Breadboards can be expensive and they’re naturally bulky. Because they’re bulky, they may unnecessarily force a minimum height for the switch. Maybe worse, they present a grid of holes that have to be filled correctly with components and wires if the circuit is expected to function properly.
The goal of this version is to provide an alternative solution to breadboards but not to require the person assembling the switch to have all the specialized equipment and skill necessary to safely solder components together. At the same time, we looked at how we could eliminate the requirement to select the proper gauge of wire and properly strip away the insulation.
In a few words, the goal is to design into the 3D object as much of the circuit design as possible and to use a liquid solder equivalent (a conductive paste) that can be injected into the 3D printed parts to join the electronic components.
This design requires two specific electronic components, a specific micro-switch and a specific headphone jack. The design was modeled in Fusion 360 to match these specific components. However, it’s easily conceivable that the design could be programmed, using OpenSCAD, to support an arbitrary set of micro-switches and headphone jacks.
Here are the specific headphone jack and micro-switch (along with links to purchase them on Amazon.com):
 |
 |
| headphone jack | micro-switch |
If the leading pin of the headphone jack is bent as it appears to be in the picture above, gently bend it straight up before inserting it into the mount so it aligns properly with the provided hole.
Here are the mounts for the headphone jack and the micro-switch.
The headphone jack mount has a base and a cover. The purpose of the cover will become clear later. The holes and slots on the surface of the mount are there to accommodate the pins and bumps on the underside of the jack. Two of the holes have a funnel shaped opening because they will be the point where the conductive paste is injected.
The micro-switch mount is simpler. Again there are two holes with a funnel shaped opening for injecting the conductive paste.
Both mounts should be printed at a small layer height to ensure that the channels are accurately printed. The mounts that were tested for this design were printed with a 0.15 mm layer height.
The horizontal arms on either side accommodate two breadboard jumper wires:
The wires come with both ends “pre-stripped” and with a plastic collar to aid insertion into the component mounts.
The two mounts have “internal plumbing” that joins the proper pins of the headphone jack and micro-switch to the ends of the jumper wires. These pictures show each component along side its internal plumbing:
|
 |
 |
 |
Each injection port has a corresponding overflow port. The overflow port provides a visual way to identify when the internal chamber has been sufficiently filled with paste:
The conductive paste (called conductive paint, by the manufacturer) comes from from Chip Quik:
The smallest syringe holds 5 grams of paste and only about 0.2 grams are needed to assemble a switch. You can purchase a nylon sleeve to fit over the end of the metal needle which will keep the paste from drying in the syringe. The paste has a shelf life of over a year. That’s enough time to print and assemble lots of switches.
The switch base has indents to aid in placing the mounts and, in the case of the headphone mount, is designed to allow the mount to slide into the wall of the base after the mount is assembled with the jack inside.
The flexure spring has been retained from the previous version with a small change. The block which held the headphone jack in place in the previous version has been removed because it is no longer needed. A cut is added to each edge to provide clearance for the clips built into the cover.
The cover has been modified with clips that fit into slots on each side of the base. These clips, along with the friction of the two surfaces, is sufficient to hold the cover to the base without the use of screws.
The post, activation surface, and activation surface mount have been preserved, unchanged from the previous version. Two post lengths are provided as STL files. One is 3 mm long and the other is 6 mm long. The shorter the length, the longer the activation distance. All the parts, together, look like this:
The STL files for this design can be found on Thingiverse at: https://www.thingiverse.com/thing:3819575
The following video shows how to assemble the switch:
Design Considerations
The conductive paint/paste is not a glue. As a result it doesn’t hold the pins of a component or the ends of the jumper wire in place. That exposes a potential weakness of the approach. Once the paste has partially or completely dried, any movement of the component or jumper could compress the surrounding paste and create an air gap between the paste and the pin/wire. To address this problem, the model adds walls and tunnels that create a rigid container for the component and the jumper.
The headphone jack is the component most likely to experience displacement forces. The jack can experience these forces in almost any direction when the headphone plug is inserted or removed. To limit the impact of these forces, the jack is locked in place by the mounting base and cover. Additionally, super glue is applied to the cover during assembly.
The micro-switch is exposed, primarily, to horizontal displacement forces when pressure is applied to the edges of the activation surface. To hold the micro-switch rigidly in place, a wall has been added to the switch mount to surround the micro-switch on all four sides.
Finally, the headphone mount and the micro-switch mount super glued to the base to prevent the mounts from moving within the switch.
