DIY compound curved screen - need help - AVS Forum
Forum Jump: 
Closed Thread
 
Thread Tools
post #1 of 133 Old 12-21-2001, 11:04 AM - Thread Starter
Member
 
stant843's Avatar
 
Join Date: Nov 2001
Location: Kentucky
Posts: 175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I have decided to construct a compound curved screen for my HT. It will have a 16:9 aspect ratio and be 8 to 9 ft wide. My projector is a ceiling mounted Marquee 8501LC. I have searched this forum and am familiar with the TORUS screen. I like the simplicity of the vacuum curved screen. Any design guidelines, Web sites with information, recommended screen materials, etc. would be appreciated. If I receive enough useful information, I will document the design and installation of the screen and post the document on this forum. Thanks
stant843 is offline  
Sponsored Links
Advertisement
 
post #2 of 133 Old 12-22-2001, 08:47 AM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
This has been something I also would love to DIY. There have been several long posts on the idea over the last few years, but no obviously doable solution...brainstorm through the posts you searched out. This was a hot topic about 8-16 months ago on the screen forum.

My screen BTW, would be the same width, and I have a 9500LC. stay in touch, maybe we could figure something out. As it is, I have some great ideas on improving on the single horizontal curve screen in a post by Y-der4me. I'd rather have torus, but at what cost?

My main question about the vacuum method is: How do they get the compound curve exact? Ii seems to me the material must have the proper shape at rest without stretch, and be kept taught with vacuum.

I can't see at all how any vacuum system could accurately stretch a flexible screen material into the shape that is supposedly so complex it uses computer ray tracing to design it.

Question #1 - does anyone know if the proper torus shape is inherent in the screen material itself (ie, if you layed the screen in its frame down flat and let the screen sag...would gravity keep the screen in its proper torus shape), or does the vacuum strecth a flexible screen material into a proper shape?

I just can't see how the "stretch method" could be accurate enough...we converge CRTs to millimeters!

OTOH, making a shaped peice of screen material to such standards seems way beyond anyone without major manufacturing capabilities.

Help us out here if you know ANYTHING...please!

Thanks

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #3 of 133 Old 12-24-2001, 09:23 PM - Thread Starter
Member
 
stant843's Avatar
 
Join Date: Nov 2001
Location: Kentucky
Posts: 175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
This compound curved screen concept is getting me confused. Why does everyone keep refering to the shape as a torus? I am not a videophile but I remember back in elementary school when a science teacher said light reflecting off a surface is focused with a parabolic shape. This is why I call the screens shape a coumpound curve - I do not know what the basic shape is. I sure wish somebody would help clear this up for me.

I am a CRT owner. I feel that CRT's produce the best image. One of their drawbacks is brightness (which limits image size). In my home theater I could use a larger, brighter image. The compound curved screen is the answer to this problem. I feel if we worked on this we could come up with a practical and economical design.

What I would like to design is a general purpose coumpound shaped screen that is adaptable to a wide variety of home theater layouts. I feel that the compound curved shape should be adjustable while viewing an actual video image. the 3 items of adjustment I would like to make are -
1. The sides of the screen probably determine the shape. (At least for that area of the screen) I would like to be able to adjust the 4 curves along the sides.
2. The screen material is fastened along the edge of the screen. The looseness or tightness of the screen material should be finely adjustable while watching a video image.
3. The screen tension from the vacuum should be regulated by a sensor. That sensor should be adjustable.

These 3 adjustments would allow the screen to be focused while viewing a video image. With all the tweaking that goes on with CRT projectors, I do not think this is unreasonable.

