Process

3D scanning:

I scanned my hand using iPad structure sensor. The sensor had a lot of trouble scanning my hand separate from my body, so I set up a board with a hole to scan my hand as an object instead. Many thanks to Ridwan for helping me with the scan.

Mad props to Ridwan for the help with scanning.

3D modeling:

A 3D model produced was processed and reduced in meshmixer. The resulting model is uploaded to sketchfab here. I want to try out wrapX for redefining the mesh.

I then made a shell around the model and split it in two. I had to be careful to split it such the two halves could fit on the model when 3d printed. The separation between the model and the shell is 2.5 mm(1/8″) this is so that I can use 2.5mm(1/8″) acrylic pieces as spacers to ensure the model is centered inside the shell then closed.

3D Printing:

I printed the models using Cura and Ultimaker 2+. Since the model is relatively smooth, the layer height does not need to be fine. Even a 4mm layer height with 6mm nozzle is good. There were a LOT of failed prints.

Take care of the orientation of the build. It can have a huge effect on the time taken for the print.

Polishing:

The print needs to be cleaned up before use. I sanded the prints with a Dremel, going slow to prevent melting. Then I covered the entire surface with epoxy. The epoxy adheres to the rough surface well and allows the silicone to be removed easily.

Adding electronics:

I stuck discs of 2.5mm(1/8″) acrylic on the model, spaced regularly and test fitted the shell. Then I stuck neopixels and a flex sensor on the model. I used silicone wires to connect the sensors so that it sticks to the poured silicone easily.

Casting:

I followed standard techniques to cast silicone using a vacuum chamber. I’d recommend this video for instructions.

What went wrong:

The silicone I used(EcoFlex-30) has a relatively low working time of ~20 minutes. So there wasn’t enough time to degas the silicone when pouring it into the thin cavity between the model and the shell. The bubbles got trapped in the silicone and ended up in the final solid cast. Also, the silicone is not very durable and tears easily. And the electronics are not flexible enough to allow the cast to be removed from the model without damage.

What I would do differently:

• Model in a different way so that the casting is in a perpendicular direction. Having a shorter distance for the bubbles to travel will allow for easier casting.
• Design the mold with air escape holes at the bottom
• I would use silicone with longer working time.
• Use shorter stretches of electronics separated by coiled silicone wire.