I would think 2 projectors would be stretching it, BUT, who knows, you guys just worked out a miracle. LOL :)
Matt Olieman
Printable View
I would think 2 projectors would be stretching it, BUT, who knows, you guys just worked out a miracle. LOL :)
Matt Olieman
This picture might make it a little clearer on how the collimated screen works for multi projectors. I have been working on this subject with my wife for a very long time and I will try to contribute to this thread.
http://i56.tinypic.com/331hveu.jpg
Joe
Great news and a big congratulations for all the great work and success you have achieved. I am very interested in anything to do with improving visuals, as I have just re-arranged the Airbus and the 3 projectors with a little better resolution and FOV. I am now using 180 degrees on nthe Triplehead with FSX. I have had to back off quite a bit on the graphics as one would expect, using that sort of screen size (3x1280x720), but it is working very well. Can even use YMML from Orbx, but need to taxi at a reasonable pace !!
Keep up the great work...............Brian W.
Hi,
This is a wonderful step in the developement of the perfect home simulator. Now don't shoot me down in flames here, but while the Milar is in the stretched form, is it possible to spray the rear surface with some type of product (Who Knows What) to make the shape permanent. Is there anyway this could be layed on it's back and filled with a product such as foam to form a permanent mould to which could be sprayed with a refective product (Who Know What). These are just thoughts buzzing through my head and feel, as silly as they sound, there might just be something that is useful for further progress.
Kind Regards
Bernie.
Actually, yes.Quote:
Originally Posted by Bernie
US patent #3,549,803 describes making a curved mirror by coating the back of a Mylar mirror with multiple layers of urethane foam and fiber glass.
Hi Mike,Quote:
Originally Posted by Mike.Powell
Does this mean that it can't be done by a hobbyist?
Kind Regards
Bernie.
That particular patent was issued December, 1970. I'd be surprised if it were still in force. You can check it out at http://www.google.com/patents/about?...J&dq=3,549,803
The trick is, how do you apply the foam & fiberglass while the mirror is under vacuum? :)
g.
Well, this has been onmy mind too. All i could come up with is to make a negative mould. then you could create as many mirrors as you want. So, you turn on the vaccum, then apply whatever you want to the outside, and then, when you take this shape away, you will have a master mould, and you can then make the mirror from that. Its the same process as making resin moulds from silicon, only on a much larger scale!
Let me know what you think
Jordan
The trick is that you don't. You flip things over so you can pressurize the Mylar from the reflective side. It means building a totally different frame for holding the Mylar.Quote:
Originally Posted by geneb
http://mycockpit.org/images/collomat...ASA%20rfd7.jpg
This is a 180 degree system (a very old one). This system has 4 projectors and the projectors are out of the sim or rear display compartment. Projecting at a angle downwards. The 5 projector system I mentioned earlier was similar to this one, except it was level D.
Matt Olieman
How curved glass is made:
http://www.youtube.com/watch?v=1tcXCfDzCjY
@Jordan - any mold has to have a surface that is mirror-perfect. That's, really, really hard to do.
Commercial flight simulators use thin film mirrors because they're very light and compared to the same size in glass, dirt cheap to make.
@Matt - was that sim built that way or was it on a motion base at one time? I've never seen a static sim with a collimated display and no base. It's weird. :)
@Mike - Smart aleck. :D
g.
I don't recall this particular one.... :(Quote:
Originally Posted by geneb
I used to travel quite a bit (cross country), I'd try to hit every Flight Sim School there was. I do remember seeing static sims with a collimated display.
Matt Olieman
What about if not sucking air out from backside of mylar, but pushing air in front side.
I think it will make same thing to mylar and you can add something urethane foam etc to backside.
I believe the patent Mike Powell mentioned specifically states that the first layer should be some sort of uniform material (a few coats of plasticized primer should do), as the voids in foam result in an imperfect surface. Otherwise, you've got the idea.Quote:
Originally Posted by makoy
What about applying the material to the rear surface while not being stretched and then using the stretching process to pull it into shape giving the first layer time to set.
Kind Regards
Bernie.
Mylar takes a lot of force to get it into the proper shape. I doubt a single layer on the back would hold the proper mirror shape. I think you would need to apply multiple layers and let them all cure and gain strength if the finished mirror is to be useful.Quote:
Originally Posted by Bernie
Hi Guys ..
Congratulations , this is a historical moment in our hobby !! cheers :D
you made a great step guys , who believed it's possible to make a DIY collimated mirror display system for our replicas !! , So we can modify & discover ways to solve the distortions in the view that appeared in the videos ..