Any ideas would benefit all front projector owners.
stant843 is offline  
post #4 of 133 Old 12-25-2001, 02:01 PM
Senior Member
 
Don Stewart's Avatar
 
Join Date: Sep 2000
Location: Stewart Filmscreen Corporation. USA
Posts: 371
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Ok guys...I have been reading Torus Screen post for over a year and have never commented as it is a fairly complicated subject. There have been many comments on single axis curved and compound curved screens that just do not hold any water. I recently read where someone was making a unity gain curved screen. A matte white curved screen will simply cross reflect on to itself and wash out its own black level. The main reason you curve a screen is to take advantage and maximize a particular screen material's gain. In order to determine the proper screen radius, the chosen screen's material gain fall off plot must be accounted for. In the case of a Torus compound curved screen, to put it simply, the goal is to turn every square inch of the screen's surface into one "uniform" hot spot. In order to achieve this, the following
design criteria must be adhered to:

1. Use an angular reflective screen material and know its gain fall off plot. You can find this info at screen manufactures' web sites.
2. Make a scale drawing of your room in Plan view and in Section view. Make sure that your projector and viewer eye points are accurately drawn.
3. You will need to draw a projector and viewer ray trace in section and plan view.

Lets start with the section view. Since most projectors are ceiling mounted, the projector's incident light rays will be offset to the screen's vertical center line. This means that the top of the screen will have a shorter radius than the bottom. (See example drawing below). Now in order to determine the screen's vertical curve, you want to angle the screen's plane, at any given point, to be one half of the angle of the ray trace between the projectors lens and viewers eye point. Plot this angle at least at 5 points. (top, bottom, center, center top and center bottom). You will find that the screens equator will be offset or closer to the bottom of the screen.
When doing the ray trace in plan view, you will have to accommodate a wider horizontal spread of light than needed in the vertical plane. Pick the center sweet seat and also the seat that is most off screen's center line. Make your projector's and viewer's ray trace for both seats and average the two different angles to determine horizontal radius. (Now you can see why we have developed software for this time consuming task).
Once the screen's top, bottom and side radiuses have been determined, a screen frame box can then be constructed. The box can be made out of plywood with the proper radiuses cut into top, bottom and sides leading edges. The screen frame box is then attached to the wall at the proper design height. The box should be caulked with a sealer to make it near air tight. A small low voltage pancake fan in a hush box is all that is required to draw the screen back ( less than 1/8 of an inch water column) to its design point. A proximity sensor is placed on a bracket at the screen's equator. The depth of the sensor in the box is at the intersection point of all three radiuses. The sensor switches a solid state relay that regulates the fan. The screen will always be seeking and searching for its design point as the vacuum turns off and on. With the proper proximity sensor, the screen's pumping action is less than one mm. We have never had a complaint on the gradual screen movement. Finally, the screen fabric chosen must have elasticity to it. Draper, Dalite, and Stewart all make screens that do. As a guide, pick a fabric that is available with tab tensioning on a roller screen. This type of screen material is unsupported and will have some stretch to it. The screen fabric must be cut to match the curved arc dimensions of the top, bottom and sides of the screen frame box. Also allow a few extra inches in each dimension so that the screen material can wrap around the sides of the box and attach to the plywood with appropriate fasteners. This wrap around will also make the screen air tight on the front edge of the plywood. Remember...a Torus screen is very unforgiving. Due to its gain and white field uniformity any imperfection on or in the viewing surface will stick out like a sore thumb. In other words, the screen fabric needs to be free of streaking, mottle, dirt, etc.
This is the one and only post that I will make regarding DIY curved screens. Please do not follow up with questions as I will not be able to answer them. BTW..Stewart Filmscreen holds over 20 patents regarding Torus screens. The law allows you to build your own for your own use. If done correctly you will have a great screen. It is the best way I know to get those extra foot lamberts for CRT projection. Good Luck.

Regards,
Don

Sorry... the example drawing that I wanted to include will not post as the message says the file is to big.
Don Stewart is offline  
post #5 of 133 Old 12-26-2001, 08:20 AM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks Don, We really appreciate that! Most of us here are basically just guys trying to get the best we can out of what can be a REALLY expensive hobby.

I guess some would like to start their own busines, but many (myself included) just like tinkering and DIY projects. Its fun, rewarding, and it gives us a chance for cutting edge improvements that we couldn't otherwise afford. Even if we don't do a "professional" job, its still something we made with our own two hands...and sometimes it looks damn good!

