I've received a sample for the PDLC film I mentioned in my last post. While the film works well, I started having concerns about the circuitry and power requirements for a matrix based on this film. I think I'd have to build a significant amount of supporting circuitry for each pixel and for the matrix as a whole. Using PDLC films will likely require the use of optocouplers and TRIACs. I'm also not sure if a typical display matrix IC, like the HT16K33 will be well suited for controlling PDLC films, since you'd effectively be using a persistence of vision effect, with the films turning on and off at a high rate.

This got me thinking of other solutions. I had considered mechanical solutions in the past, with miniature 3D printed louvers or a sliding mechanism to construct a color changing 1-bit pixel, but those designs got complicated and unreliable. I also wanted to avoid using conventional motors, since that is a large added cost and complexity.

Another thought I had was to use polarized films and a mechanism using shape-memory alloy wires (a.k.a. SMA, muscle wire, nitinol). If I could build a mech that rotates a circular polarized film, relative to a stationary one, in 90 degree increments, this would satisfy many of the goals of the project.

The tricky part is getting the design right. Working with SMA wires is tricky and the prototypes I've made so far have had various issues. I think I'm getting closer to working solution though, so this seems like a good path to go down.