How to create a Zero-Force Compliant Surface for FTIR Based Multi-touch on a Back Projected Table

Materials needed for zero-force:

SORTA-Clear 40 Silicone Rubber - North America / Germany

Mylar - Available at any art store or craft store. Also known as 'Sketching paper' or 'Architekt Papier' in Germany

See here for a photo.


Mix 908 grams (one tub of Silicone) + 90.8 grams (one bottle of Rubber) together for a 0.5mm surface on a 100cm by 75 cm table. Follow Instructions on the packets for details concerning how to do this.

Getting the Silicone Rubber even is non-trivial. I had two pieces of metal on each side of the table as 'tracks' , about 2 mm high. I then dragged a metal piece across these tracks to slice the Silicone Rubber and spread it out. When you reach the end of the table, flip the metal stick over and repeat. You'll do this a lot.

There will be gaps. Use business cards to fill them and spread out the Silicone Rubber.

When it looks good, do the same thing in an orthogonal direction until you get a nice even glass like surface. The process here is a little different. There are no tracks to guide the movement. With the tracks on, it is hard to get the pressure even and make the Silicone Rubber even as well. So, I took a large tube the one that the Rosco Grey came rolled on, and 'pushed' it over the surface. Note, I did not roll it I pushed it. This distributes the Silicone Rubber evenly. Use about 40% force (if 100% is pushing on the table at full force) and push across the table to one side and then repeat until the surface is like glass.

Note, this step is the 'final' step in the pouring. When I did this, I had streaks, or visible lines in the Silicone Rubber. In the end, they did not show up on the projection surface so don't worry so much. It will never be perfect this is clearly something a machine should do. Also, the Silicone Rubber spreads itself out naturally. It evens out over time and these streaks are less pronounced. See the top of this photo to see the streaks.

Note, mixing the solutions introduces air bubbles. If you pour thin, the air bubble rise to the surface and are released. For thicker mixes you need to suck them out somehow. Good luck!

I cannot stress this anymore: do all of this fast. The faster you do this, the more the mix evens out naturally. After 15 min the mix starts to get sticky and less workable. I suggest practicing on a smaller panel first. However, make sure to get the mixtures combination right. Note, you must mix by WEIGHT and not by volume. To avoid this, just use one tub Silicone to one bottle Rubber. Easy.

Let it dry for 24 hours at least. I put mine in the Sun for the last 12 hours. The Silicone Rubber must not be sticky, otherwise the Mylar will stick. If you do not mix the right proportions, it will be sticky and will not work. The projection surface will stick to it and you'll have to start over.

Note, I did not peal the Silicone Rubber off the surface. Ever. It is stuck on the acrylic and there is no intentional air gap in between. By 'intentional', I mean that there is likely a thin, barely noticable, pocket of air that results from the pouring and curing process. For all practical purposes, the Silicone Rubber is adhered to the surface.

Voila. You're done.

Put the Mylar on top.

Put some Transparent Vinyl on top, plastic, or whatever, to protect the Mylar (if you want). It doesn't matter if you're doing it for yourself, you don't need this. However, for an exhibit or installation you need robustness.

Some comments: it feels fantastic. Mylar has a great texture. I put Rosco Grey on top, it still works, just darker (obviously). It feels better than Mylar — Vinyl Rules! The performance is uniform across people. I tested with 15 people and it was zero-force for each person. With bare acrylic, performance varies by the amount of naturally occuring oil each person has in their skin.

Oh, pens work too!

That's all for now. Any questions, let me know. Or, check out my homepage to see why I'm doing with all this.

How to top-project

David Smith at Queen's University in Canada has done this. He's a colleague of mine. We did a similar process, he had top-down projection because the physical setup is cleaner (back-projection with 4 sides available needs a raised floor. No raised floor removes 1 edge of the table for projection). See the video here.

As well, he poured his much thicker so there is more haptic feedback. I wasn't so concerned with this. Email him if you have questions, his webpage is here.

Created by holman. Last Modification: Thursday 19 of April, 2007 12:36:23 by holman.

Media Computing Group at RWTH Aachen