Toys4Therapy
Documentation
Week 12
Week 13
Week 11
Introduction
Documentation
Trail of Evidence
Recess
Week 14
Trail of Evidence
Documentation
Week 15
Trail of Evidence
Documentation
Week 16
Trail of Evidence
Documentation
Week 17
Documentation
Trail of Evidence
Week 18
Trail of Evidence
Documentation
Week 19
Trail of Evidence
Documentation
Week 20
Trail of Evidence
Documentation
In order to get an impression of exactly what the product will look like, and to determine the different dimensions of the parts, we made a model of the end product.
Silicone rubber research
We started researching silicone rubber itself. What exactly is it? On the site of "Smooth-On" a lot of information could be found, also about the product "Dragon Skin™ FX- Pro", the product that Mariana Pinheiro also used.

Dragon Skin™ FX- Pro features:
• Soft - Shore 2A hardness allows for softer appliances using less SLACKER™ deadening agent
• Low Viscosity - combined with long working time reduces bubbles and makes pouring easier
• Cures Dry - will not leach silicone fluid
• Long Working Time - 10-12 minute working time is good for larger pours or injections
• Fast Cure - 40 minute cure time allows for quick turnaround of castings or molds
• More Workable - Do More - has a ‘gel’ stage - slushing the material into prosthetic molds is easier

For more information about silicone:
• https://www.smooth-on.com/category/platinum-silicone/

For more information about Dragon Skin™:
• https://www.smooth-on.com/products/dragon-skin-fx-pro/
Model of the end product
Designing the electronics
To start of with designing the electronics a list of required feutures was drafted to get a general idea what the device should be capable of.

- Generates sounds
- Processes 24 + 4 button inputs
- Controls multiple RGB LED's
- Controls two Displays
- USB Port for programming

The general design was divided into two types of boards. The LED boards which would mount the LED's, switch and possibly some power regulation. And the main board which would handle all the other functions.

Some early component decisions were made. The RGB LED's could have been of the passive variety with an external LED controller, but this option was more expensive than using LED's with an embedded controller. Usually I prefer the APA-102C LED's since their dedicated clock line allows for lenient timing specifications, however due to their high price point the cheaper WS2812B LEDS were selected.

The displays selected are a copy of the display used in the Nokia 5110. These displays are popular with DIY projects because they are affordable, easy to control and there are plenty of examples and libraries available.

The buttons would be connected in a row and column style array. So instead of having every single button use up a pin on the microcontroller, one pin is used per row and one per column.

Next up the general power topology was designed. All components are eventually powered by a low dropout linear regulator, because unlike switching regulators they don't typically suffer from feedback loop instability. So on the odd chance that the buck converter becomes unstable and decides to overvolt everything that is connected behind it, I wont have to resolder ~100 components.
Because the LDO's only have to drop ~1V they don't generate a huge amount of heat. Nonetheless, when all the LED's are powered up fully they draw more than 5 Amps at 5V in total, so it was still required to mount individual LDO's on every led board instead of a single one on the main board, since this spreads the heat more effectively.

The schematic centers on an XMega microcontroller. Because the schematic does not contain advanced analog circuits, high speed signals or digital glue logic.
It is very straight forward to debug.
An ATTiny was used to generate audio signals, in the ToE this decision has been discussed in more detail.

All 25 PCB's were milled in an LPKF machine. Instead of punching the via's we opted to solder them with a piece of copper wire, since it makes a more reliable connection. After all the PCB's were soldered they were measured and tested.
While looking for an alternative to making a mold, I started looking at what is possible in the Makers Lab. That's how I ended up at the vacuum former, you can make a mold with the vacuum former without any imperfections (which I did have with the use of clay). Since the bottom of my object is flat, I only needed a one-piece mold. This is exactly what the vacuum former can do.

For the shape of the mold I cut a piece of MDF wood in a round shape, on top of this I put some nuts (moeren).
Trail of Evidence
As a test we made a clay mold, in a wooden MDF piece. This did not quite work as expected, because of the use of clay the surface of the rubber silicone did not become completely smooth.
pdf
"Dragon Skin™ FX-Pro is a soft, stable, high performance platinum silicone rubber specifically designed for creating silicone makeup appliances and skin effects. Dragon Skin™ FX-Pro is compatible with Skin-Tite™ silicone adhesive as well as medical grade pressure sensitive adhesives for adhering silicone appliances temporarily to the skin. Cured material is skin safe and certified by an independent laboratory."
Technical Data:
Further research on YouTube
After having had a reasonable basis of information about silicone rubber, we went looking for some more information about silicone rubber on YouTube.
A video explaining how to make a mold, this video also provides useful tips.
A video that shows how strong the material is. The material is stretchable, but transforms back into its own form. Exactly what we need!
Buying silicone rubber


We bought the silicone rubber at FormX, a shop that sells design materials. Here on the left you can see the materials we bought.

The shop address is: KNSM-Laan 810, 1019 LT Amsterdam.

Website: http://formx.nl/index.php
Testing silicone rubber


Making silicone rubber
The product consists of the following (within the meaning of the exterior):
• 24 discs
• 4 buttons
• 2 LED screens
• 1 bottom and top
• 2 side length
• 2 side width
Plastic mold


Some examples of the molds I made with the vacuum former.

Liquid silicone rubber is poured into the molds, after about 20 minutes the silicone rubber will become a solid form.
On the left is the result of silicone rubber in a solid form.
Silicone rubber result


zip
Led board schematic and PCB
zip
Main board schematic and PCB
zip
3D prototype