Film-Tech Forum ARCHIVE: Screen frame idea

This is topic Screen frame idea in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=005753

Posted by John Walsh (Member # 168) on 01-13-2004, 10:48 PM:

I had an idea for a screen frame; I was wondering if anyone ever tried anything like this before.... although it seems just about everything has been tried in this business at one time or another....

The idea is to curve the screen frame differently at the top than the bottom to match the throw. Since the throw is usually longer at the bottom than the top, this 'corrects' for the required curve.The frame would be designed to accept a regular, rectangular shaped screen, and masking would trim off all sides.

I may have a job that has a big screen, short throw, and the projector is located up high, and thought I'd experiment.

Posted by Gordon McLeod (Member # 33) on 01-13-2004, 10:55 PM:

the cinespheres screen does that as well as having a verticaly curved top member

Posted by Edward Jurich (Member # 1787) on 01-14-2004, 11:15 AM:

You may wind up with some distortion in the center portion using a standard screen. The top, bottom and edges of the screen will follow the curved shape of the frame but the center will tend to flatten itself. I once tried tilting a 40 X 22 foot screen back on a steep throw. After awhile the frame started to bend. If you're going to tilt then you need a sturdy frame.

Posted by Tim Reed (Member # 137) on 01-16-2004, 08:44 AM:

That's a very good idea, John! Maybe there's a way to keep it from flattening in the center, I can certainly visualize that happening. I guess it would have to be made to relatively close tolerances.

Posted by John Hawkinson (Member # 1135) on 01-16-2004, 09:34 AM:

Preface: I don't know a whole lot about this, so this is speculation.

I would expect that if you tried to curve the top differently from the bottom, you would get worse than a center flat spot -- you'd get wrinkles or something ugly (and maybe an increased tendancy to rip) at point where the two curves come together.
Example: try doing this with a piece of paper while keeping the paper square.

Question: Why can't you just curve the screen uniformly and then tilt the screen forwards to compensate for the vertical keystoning? Is that too imperfect such that this solution would give a better results?

--jhawk

Posted by Dave Macaulay (Member # 813) on 01-16-2004, 10:45 PM:

A modern solid vinyl screen would have no problem stretching to fit the frame without wrinkles. I don't think one can even still get the ancient cloth-backed screens, I've removed a few and hope I never have to again... they weigh a LOT and after the 50 years or so they must have been up there they were encrusted with some very nasty dust and smoke residue.

Posted by John Walsh (Member # 168) on 01-17-2004, 08:42 AM:

I think what everyone means is the screen would 'bow' probably in the lower center something like this:

The Cinespheres is an IMAX screen, right? I wasn't thinking of something that big; only about 35-45 ft wide.

I'm guessing of course, but I was thinking that any 'bowing' problem would be more than offset by the overall reduced distortion. We did a theater recently where we followed the Schneider suggestions for curving the screen. The result was just what you'd expect; at one place a projected line looks straight, but above/below that point, the line curves up/down. So, I was trying to get rid of that, too. A CAD system would probably be useful to determine how to fit a rectangular screen on a curved frame.

EDIT: Sorry about the image size; I sampled it to 200 pixles and tried to reload a smaller one, but it's still big....

Posted by John Pytlak (Member # 331) on 01-17-2004, 10:04 AM:

Ideally the screen curve should be optimized by ray tracing --- each part of the screen surface should reflect the projected light back to the center of the seating area. See SMPTE Recommended Practice RP95, Installation of Gain Screens.

This can be tested in practice by holding a small front-surface mirror parallel to the screen surface at various points, and verifying that the projector light is reflected back to the center of the seating area.

Or a one-foot square of very thin "MYLAR Mirror" can be stuck to various areas of the screen by "static cling", and used to precisely verify where the light is being reflected:

MYLAR Mirror

http://www.rondexter.com/professional/lighting/lighting_tricks.htm