Merry Christmas.


Now, if you HAD a torus we could afford...I would be VERY willing to forgo my personal sense of achievement and just sit and watch movies!

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #6 of 133 Old 12-26-2001, 09:36 AM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Now, to all the non-pros out there...Lets assume we use a 2.0-2.5 gain Stewart/Draoer/etc material as suggested....follow me here and correct me if I'm wrong...

The worst hot spot with the high gain material would be with NO curve at all. That's the reason we're doing this. Alan now has a single curve that eliminates much of the problem, so even if we didn't get the ray tracing 100% right, a well made (home made) torus would still be MUCH more uniform than with no curve.

So it appears its the radius of the frame that dictates the shape of what is essentially flat but strechy material. Since the tracing tries to cover a seating AREA, not a point, even the computerized ray tracing must be an average! Straying slihjtly from this average may cause slight changes here and there in the seating areas, but unless you made a huge and sudden change in your radius, I don't think it would be noticed during viewing.

This leads me to believe again that a slight error in frame radius would still be very good, just not as good as if you had Don's resources.

Since he is using very low pressures, the stretch on the screen material must be minimum...only being needed to allow the material to ""average out" the horizontal and vertical curves of the frame.

Either plywood or plastic could make the frame. It could also have a laminate back to make it airtight and still be flexible in placement options (ie a movable lightwieght box).

The two areas I need further iseas on are the proximity switch and mounting the material to the frame. Here a slight error in pulling the screen too tight at some point on the radius would make a bowing in at the center of the screen.

For now, I assume the flat screen shape ( ie if you lay it flat on the ground) is rectangular, and the Final screen shape comes solely from the frame and vacuum. (ie fix the strsight edge of material to the curved frame on all 4 sides.

I don't want to give up on this idea guys...its got to be easier than designing a HTPC, and look at what several members have done over the years in that field!

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #7 of 133 Old 12-26-2001, 08:39 PM
Advanced Member
 
KFung's Avatar
 
Join Date: Apr 2000
Location: Ottawa, Ontario, Canada
Posts: 879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
You do want to be careful though, I gather most people who would attempt this kind of screen are working with a CRT. By the nature of it's short throw and 3 lenses it is one of the most challenging cases, ironic since a CRT would likely benefit the most from a TORUS.

The short throw means that would will probably have a fairly tight radius for your curves (i.e. smaller radii) to direct the light back to the audience. The 3 separate guns means that you will probably need pretty good calculations/drawings for the curves to prevent colour-shifting, you also wouldn't want to push your design too hard with too wide a seating area (having the first row very close and wider than the screen). The optimum would obviously be a single seat, vertical risers will usually not present and difficulties, but as above you don't want too wide of a seating area (3-4 seats maybe? That's just a guess). Errors in calculations are probably OK, but you will definitely want the curves you cut for the sides to be smooth and even, bumps in the sides would show up and propagate along the surface of the screen a little. As Jeff says, changes in radius along the curve are probably OK, but you will want it to be a smooth transition. Also precise focus will be an issue because of the short throw (which reduces depth of focus) and the increased curvature. I'm not to familiar with the optics in CRT's, but I'm fairly certain that some models have very robust focusing options that let you focus the corners separately from the centre, this would be very useful in getting optimum results. I think Peter/Cineramax had some problems with this in his previous installs.

Regarding Jeff's comments on using laminate for the back, you will also have to brace it so that it remains undistorted under pressure because you will likely be mounting the sensor at this point.

Jeff says it right though, if you can live with the image that 2.0 gain screen produces (with hotspotting and colour-shifting, visible or not), a curved screen will almost certainly be an improvement. Good craftsmanship will help a lot in the end result, though.

Good point on mounting the screen material, I don't know how much of an effect it will have in the final product, but it's something to keep in mind.

