This paper addresses an issue in energy harvesting that has plagued the potential use of harvesting through the piezoelectric effect at the micro-electro-mechanical systems (MEMS) scale. Effective energy harvesting devices typically consist of a cantilever beam substrate coated with a thin layer of piezoceramic material and fixed with a tip mass tuned to resonant at the dominant frequency of the ambient vibration. The fundamental natural frequency of a beam increases as its length decreases, so that at the MEMS scale the resonance condition occurs orders of magnitude higher than ambient vibration frequencies, rendering the harvester ineffective. Here, we propose a new geometry for MEMS scale cantilever harvesters with low fundamental frequencies. A “zigzag” geometry is proposed, modeled, and solved to show that such a structure would be able to vibrate near resonance at the MEMS scale. An analytical solution is presented and verified against Rayleigh’s method and is validated against a macroscale experiment. The analysis is used to provide design guidelines and parametric studies for constructing an effective MEMS scale energy harvesting device in the frequency range common to low frequency ambient vibrations, removing a current barrier.
14 July 2011 in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
This research presents an experimental and theoretical energy harvesting characterization of beam-like, uniform cross-section, unimorph structures employing single-crystal piezoelectrics. Different piezoelectric materials, substrates, and configurations are examined to identify the best design configuration for lightweight energy harvesting devices for low-power applications. Three types of piezoelectrics (single-crystal PMN-PZT, polycrystalline PZT-5A, and PZT-5H-type monolithic ceramics) are evaluated in a unimorph cantilevered beam configuration. The devices have been excited by harmonic base acceleration. All of the experimental characteristics have been used to validate an exact electromechanical model of the harvester. The study shows the optimum choice of substrate material for single-crystal piezoelectric energy harvesting. Comparison of energy scavengers with stainless steel substrates reveals that single-crystal harvesters produce superior power compared with polycrystalline devices. To further optimize the power harvesting, we study the relation between the thickness of the substrate and the power output for different substrate materials. The relation between power and substrate thickness profoundly varies among different substrate materials. The variation is understood by examining the change of mechanical transmissibility and the variations of the coupling figure of merit of the harvesters with thickness ratio. The investigation identifies the optimal thickness of the substrate for different substrate materials. The study also shows that the densities of the substrates and their mechanical damping coefficients have significant effects on the power output.
18 December 2017 in Smart Materials and Structures
In this paper, a physical platform is proposed to change the properties of phononic crystals in space and time in order to achieve nonreciprocal wave transmission. The utilization of magnetoelastic materials in elastic phononic systems is studied. Material properties of magnetoelastic materials change significantly with an external magnetic field. This property is used to design systems with a desired wave propagation pattern. The properties of the magnetoelastic medium are changed in a traveling wave pattern, which changes in both space and time. A phononic crystal with such a modulation exhibits one-way wave propagation behavior. An extended transfer matrix method (TMM) is developed to model a system with time varying properties. The stop band and the pass band of a reciprocal and a nonreciprocal bar are found using this method. The TMM is used to find the transfer function of a magnetoelastic bar. The obtained results match those obtained via the theoretical Floquet–Bloch approach and numerical simulations. It is shown that the stop band in the transfer function of a system with temporal varying property for the forward wave propagation is different from the same in the backward wave propagation. The proposed configuration enables the physical realization of a class of smart structures that incorporates nonreciprocal wave propagation.
This manuscript investigates energy harvesting from arterial blood pressure via the piezoelectric effect for the purpose of powering embedded micro-sensors in the human brain. One of the major hurdles in recording and measuring electrical data in the human nervous system is the lack of implantable and long term interfaces that record neural activity for extended periods of time. Recently, some authors have proposed micro sensors implanted deep in the brain that measure local electrical and physiological data which are then communicated to an external interrogator. This paper proposes a way of powering such interfaces. The geometry of the proposed harvester consists of a piezoelectric, circular, curved bimorph that fits into the blood vessel (specifically, the Carotid artery) and undergoes bending motion because of blood pressure variation. In addition, the harvester thickness is constrained such that it does not modify arterial wall dynamics. This transforms the problem into a known strain problem and the integral form of Gauss's law is used to obtain an equation relating arterial wall motion to the induced voltage. The theoretical model is validated by means of a Multiphysics 3D-FEA simulation comparing the harvested power at different load resistances. The peak harvested power achieved for the Carotid artery (proximal to Brain), with PZT-5H, was 11.7 μW. The peak power for the Aorta was 203.4 μW. Further, the variation of harvested power with variation in the harvester width and thickness, arterial contractility, and pulse rate is investigated. Moreover, potential application of the harvester as a chronic, implantable and real-time Blood pressure sensor is considered. Energy harvested via this mechanism will also have applications in long-term, implantable Brain Micro-stimulation.
This paper presents development of a three dimensional finite element model for simulations of a conformable piezoelectric sensor utilizing COMSOL Multiphysics. The sensor has a multi-layer structure composed of four circular piezoelectric elements arranged in an array structure laminated on a soft substrate and is capable of providing a strain mapping of soft tissue surfaces for spatiotemporal biokinematic assessment of the facial skin. Here, we provide the finite element method (FEM) for the sensor to predict its electromechanical behavior. This paper studies the effect of the design parameters such as dimensions of the piezoelectric sensor and the substrate on voltage sensitivity and sensor compliance. The FEM model is established to understand the underlying physics and guide the mechanical characterization of the system. The developed model is experimentally verified through two series of tests. The first set of tests involve comprehensive in vitro mechanical testing to provide accurate measurements of strain during compression, stretching, and bending. The second set of tests present in vivo experiments on healthy and amyotrophic lateral sclerosis subjects. The experimentally verified FEM model provides a detailed insight into analyzing the response of the sensor which establishes new design rules for next generations of conformable piezoelectric sensors.
27 February 2019 in Journal of Mechanisms and Robotics
This paper presents the design of a two-degrees-of-freedom (DoFs) variable stiffness mechanism and demonstrates how its adjustable compliance can enhance the robustness of physical human–robot interaction. Compliance on the grasp handle is achieved by suspending it in between magnets in preloaded repelling configuration to act as nonlinear springs. By adjusting the air gaps between the outer magnets, the stiffness of the mechanism in each direction can be adjusted independently. Moreover, the capability of the proposed design in suppressing unintended interaction forces is evaluated in two different experiments. In the first experiment, improper admittance controller gain leads to unstable interaction, whereas in the second case, high-frequency involuntary forces are caused by the tremor.