I'm so sorry my english isn't well btw it was so tiring to translate words in your last 169 posts :| .. I have read all posts as i can & I have already got the general idea ... I'm trying to make a frame from wood with a vertically - horizontally FOV 40 - 180 degrees , & I'm going today to a mirror factory to ask them if they can made a mirror that can be formed on the wooden frame form .. but in all your posts you were talking about "Mylar" and you didn't talk about making the mirror at a factory or a workshop .. whatever .. is it expensive to order a mirror by the shape we want ?? or they can't do it in an accurate optical properties ??
another question .. how the vacuum form the mylar ?? please explain because i didn't get how it form the shape of it ..
Thanks for your nice thoughts & posts ..
Did you look at all the videos? Gives you a good idea how this is done and the genius of it all. :) :) :) It's not as simple as to ask a mirror company to build this, there's a bit more to it than just that.Quote:
Originally Posted by SU-Medo
Most of your questions will be answered in this thread. It's quite informative.
Matt Olieman
Thank you Matt for your post ..Quote:
Originally Posted by Matt Olieman
I watched all the videos , but I didn't understand how the vacuum form the Mylar on the frame !!!
but i noticed that the rear-projection screen isn't uniform & a bit distorted .. is this the reason why the picture on the mylar isn't uniform ?? or the reason is that the mylar itself ??
I went to the factory , they can do it .. but the cost made me Deaf !!
Plz tell me the theory of using a vacuum .. ( what it do )
Thanks
@ SU-Medo, The questions you are asking are all answered in this thread. I understand the difficulty with the language barrier, but it's here. Please try to translate using Google translator, perhaps that will help.Quote:
Originally Posted by SU-Medo
Also read Mike Powell's articles related to collimated displays: HERE
Matt Olieman
Quick summary - The frame forms the outline of a a spherical section and when the mylar is properly placed on the frame and you apply a vacuum source to it, it pulls the mylar into the spherical section shape that the framework "describes".
This is what it looks like as a "solid":
http://www.diy-cockpits.org/coll/ima...h40v_model.jpg
As soon as you turn off the vacuum, the mylar "relaxes" and no longer forms the mirror. This is shown especially well in the first video that Wayne posted that shows the first time we applied vacuum to the mirror.
Also, we're not using rear-projection. :)
The majority of the weirdness you see in the videos is the result of the kraft-paper wrapped screen. It will get MUCH better when skinned with material that's not doing its level best to fold, crease, wrinkle and tear itself while being wrapped around the screen framework. :)
g.
Matt .. thank you for Mike Powell's articles , I appreciate that for you .. :):)
Thank you so much for the summary , I got it now .. :)Quote:
Originally Posted by geneb
Whatever the type of projection ( rear or forward ) , I saw a big ball from white acrylic in a shop , It's used for containing a lamp for lighting in streets .. It was big enough ( by guessing it has a radius about 3' feet ) , I think we can cut a part similar to the curvature of the Mylar mirror's frame to use it as a rear/forward-Projection screen .....
I wondered about using something like SU-Medo mentioned (the acrylic white ball).
Rear projection would still be a preference, well, for me, anyway..... :)
I've played around with sheets of Lexan, forming shapes. Lexan is easy to form and takes a little heat, it's also easy to repair when scratched. Anyway, I thought about this when I saw the forming of glass video, with lexan it's the same method, but less heat.
Matt Olieman
Quote:
Originally Posted by Matt Olieman
Here is some info on paints for front projection systems you all might find useful for the collimated projection screen or any DIY front projection system for that matter
http://myplace.frontier.com/~vzepdzs...t%20screen.htm
I'll be using it on my 180 wrap around, not collimated but okay for now. Playing with mirrors geometries to shorten throw distances, idea is to bend a surface into shape with something like Lexan then skin it with a one mil mylar sheet to create a FS mirror. The amount of bending is amazingly little in the vertical, radius of curvature on the order of 15 to 17 feet Haven't tried it yet, but with a conventional mylar mirror was able to cold bend it to produce the following mesh pattern shown,, this is with a VERY rough first order mesh distortion program which will be fine tuned and enhanced to handle localized warping as needed and edge blending
JW
Hi,Quote:
Originally Posted by castle
Here is a link to the use of a curved Sintra sheet and a mirror. The information may be of interest to you if you haven't already seen it.
http://www.biagettis.com/737simproje...roject_008.htm
Kind Regards
Bernie.
Nice work!Quote:
Originally Posted by castle
Yes, I saw that. Thank you.Quote:
Originally Posted by Bernie
Two aspects of that approach I'm trying to improve on. First, using a curved mirror and screen rather than a flat acrylic provides 180 degree with a screen radius of six feet (~2M) This allows you to increase the FOV in the vertical; important for the side view where the side windows are closer plus by placing the mirrors at the edge of the cabin, you reduce or eliminate the problem where the cabin edge might shadow the reflected image, second FS mirrors to eliminate image ghosting can be expensive in larger sizes and customized curved ones can be very expensive. You can create a FS mirror from a conventional mirror by removing the protective backing. It can be done with some care, but that will be a flat surface.