Regards,

Kam Fung
KFung is offline  
post #8 of 133 Old 12-26-2001, 09:44 PM
AVS Special Member
 
deronmoped's Avatar
 
Join Date: Apr 2001
Location: San Diego, Ca.
Posts: 3,483
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I was thinking of building a compound curved screen too. I believe the vacuum method is going to be the way to go.

There are four concerns I have though.

1) Will the screen material be smooth in the corners when it is stretched?

2) Is there screen material wide enough without a seam, at say 2.5 gain?

3) How fast Will the screen material lose its resilience?

4) Is it worth it?

The forth reason is the biggest problem for me, as I have already constructed a 161" 16X9 Silver painted curved screen. The gain is at 2.5. And I'm real happy with the results.

With all the work it took me to get this screen right, I think it would have been easier to do a vacuum screen.

Deron

CRT Videophile
deronmoped is offline  
post #9 of 133 Old 12-27-2001, 05:03 AM
Senior Member
 
Iceman's Avatar
 
Join Date: Oct 2000
Location: Stockholm, Sweden
Posts: 439
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Small differences from the theoretically optimal shape will not be a problem in practice. By using the "vacuum method" the shape will always differ slightly from optimal, and while differences like these easily show up in calculations and perhaps also on white fields, with real-life material there is no way for a sane person to notice them.
Iceman is offline  
post #10 of 133 Old 12-27-2001, 04:52 PM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
That's my point exactly! No it won't be easy...but what about this hobby that's really cutting edge (and worth it) is?

PS - Alan has a new Stewart made curved (not torus) screen, silver, 2.5 gain...and he says its the best he's seen with NO hot-spotting or color shift.

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #11 of 133 Old 12-27-2001, 06:09 PM
Senior Member
 
Don Stewart's Avatar
 
Join Date: Sep 2000
Location: Stewart Filmscreen Corporation. USA
Posts: 371
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally posted by Jeff Smith


PS - Alan has a new Stewart made curved (not torus) screen, silver, 2.5 gain...and he says its the best he's seen with NO hot-spotting or color shift. [/b]

I think its more like 3 gain plus.
Don
Don Stewart is offline  
post #12 of 133 Old 12-27-2001, 09:51 PM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Don,

Alan was quoting the 2.5 from the Stewart "owners manual",,,but from what he described, it had to be a high gain....regardless, it sounds good. He spoke of the silver as being somewhat delicate and fragile, he also mentioned he was going to try to get a 2.0 for the same screen frame to compare.

Any comments based on experience with this curved frame and the two materials? Also, are they available yet...even for beta testers? I'd be using a 9600LC.

Believe me, I really want to be able to just buy something...but then there's the price tag to deal with...

If I can get even 1/2 the benefit of a torus screen, that would be worth consideration.

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #13 of 133 Old 12-27-2001, 11:52 PM
AVS Addicted Member
 
CINERAMAX's Avatar
 
Join Date: Sep 2000
Location: From Coconut Grove FL to the rest of the globe with TLC.
Posts: 12,738
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 243 Post(s)
Liked: 78
CINERAMAX is online now  
post #14 of 133 Old 12-27-2001, 11:54 PM
AVS Addicted Member
 
CINERAMAX's Avatar
 
Join Date: Sep 2000
Location: From Coconut Grove FL to the rest of the globe with TLC.
Posts: 12,738
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 243 Post(s)
Liked: 78
CINERAMAX is online now  
post #15 of 133 Old 12-28-2001, 09:30 AM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Peter,

If someone sits outside your indicated cone of viewing is the hot spotting better or worse than a lesser gain non-torus screen? Granted its not recomended, but unavoidable at Super Bowl parties.

Can the rank and file, working class, used 9500LC owner get one from you reasonably affordably... like about9' wide?

PS - to Don and Peter, I'm a doctor...I promise not to go into this business...I just like tinkering, DIY, and quality toys.

