30 degree mirror-collimated display (10 up, 20 down forward, 0 up 30 down to the side).
I realize that this is ambitious, and that I may be jumping in the deep end with a pair of lead boots, so I'm looking for advice from anyone who may have tried something similar. Whether it worked well or failed miserably, I'd like to learn from your experience.
I originally found this site through this thread:
http://www.mycockpit.org/forums/showthread.php/1121-Brainstorming-for-a-collimated-mirror-display
While a couple people looked like they were taking a shot, the thread died out before anything came to fruition. I took some time to read through several patents describing methods to form an acceptable spherical mirror, some of which briefly mention the proper screen shape to go along with it. Referencing those patents against the discussion in the above thread and with some research into the optical properties of spherical mirrors (particularly where off-axis spherical aberration is concerned), I think I may be able to pull it off.
Some things I've learned through research:
- Mylar does not have uniform properties in the bottom-end of the stress-strain curve, and needs to be pre-stressed beyond this range in order to get even curvature at the fixed edge of the mirror frame.
- While the nominal focal sufrace of a spherical mirror is at R/2, this does not apply for significant off-axis viewing positions. As the eyepoint moves further off center, the focal surface moves outward.
- A centered spherical screen with minimum curvature to properly form an image across the entire field of view will only be correct for one elevation, typically close to horizontal. The screen is closer to the mirror than necessary at points above and below the horizon, resulting in an image which will have convergence progressively closer as you move away from the horizon. This is not necessarily bad, as it will make the ground appear at less-than-infinity when the horizon is at infinity.
- An aspherical screen can be constructed with a geometry which will result in more of the image formed at or near infinity. The correct geometry has the side-benefit of permitting a wider vertical field-of-view.
- for a mirror roughly 60" in radius (as big as I can fit in my office without the wife saying no), the difference in screen position to form an image at infinity differs from a position which forms an image at 50 feet differs by only 3/4". Too close, and I might as well just use a direct projection onto a cylindrical screen. Too far, and the image doesn't form.
So...
Before I commit too much time and money to a doomed project, I'd like to learn from your experience.
I know that a few of you are also located in western Washington state, and I'd love to get in touch. I'm about 25 miles south of Seattle - if you're somewhere in the area, feel free to drop me a message directly!
Thanks!
I originally got into flight simming back in the days of Falcon 3.0 on a 486DX2-66, and grew along with the MSFS series. As I was never able to afford the latest-n-greatest PC technology, I usually flew head-down IFR-type flights in cessnas and the like, with the graphics cranked to minimum to keep the frame rate up.
When I finally entered the working world and started bringing home a respectable paycheck in mid-2000, I got my private pilot's license. On the intro flight, the instructor handled the takeoff, communication and navigation, but I had the controls for the rest of the flight, including a few touch-and-go's and the final landing (the instructor was ready with his hands by the yoke for the landings, but claims he never touched it). He seemed surprised when I told him that my only experience was a few too many hours with MSFS.
When life got busy and things got more expensive after 9/11, my license started gathering dust, and is now at least 6 years out of current. I've started simming again recently to get my procedures back in line before I make any serious attempt to get back to real-life flying.
One of my main concerns is pattern activity, something I've always felt a bit lacking in the sim with just a forward view making the turn timing a bit on the iffy side. And so... on to why I'm here.
I've tried looking around with the POV hat, but I feel it detracts too much from the sense of immersion.
I've tried the multi-monitor approach, but it never feels quite right.
I've tried FreeTrack, but it makes me motion-sick.
Which brings me to true wide field-of-vision displays.
I feel like if I'm going to do it, I want to do it right.
I want a full 180-by-30 degree mirror-collimated display (10 up, 20 down forward, 0 up 30 down to the side).
I realize that this is ambitious, and that I may be jumping in the deep end with a pair of lead boots, so I'm looking for advice from anyone who may have tried something similar. Whether it worked well or failed miserably, I'd like to learn from your experience.
I originally found this site through this thread:
http://www.mycockpit.org/forums/showthread.php/1121-Brainstorming-for-a-collimated-mirror-display
While a couple people looked like they were taking a shot, the thread died out before anything came to fruition. I took some time to read through several patents describing methods to form an acceptable spherical mirror, some of which briefly mention the proper screen shape to go along with it. Referencing those patents against the discussion in the above thread and with some research into the optical properties of spherical mirrors (particularly where off-axis spherical aberration is concerned), I think I may be able to pull it off.
Some things I've learned through research:
- Mylar does not have uniform properties in the bottom-end of the stress-strain curve, and needs to be pre-stressed beyond this range in order to get even curvature at the fixed edge of the mirror frame.
- While the nominal focal sufrace of a spherical mirror is at R/2, this does not apply for significant off-axis viewing positions. As the eyepoint moves further off center, the focal surface moves outward.
- A centered spherical screen with minimum curvature to properly form an image across the entire field of view will only be correct for one elevation, typically close to horizontal. The screen is closer to the mirror than necessary at points above and below the horizon, resulting in an image which will have convergence progressively closer as you move away from the horizon. This is not necessarily bad, as it will make the ground appear at less-than-infinity when the horizon is at infinity.
- An aspherical screen can be constructed with a geometry which will result in more of the image formed at or near infinity. The correct geometry has the side-benefit of permitting a wider vertical field-of-view.
- for a mirror roughly 60" in radius (as big as I can fit in my office without the wife saying no), the difference in screen position to form an image at infinity differs from a position which forms an image at 50 feet differs by only 3/4". Too close, and I might as well just use a direct projection onto a cylindrical screen. Too far, and the image doesn't form.
So...
Before I commit too much time and money to a doomed project, I'd like to learn from your experience.
I know that a few of you are also located in western Washington state, and I'd love to get in touch. I'm about 25 miles south of Seattle - if you're somewhere in the area, feel free to drop me a message directly!
Thanks!