The actuator with the zig-zagging pattern of SMA wire has a big, intrinsic disadvantage: the strain it can sustain without suffering permanent deformation varies throughout the stroke length. There's no way to work around this, the math sucks.

I've got another model that I've come up with that solves this, running the wire through two pulleys attached to long support arm. Using this setup I could run 60cm of wire, which will contract about 3 cm, between two points that are a few cm apart. As long as I can keep the pulley system compact enough, I should be able to reach the 30% contraction goal.

This I printed out and tested a prototype. The goal was to verify that the pulleys will spin and get a feel of what the end device will look like. It worked.