Session: Government Agency Student Posters

Paper Number: 172643

Design and Optimization of a Compliant Mechanism for Energy-Efficient Prosthetic Legs 

Compliant mechanisms are well suited for physiological applications due to the compliant nature of many biological systems. Prosthetic knees and ankles in particular are excellent candidates for the reduced complexity and added flexibility that compliant mechanisms offer. These advantages are compounded when compliant mechanisms add parallel stiffness to actuators used in active prostheses. Previous and ongoing work has shown that parallel torsional springs can reduce actuator energy and torque requirements in powered joints for a variety of activities, thereby facilitating lighter, more compact, and more efficient prostheses. It is therefore desirable to design novel, energy-dense and volume-efficient compliant mechanisms capable of replicating prosthesis joint motion while adding stiffness in parallel to an actuator. Furthermore, it is important that these compliant mechanisms are accurately modeled and have a tunable stiffness so that maximum actuator efficiency can be achieved for a variety of user masses.

This work characterizes a novel torsional spring compliant mechanism capable of replacing an ankle joint and adding optimal parallel stiffness. This mechanism consists of a series of thin-walled split tubes of increasing radii connected in series. These tubes are nested within one another to conserve volume, and the mechanism gains its motion through the torsional deflection of the split tubes, qualifying it as a compliant mechanism. Mechanics models allow these deivces to be designing to specific targets based on the users needs and activities.

In this work, the development and design of the nested split tube torsional spring compliant mechanism are described, and various prototypes are demonstrated. We present a simple mathematical model and further develop additional more complex models that make fewer simplifying assumptions. We also develop an inexpensive manufacturing method for metallic prototypes and subject four identical specimens to torsional testing to verify repeatability and model accuracy. Additional prototypes fabricated from polymers are also subjected to torsional testing and added to the data set. We conduct a finite element analysis of this model and compare it with other results. We then compare our mathematical models to test data to verify appropriate assumptions and produce a design tool to achieve tunable stiffness and support future work. Finally, we develop a customized genetic algorithm to design efficient nested split tube mechanisms under user-defined constraints.  The algorithm pulls from a database of readily available parts to minimize cost. There is a tradeoff in this mechanism, as with many torsional springs, between stiffness and range of motion. This algorithm can maximize stiffness for a given range of motion requirement and vice versa. Various individuals that meet the requirements can be produced, allowing the designer to select a suitable mechanism based on mass, volume, complexity, or other considerations.

As expected, the range of motion increases and stiffness decreases as more tubes are added to the nested split tube joint. Steel prototypes with an outer diameter of 1.5”, length of 6”, uniform wall thickness of 0.063”, and a total of 4 nested tube layers were built. The first of these prototypes has been tested and is capable of at least 30 degrees of rotation without plastic deformation. Further testing revealed a total angular deflection of at least 90 degrees in one direction with some plastic deformation but without failure, demonstrating the robustness of this design. As expected, connection points between tubes are subjected to the largest stresses and are the principal failure locations. A mechanics model that assumes that each tube is in pure torsion and does not experience any axial stresses due to restrained warping best predicted the trends in the data. Future work could include further experimentation with different materials as well as an analysis of joint behavior and model accuracy as a function of tube length. Overall, the viability of this mechanism as a potential robust solution to prosthetic limb design as well as a potential solution for a variety of other applications is demonstrated.

This material is based upon work supported by the National Science Foundation under Grant No. 2344766 and is in collaboration with the University of Notre Dame.

Presenting Author: Collin Klomp Brigham Young University

Presenting Author Biography: Collin Klomp is a master’s student in mechanical engineering at Brigham Young University, working in the Compliant Mechanisms and Robotics Lab under Dr. Larry Howell. His research centers on designing compliant mechanism solutions for prosthetic ankle challenges.

Authors:

Collin Klomp Brigham Young University
Edgar Bolivar-Nieto University of Notre Dame
Nathan Usevitch Brigham Young University
Larry Howell Brigham Young University