Masking system / Adding multiple stop positions to motorized screen

I have a 150" DA-Lite Cosmopolitan Electrol which is mounted on a wall with a window and a balcony door. Now a (motorized) masking system should be applied. There are only horizontal masks necessary because the screen is used with constant width.


If possible I would think about a masking system like this: http://www.kathiejohnson.com/masking.html


But because of the balcony door it is not possible to have parts of the system permanently mounted on the floor. Between the bottom of the screen cabinet and the top of the open door there is only a gap of 1.5". It should be possible to install the upper part of the system showed at the link above in front of the screen cabinet. The bottom mask should be realized with the bottom mask which is painted on the screen.

But how to extend the bottom mask to the floor?

One possibility is to mount a drapery roller blind behind the screen cabinet, which is capable to cover the hole wall downto the floor. So the area under the bottom edge of the screen is allways black. But I don't want to do this really because there is no place behind the screen cabinet for the roller blind.

Any ideas except moving the screen a few inches forward?


The other problem is how to add multiple stop positions to the screen. My version of the screen has only the 2 end positions.

My currunt idea is to scan the rear of the screen with a reflective photoelectric barrier. If this detects one of the white dots which are glued to any desired stop position, than the movement of the screen is stopped with a relay. Starting from such a position the movement can be restarted in each direction till it stops again at the next dot. This would be the interactive but very simple method.

Another idea is to install an array of 2 or 3 reflective photoelectric barrier horizontally to encode the current stop position with dot patterns. So a simple logic would be able to control the system automaticly. It can detect where the screen is, decide in which direction the movement has to be started and at which detected dot pattern it has to be stopped. So it is possible to select an aspect ratio and the logic makes the rest.

I expect following problems:

If the dots are to small, they can't be detected shurely.

If the dots are to big, the real stop position depends on the direction of the movement.

This could be solved by additional dots marking braking positions do reduce the speed, but I don't want to make it to complex. I'am seeing myself programming a microcontroller ;-)


These are my current ideas. Did anybody something like this? Any suggestions to refine this method or to make it less complex?


Jochen