Arduino driven Hand Actuator

myminifactory

When I discovered what eNABLE does and started learning about the problems eNABLE community members face when designing, printing, and assembling prosthetic hands in any variant, I was able to identify a couple of common issues that plagued almost all cases, and I decided to start working on a way to address those challenges. I first identified the scale/size issue; when a printed hand involves screws or bolts and other standard-sized hardware, it becomes extremely difficult to properly scale to a recipient's custom size. Then there is the mechanical aspect of it; all systems I have seen use the force the recipient has in their forearm/wrist to open/close it. This approach has several disadvantages since every recipient is different and a design that may seem effective for some may not work at all for others, (different flexion and extension) and the strength exerted by it varies accordingly, but always within certain limits (never too strong) dictated by the forearm/wrist itself. We encounter limitations with 3D printers as well, which cannot accurately print smaller features, and also, the plastic itself reaches a limit in its mechanical properties as features get smaller. Additionally, there's the issue of transradial amputation, where the above (wrist-actuated) does not apply. Furthermore, the thing about individual fingers being controlled all at once seems very disadvantageous, since, in most cases, all fingers share a common pivot point, which makes grasping non-cylindrical objects extremely difficult, if not impossible. So I started working on these topics and designed this piece that I present to you here. I designed an enclosure where the mechanical side of things happens, that doesn't need scaling, and that can interact with any type of 3D-printed hand (PTC), eliminating the problems with varying hardware sizes. I also designed a pulley system that multiplies the work that the actuating servo can produce, along with a second pulley system that allows the fingers to move somewhat independently, giving them more freedom to grasp conical objects or other non-cylindrical ones. This enclosure can be axially (collinear) mounted on the recipient's residual limb, if there's enough room for it between the elbow and the wrist, or it can be mounted on the underside of the residual limb via a couple of adapters. (Everything was designed to be 3D-printed without any support, making it easier for anyone to assemble it with minimal knowledge/skill required.) I also tried to remove as much material as possible so it could be as light but as strong as possible. All 3D-printed parts weigh 178 grams. The heaviest parts are the battery and the linear servo that controls the movement. The movement is actuated via an Arduino Micro microcontroller that reads the values from a Myoware sensor placed on the biceps muscle group. The Arduino actuates the servo, which moves the fingers via a multiplying pulley and a tensile system that allows independent motion between the fingers. There's an Arduino sketch (code/program) I wrote, which calibrates the sensor each time the system is turned on, so if the readings from the Myoware sensor vary depending on its placement on the recipient's arm, the system adapts to this. The hardest part of this system while assembling it is that it requires soldering, and that's it, pretty simple. I tried to make it as simple, light, efficient, cheap, and strong as possible. I am using red PLA, printed at 195c, with .3mm layers, and varying percentages of infill, all printed on a Makerbot Replicator 2, using a Flexy-Hand (by Gyrobot) to demonstrate the working principles of my design. Clearly, this design can and will work with any PTD design that actuates via pulling strings. You'll find the complete list of materials with its recommended print settings and characteristics in the files attached, as well as the Arduino code involved and instructions for assembly. Even when the challenge is to create a PTD with several technical requirements, and my design doesn't fully comply with all of them, I think it's a great time and opportunity to share this work that I believe can be very helpful (mainly in concept but also in execution). I really hope it’s something that contributes to the eNABLE community. Thanks for reading, and again, please feel free to comment/ask me on this design. José Santos Lizcano.