Bio-implantable medical devices need a reliable and stable source of power to perform effectively. Although batteries can
be the first candidate to power implantable devices as they provide high energy density, they cannot supply power for long
periods of time and therefore, they must be periodically replaced or recharged. Battery replacement is particularly difficult
as it requires surgery. In this paper, we develop a micromachined ultrasonic power generating receiver with a size of
3.5mmx3.5mm capable of providing sufficient power for implantable medical devices. The ultrasound receiver takes the
form of a unimorph diaphragm consisting of PZT on silicon. We dice bulk PZT with a thickness of 127 μm and bond the
diced pieces to a silicon wafer. In order to get a 50 μm thick PZT layer, which is needed for optimal power transfer, we
mechanically lap and polish the bonded PZT. We numerically investigate the performance of the fabricated receiver with
inner and outer electrodes on the surface of the PZT. Using COMSOL simulations, we analyze the effect of different sizes
of inner and outer electrodes under the actuation of the inner electrode in order to find the optimum electrode sizes. We
show that when the transmitter is generating an input power less than Food and Drug Administration limits, the receiver
can provide sufficient voltage and power for many implantable devices. Furthermore, the process developed can be used
to fabricate significantly smaller devices than the one characterized, which enables further miniaturization of bio-implanted
Proc. SPIE. 9801, Industrial and Commercial Applications of Smart Structures Technologies 2016
KEYWORDS: Chest, Stars, Data modeling, Quartz, Energy harvesting, Transducers, Motion models, Optimization (mathematics), Commercial off the shelf technology, Systems modeling, Information operations, Instrument modeling
This paper presents the characterization of the micro-generators embedded in Commercial-Off-The-Shelf (COTS) watches
based on a generalized rotational energy harvester model which predicts the upper bound on energy generation given
certain system constraints and specific inputs. We augment this generalized model to represent the actual micro-generator
used in the Seiko Kinetic watch with realistic damping coefficients which allow us to identify optimizations to move the
system output towards the upper bound. We have developed a mobile data logging platform which captures 6 DOF inertia
data and the voltage output from the micro-generator simultaneously. We have asked 6 subjects to conduct a series of daily
activities with the platform worn on different locations of the body. This effort not only serves as the experimental
validation of our model but also provides insight into the state of the art in wearable kinetic energy harvesting devices that
are commercially available. Finally we identify the opportunity for improvement on energy generation and show that we
can increase the power by reducing the mechanical damping in the system, which might require an alternative mechanism
with inherent lower friction.
A tremendous amount of research has been performed on the design and analysis of vibration energy harvester architectures with the goal of optimizing power output; most studies assume idealized input vibrations without paying much attention to whether such idealizations are broadly representative of real sources. These “idealized input signals” are typically derived from the expected nature of the vibrations produced from a given source. Little work has been done on corroborating these expectations by virtue of compiling a comprehensive list of vibration signals organized by detailed
classifications. Vibration data representing 333 signals were collected from the NiPS Laboratory “Real Vibration”
database, processed, and categorized according to the source of the signal (e.g. animal, machine, etc.), the number of dominant frequencies, the nature of the dominant frequencies (e.g. stationary, band-limited noise, etc.), and other metrics.
By categorizing signals in this way, the set of idealized vibration inputs commonly assumed for harvester input can be
corroborated and refined, and heretofore overlooked vibration input types have motivation for investigation. An initial qualitative analysis of vibration signals has been undertaken with the goal of determining how often a standard linear oscillator based harvester is likely the optimal architecture, and how often a nonlinear harvester with a cubic stiffness function might provide improvement. Although preliminary, the analysis indicates that in at least 23% of cases, a linear harvester is likely optimal and in no more than 53% of cases would a nonlinear cubic stiffness based harvester provide improvement.
The rapidly decreasing size, cost, and power consumption of wireless sensors has opened up the relatively new research field of energy harvesting. Recent years have seen an increasing amount of research on using ambient vibrations as a power source. An important feature of all of these generators is that they depend on the resonance frequency of the generator device being matched with the frequency of the input vibrations. The goal of this paper, therefore, is to explore solutions to the problem of self-tuning vibration based energy harvesters. A distinction is made between “active” tuning actuators that must continuously supply power to achieve the resonance frequency change, and “passive” tuning actuators that supply power initially to tune the frequency, and then are able to “turn off” while maintaining the new resonance frequency. This paper analyzes the feasibility of tuning the resonance frequency of vibration based generators with “active” tuning actuators. Actuators that can tune the effective stiffness, mass, and damping are analyzed theoretically. Numerical results based for each type of actuator are presented. It is shown that only actuators that tune the effective damping will result in a net increase in power output, and only under the circumstance that no actuation power is needed to add damping. The net increase in power occurs when the mismatch between driving vibrations the mismatch between driving vibrations the resonance frequency of the device is more than 5%. Finally, the theory and numerical results are validated by experiments done on a piezoelectric generator with a smart material “active” tuning actuator.