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #16 of 133 Old 12-28-2001, 12:36 PM - Thread Starter
Member
 
stant843's Avatar
 
Join Date: Nov 2001
Location: Kentucky
Posts: 175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
The information you supplied was very useful to me. I am completely new to projectors and home theater. I finished a room in my basement and decided to turn it into a home theater. I have been reading everything on this forum for about 6 weeks and am putting together a HTPC for my Marquee 8501LC. To me there are 3 things in the video chain. The source, the projector and the screen. As I piece together my home theater I am beginning to realize that the screen is just as important as the video source and the projector. I have recently realized that while watching the projected image a viewer spends all their time looking at the screen. What is interesting to me is how some people will spend all their resources on the source and projector and completely neglect the screen.

Anyway, the reason I am writing this is because I have decided to use your screen material in my DIY compound curved screen. I have looked at Da-Lite, Draper, etc. I am sure they make a good product but I have been able to get more technical information on the optical properties of your screen materials. That gives me confidence that if I use your material, the finished screen will perform optically as designed.

I know you said you will not respond any more about this subject but please just give me some specific information. Could you list the screen materials (fabrics) you make that are ok to use with this type of screen. The only material property I am interested in is the one concerning stretch in the vacuum box. I am uncertain on this point. When I order your material or speak with a salesman, what should I be asking for?
stant843 is offline  
post #17 of 133 Old 12-28-2001, 12:40 PM
AVS Addicted Member
 
CINERAMAX's Avatar
 
Join Date: Sep 2000
Location: From Coconut Grove FL to the rest of the globe with TLC.
Posts: 12,738
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 243 Post(s)
Liked: 78
To discuss the possibility of a TORUS kit...
CINERAMAX is online now  
post #18 of 133 Old 12-28-2001, 03:44 PM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Stant843.

I'd love to talk with you and swap suggestions. I have thought this out quite a bit as well and spoken to others about it.

My email is jeffandpamsmith@home.com. I'd like to find a time to call you and brainstorm.

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #19 of 133 Old 12-28-2001, 07:07 PM
AVS Addicted Member
 
CINERAMAX's Avatar
 
Join Date: Sep 2000
Location: From Coconut Grove FL to the rest of the globe with TLC.
Posts: 12,738
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 243 Post(s)
Liked: 78
I suspect you should talk to other marquee owners and get a power buy on the ray trace, the material, the TORUS EXTRUSIONS, and a proper electrical design.

Such a power buy goes thru AVS.

Why get a DEFLATO when you can have a TORUS.
CINERAMAX is online now  
post #20 of 133 Old 12-28-2001, 08:32 PM
TEW
AVS Special Member
 
TEW's Avatar
 
Join Date: Aug 1999
Posts: 1,447
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Another idea for a DYI Torus???

Would a large satelite dish, say 12ft in diameter or so, provide a reasonable mold to construct a Torus screen using a fiberglass chop gun?
You could mask off the top and bottom quarters of the circle, fiberglassing only the center area from side to side.
Would this yield anything closely resembling a Torus shape?

TEW is offline  
post #21 of 133 Old 12-29-2001, 09:08 AM
TEW
AVS Special Member
 
TEW's Avatar
 
Join Date: Aug 1999
Posts: 1,447
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Or . . . .
How about a large balloon as a mold?
I've seen ads for military surplus balloons
that are up to 10 or 12 feet in diameter, and they are pretty cheap too.
I would assume these large baloons would be strong enough to support a gel-coat being painted on the top cross section of the balloon. The size or diameter of the balloon would dictate the with of your screen. You would need a balloon with a width greater than the screen size you want because you couldn't mold your screen in a full half circle. But you might be able to get an 8 or 10 ft screen from a 12 ft balloon.
The vertical size of your screen would be determined by the swath of gelcoat you apply in the vertical, so you would simply make that the correct dimension to give you a 16:9 screen size relative to the width.
Sounds pretty easy to me. Just apply a release agent to the surface of the balloon, paint on a good gel coat over a cross section of the baloon, let it dry, and use hand laid fiberglass mat and resin over the dried gelcoat.
How would you deterine the right size balloon radius to get the screen size you wanted?
Any thoughts?