If using a thin mylar skin stretched on a concave surface works, it opens the door for an effective DIY low cost solution and also one to customize a mirror shape. And then create the appropriate distortion mesh designed for the precise arrangement of cameras, mirrors, and screens.
Cheers
JW
Wayne stopped by for a bit today and we were able to get some "pre-saturday" work done. :D
These parts....
http://www.diy-cockpits.org/coll/ima...rame_parts.jpg
Will combine to build this...
http://www.diy-cockpits.org/coll/ima...n_assembly.jpg
The rendering above is the 225 degree screen for the new mirror. The spaces between the ribs will be filled with 2" insulation foam that's been hot-wire cut to shape. The framework is made from 7/16" plywood and the thin ribs are to be cut from 1/8" hardboard. Not shown in the drawing is the supporting structure that will be on the top & bottom of the screen.
It turns out that my original screen design will work out with 56" wide mylar, so the new mirror will cover a vertical FOV of 40 degrees and right around 220 degrees or so horizontally. The diameter remains the same at 8 feet.
The screen as designed is about 52" or so across at the widest point of the base ring. It's a bit wider than that overall because of the ribs - but not large enough to properly do rear projection using a vacu-formed screen.
More tomorrow!
g.
Here's a link that shows what we got done this past Saturday:
http://www.diy-cockpits.org/coll/col...lay_page2.html
Hopefully the server won't fall over again today. (motherboard is fading I think)
It turns out the only active patent we have to worry about covers the "ears" that help form the mirror. I don't know that there is a practical way around that one, but we'll look into it further at some point.
g.
Thanks for the link and pix, Gene. GREAT work!
Thanks Mike. I've been looking into 1200x800 projectors and boy... ka-ching! :(
g.
Yes Gene, thanks for the pics. :) I tried to look for more, but you've got them well hidden :p
Projectors.... I don't know which ones your looking at, but they've come done quite a bit from a few years ago. Yes, I know, still a bunch. I thought they would be coming down quit a bit more by now. :(
I'm amazed at your workmanship..... WOW, you've got the touch!!!! I can't wait for more pics.
Matt Olieman
I saw this yesterday on discovery, did not know that the projector had a mirror between it and the screen that reflects to the vacum mirror.
http://science.discovery.com/videos/...simulator.html
Thanks Matt. I'm looking at the Epson EX7200 right now. They can be had as low as $650, but it's still painful, specially if I need to throw in a new TH2G as well. If it turns out two more GTX260 video cards are cheaper than one TH2G.... :)
I'm going to start the design work on the 225 degree mirror framework this evening. I'm going to be puzzle-piece'ing quit a bit of it. The design needs to be both portable and easily to assemble on-site at Wayne's house. It won't be nearly as deep as the 60 degree frame though. I'm going to design it so that the base is between 2 and 4 inches deeper than the spherical arc in order to keep it as small as possible. The back will be round instead of squared off like it is and there will be a couple of different vac points as well as a panel that I can access the interior with. I'm going to be experimenting with controlling the vacuum by Arduino and a barometric pressure sensor (did I mention this already?!). I'm also going to try a much smaller vacuum pump that has a very small CFM rating (cubic feet per minute) flow, but can draw down a higher vacuum. The Arduino would control the power to the pump via relay as well as a solenoid driven vacuum valve.
g.
Very nice Gene!Quote:
Originally Posted by geneb
Fry's had a sale on two months ago on some Viewsonics PJ-551Ds for $495.
Resolution is 1280x1024, so I picked up two. Should check, but think that is native res and thery are DLPs, still in the boxes. Once I get the mounts and mirrors up and the mesh software working will fire them up. Maybe before Christmas..
Not fair you have all those neat tools like a CNC and cutting table to work with ;-). I'm stuck with a table and band saw and a drill press..
Marvelous work, Gene
JW
Castle, the native resolution of that projector is 1024x768. It make take a high resolution input signal, but it'll scale it down to fit. You'll get the highest image quality if you feed it at its native resolution.
Here's the stats from the Projector Calculator:
http://www.projectorcentral.com/View...ulator-pro.htm
The screen was Wayne's trial by fire with AutoCAD. He did good. :D
g.
Just a quick update - this is what the mirror "solid" looks like;
http://www.diy-cockpits.org/coll/ima...rame_solid.jpg
This will be cut apart into the bits needed to make it into a framework. The top is 4" wide and the gap is about 68" wide. The mirror frame covers 225 degrees.
[Update! I screwed up - the arc should only be 190 degrees.]
g.