Stacking Up Against the Requirements
This design inherits many of the features of previous designs. Only significant changes will be described in detail below.
1 – support being constructed by individuals closest to the individual with a disability (e.g., therapists, AT professionals, family members):
- can be 3D printed on a consumer accessible printer (i.e., printer cost <= $250)
Yes. Note that printer cost has fallen significantly and print quality has risen significantly from the time when this requirement was originally conceived.
- each instance of the switch can be printed and assembled for less than $20
Yes, this is the lowest cost design so far. Cost: $5.09 (plastic and power – $1.50, electronic components – $2.79, paste – $0.80)
- can be assembled and disassembled with no special skills or equipment (e.g., does not require soldering skills and equipment, components bolt/snap/tape/glue together)
Yes this is the simplest design to assemble so far. Inject paste, insert components and wires, glue mounts, snap cover to base.
2 – support trouble shooting via simple part-replacement techniques
Yes, though once parts are glued, many parts must be reprinted or replaced. The low cost to print the parts and produce a new switch reduces the need for this requirement.
3 – support personalization of aesthetics (e.g., user-specified colors and shapes, activation surface texture, and tactile activation travel characteristics)
Yes.
4- support simple customization of features and functions:
- activation distance and activation pressure
Yes.
- support for externally powered elements (e.g., adapted toys)
Yes.
- support for internally powered elements (e.g., visual /auditory / tactile feedback, wireless interfaces)
No. This design could support internal power but would require additional mounts and a new micro-switch mount design.
- support for common mounting and positioning solutions
No, but designing a mount that attaches to the base in a manner equivalent to the cover would be relatively simple.
- supports simple scaling of the components to meet a variety of applications and abilities
Yes, but even more so if this design were implemented in a 3D modeling language like OpenSCAD.
4 – support simple modification to better fit the needs and desires of the individual with a disability without the destruction of the component that is being replaced
Not the headphone jack because it is super-glued inside its mount.
5 – support simple, end-user replacement of components with limited life (e.g., batteries)
Yes, though this design doesn’t include such components.
6 – exhibit high reliability simply by avoiding known, low reliability configurations and components
Yes. The jumper wires can be low reliability if bent repeatedly. The designs for the component mounts hold them securely in place so there shouldn’t be any movement of the wires once they’ve been attached to the mounts.
7 – support simple and low/no-cost design customization if the supplied designs fail to fit the needs and desires of the individual with a disability
No because many parts are super-glued together.
8 – many aspects of the design should be reusable and extensible across new switch types, new form factors, and alternative technologies. An optimal design will have reusable elements that facilitate porting the concepts (or at least don’t hamstring the porting) to other implementation technologies (e.g., injection molding, laser sintering, stereolithography) or other form factors – e.g., how many of the POC design elements could be reused.
Yes, the concept of an injected conductive paste could have lots of applications and facilitates the concept of designing the circuit into the 3D model.
Caveats
The Chip Quik paste can be purchased in two sizes: 5 grams and 10 grams. Given that only 0.2 grams are needed to assemble a single switch, that leaves a lot of paste in the syringe that isn’t getting used. At $15.95 + $4.00 shipping for the 5 gram syringe, that can result in an expensive way to create a single switch. The paste should be good for over a year but if you’re only going to create a single switch, this approach may not be right for you. On the other hand, if you anticipate possibly creating several switches, the money’s not wasted.
As the paste dries, the internal resistance drops. It starts out with almost infinite resistivity which falls to a minimum of about 35 ohms in 3 days. That means you won’t be able to test your connections until at least 3 days after assembling the switch. That can be frustrating if you’d like to experience the immediate feedback of your work. There may be ways to speed up the drying process but because you’re using super glue to immobilize the mounts, you need to wait at least 12 hours for the glue to cure.
The post and activation surface mounts must be 3D-printed upright. That means that they’re sensitive to shearing forces. These forces are most likely to occur when the activation surface is over-rotated after assembly. To eliminate this risk, the activation surface mount should be super-glued to the flexure spring and the activation surface should be super-glued to the activation surface mount. Does this violate the requirement to be able to swap out the activation surface and the flexure spring? Yes it does. You can still swap out the post to change the activation distance. The bottom line is that you should experiment with activation surfaces and activation force (the flexure spring) and once you determine the best options for those two items, go ahead and super-glue the activation surface mount to the flexure spring and the activation surface to the activation surface mount.