TEW is offline  
post #22 of 133 Old 12-29-2001, 10:11 AM
AVS Special Member
 
Jeff Smith's Avatar
 
Join Date: Apr 1999
Location: On a large overstuffed couch in Virginia Beach, Va
Posts: 1,203
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I originally thought of a solid screen too. I think the radius would be different enough from either a balloon or dish that if you were going the solid route, the best bet would be to make the screen as we're talking above (vacuum method), then just lay it flat on the ground and use that as your mold for fiberglass.

I would want it to be light enough to handle, but sturdy enough to keep its shape. If your idea works, Tim, it should work as well with a "proper" shape (as per Don and Peter). The material in the "mold screen" could be chosen for shape characteristics only without regard to gain...ie cheaper, less stretch.

Once you have a mold, you could use that to form the "real screen". Perhaps even using standard high gain screen material laying face down over the initial mold, then smoothed carefully with proper stretch as needed, and sprayed from behind with the fiberglass to give support. Boats are done in this fashion, with the gelcoat (or finish coat) going on first.

If this works, it would be much easier to temporarily set the final screen material accurately in shape over a mold, and THEN make it fixed permanently into that shape ...rather than making the shape frame first, then trying to glue the viewing surface (the critical one) to the support surface without blebs, wrinkles etc.

I once was against the vacuum idea, but think it could actually be much more lightweight than fiberglass...even with a box frame. I do like the fact that once the fiberglass is set, the shape is fixed. OTOH with the vacuum idea, you could re-cut your frame sides and fine tune your curve.

The other problem with a solid screen is what to use for the surface. Gluing a standard high gain material to it would be very hard to do smoothly with the compound curve. And you would want a high gain...that's the whole point of torus. If KBK could come up with a 2-3 gain paint with good even characteristics then painting would definitely be best.

Todays homework, class, is to find out if a fiberglass or plastic material can be laid over the back surface of any commercial high gain screen material without damaging it. If resin doesn't
eat the material, I think the application would be smoother and more blemish free than gluing to a compound curve shape.

Next, everyone email Ken Hotte until he thinks of a suitable paint.

THEN, we have to pick a size that several of us can use!!!

I like Peters idea of getting info from him or Sigma...but really, if ANYONE with a home theater and high end CRT has already bought this from Sigma, etc., then the numbers would generally be usable to most of us...they wouldn't even have to start over.

They could sell us the specs from Jonathan $. Doe Esquire III, who has ABOUT the same screen and seating area size, and probably wouldn't care if we used the info if we just asked politely.


(God I hope the server isn't down after I typed all this!)

Thanks,

Jeff

Finally went digital: RS20, ISCO IIIs, DIY 10.5' wide torus screen, Stewart StudioTek 1.3 G3 with 4 way masking and adjustable image size
Jeff Smith is offline  
post #23 of 133 Old 12-29-2001, 11:09 AM
AVS Special Member
 
deronmoped's Avatar
 
Join Date: Apr 2001
Location: San Diego, Ca.
Posts: 3,483
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
If you are going to make a screen off a mold I would think the diameter would have to be on the order of 35-40 feet! I used a 18 foot radius on my screen and that is quite a radical curve.

As for the finished surface, it has to be perfect with higher gains, as any variations will show up as light and dark areas.

If you use a silver paint it will give you a gain of about 2.5. To get it to go on even make yourself a brush as big as the screen. I did it and it works great.

The Man with the worlds biggest paint brush!

Deron

CRT Videophile
deronmoped is offline  
post #24 of 133 Old 12-29-2001, 11:28 AM
TEW
AVS Special Member
 
TEW's Avatar
 
Join Date: Aug 1999
Posts: 1,447
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Jeff
I was thinking a good polished gel coat might provide a sufficient screen surface itself.
Or, maybe KB's paint would do the job.

I am not familiar with the "vaccum method" discussed here.
Can you sum it up in a few sentences?

TEW is offline  
post #25 of 133 Old 12-29-2001, 02:35 PM - Thread Starter
Member
 
stant843's Avatar
 
Join Date: Nov 2001
Location: Kentucky
Posts: 175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
A frame is built and mounted to the wall. The frame forms a box and it is made airtight with some caulking. There is a small fan (like a computer cooling fan) mounted in the frame. The fan blows the air out of the box. Then screen fabric is loosely fastened around the box. When the fan is operated, the screen material is sucked back into the box. This gives the screen its compound curved concave shape. The vacuum method is desirable because the shape of the screen can be changed quite easily.

I do not know what basic shape is actually being formed. Does anybody know what the shape is? For example is it a section of a torus or is it a parabola or is it something else? This is an important detail. As you all know, a parabola will precisely focus parallel rays going into it whereas a spherical shape will inprecisely focus rays coming from all directions. What is best considering the screen is not a perfect mirror?

As Don indicated above, The compound shape consists of three curves. The upper vertical, the lower vertical and the horizontal. Also, the upper and lower vertical curves do not necessarally meet in the middle of the screen.

If we could determine the basic shape of the screen, it would be a simple matter to write a computer program that would calculate the details based on projector position, screen position and seating arrangement.
stant843 is offline  
post #26 of 133 Old 12-29-2001, 03:49 PM
Advanced Member
 
KFung's Avatar
 
Join Date: Apr 2000
Location: Ottawa, Ontario, Canada
Posts: 879
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Stant,

By virtue of the name (TORUS) I think we can surmise that it is a section of a torus. This would imply that there are only two radii associated with the screen (the horizontal and the vertical). When you draw it out, the vertical dimension is probably the easiest of the two because the heads are clustered on nearly linearly (see Peter's pictures). As Iceman mentioned this would optimally be very close to a parabolic curve but a straight radius would probably be close enough. The vertical curve would most practically be a single radius.

The two horizontal radii are the ones that are different. Conceptually one would construct a torus that would focus the light from the projector to the audience, then project the viewable image area from the projector onto this torus. The section formed by the viewable image projected onto the torus would be the shape of screen that you would want. Because of the way most front projectors are setup you would find that the bottom of the screen would be closer to the audience and the top farther, the radius on the bottom edge would be subtly different than the top.

TEW, et al.

I believe someone has made a curved screen using fibre-reinforced resin (Mike2, if I recall correctly). I don't know how practical a balloon would be, it would be difficult to find/make one with the correct curvature. Mike make a plywood mold and hand-laid the fibreglass with the resin (unsaturated polyester, I believe). I don't recall for sure, but I believe he then bonded a commercial screen surface to the fibreglass.

It would be very practical to construct a rigid mold and prestretch the fabric over the mold before putting on a gel coat and laying in the fibre. I don't have any direct experience with the process, but the gel coat should probably adhear well enough to the fabric to prevent any delamination later on.

Also, I don't believe epoxy-type resins will unduly affect the screen material, but I can't be sure.

A gel coat should provide a smooth enough surface to serve as a base for a screen surface, but it won't be suitable as the final surface without a coat of paint. You would need a high gain paint of some kind, but more importantly you would need to determine the gain curve to optimise the shape of your screen. KBK may have the proper instrumentation to do this if he has a high gain grade of paint, but most of us will not.

You don't need to "buy" a design if you don't want to, Don has been generous enough to give us enough information to determine the radii ourselves.

Regards,

Kam Fung
KFung is offline  
post #27 of 133 Old 12-30-2001, 05:35 PM - Thread Starter
Member
 
stant843's Avatar
 
Join Date: Nov 2001
Location: Kentucky
Posts: 175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I've reread Don Stewarts post and perhaps the basic shape is not really important. We should just consider the screen as a lot of little mirrors. The mirrors should be positioned so that they reflect the image to the prime (middle?) viewing area. We can divide the screen into as many little mirrors as we want, then using a computer program we could calculate the orientation of each imaginary mirror. Assume our screens have a 9 x 16 aspect ratio. If one inch centers are acceptable then a Visual Basic macro could fill in the cells of an Microsoft Excel spreadsheet with numbers to accomodate a screen as large as 256" by 455". The numbers (lets assume a unit of millimeters rounded to the nearest millimeter) would be a depth from the viewing surface. Some point in the proximity of the middle of the screen would be 0. As we progress to the edges of the screen the numbers will increase. As a whole, this will describe the curvature of the screen. Because the screen has an optical diffusion (I am not sure of the correct term) the other seats in the viewing area should see a good image. I feel if we work together we can develop such a program. With some input I would be willing to write the application and post to this forum.

I get the impression that contributers to this thread do not trust the vacuum method to provide the correct compound shape for the screen. I like the idea of having a compound shaping method that is a little more deterministic as well.

I was walking around Home Depot pricing the materials for a new deck when I had an idea. It may be tedious but what if we glued some peg board to some underlayment. This would establish the basic screen size. The pegboard could be used as a grid. Then pieces of dowel of the appropriate length could be cut and glued into the pegboard surface. This would be the screen size and approximate curve we need, adapted for each home theater according to the spreadsheet calculations.

If some of these ideas seem OK, perhaps someone could contribute some ideas to make this a reality. For example, what can be laid over the pegs to take on the continuous shape we need? What screen fabric should we attach to this surface? How will we attach it? Does Ken have a ScreenGoo that we could employ? The pegs (there could be thousands of them) what is the quickest or easiest way to get them fabricated? The vacuum box is not dead - a vacuum could still be used but the pegs would provide a more accurate position or stop for the screen material.

All comments or ideas are appreciated. I am confident we will develope a practical and economical compound curved screen design that all members of this forum can take advantage of.
stant843 is offline  
post #28 of 133 Old 12-30-2001, 06:54 PM
TEW
AVS Special Member
 
TEW's Avatar
 
Join Date: Aug 1999
Posts: 1,447
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
The peg board idea sounds really hard. If I'm understanding you, the pegs would have to each be cut to the right lenght AND the right angle in order to glue flat to the screen surface.

As for the vaccum box, I'm still having some difficulties visualizing how this is supposed to work. I dont' see how a flexible screen material could be expected to draw in at a curveature when the vaccum is applied. What am I missing?

I think some solution using hand laid fiberglass will be the easiest way to do this.

How exacting will these calculations on the curvature have to be to get good results?
I have been assuming this is basically a rounded inside surface, much like the inside cross section of a balloon.
If that's the case, you would need to determine the depth to the center of the screen, as measured from the outer edges, both in the horizontal and vertical, to facilite the optimum reflectiveness given the viewing area you were focusing the screen at. Once you determine that, wouldn't the resulting natural curve of the screen surface be what you were aiming for?

TEW is offline  
post #29 of 133 Old 12-30-2001, 08:32 PM - Thread Starter
Member
 
stant843's Avatar
 
Join Date: Nov 2001
Location: Kentucky
Posts: 175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
TEW:

The idea of using pegs is not hard, it is tedious. I have no problem with the idea of hand laid fiberglass. I think it is a great idea.

But let's back up, how do you propose we construct a form for the fiberglass with the compound curve that we need. It is reasonable to assume that each home theater requires a slightly different screen shape. Do you know how to easily produce a form for these different shapes? Please share your ideas with us.

Calculating the shape we need is not the problem. The real problem is how to reproduce the compound curved shape from the calculations.

Thanks for any suggestions.
stant843 is offline  
post #30 of 133 Old 12-31-2001, 02:15 AM
TEW
AVS Special Member
 
TEW's Avatar
 
Join Date: Aug 1999
Posts: 1,447
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Well, that's the hard part. Any method of building a mold with curves both in the H and V is going to be difficult. A universal mold will be all but impossible because many of us will need different sizes. I'd like to brainstorm this and come up with a METHOD everyone can copy without too much difficulty.
What did you think about using the cross section of a satelite dish as a mold?

TEW is offline  
Closed Thread Screens

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off