Bernard H Stark

Review of Power Conditioning for Kinetic Energy Harvesting Systems

In this paper, a summary of published techniques for power conditioning within energy harvesting systems is presented. The focus is on low-power systems, e.g, <;10 mW, for kinetic energy harvesting. Published concepts are grouped according to functionality and results contrasted. The various techniques described are considered in terms of complexity, efficiency, quiescent power consumption, startup behavior, and utilization of the harvester compared to an optimum load. This paper concludes with an overview of power management techniques that aim to maximize the extracted power and the utilization of the energy harvester.

An Experimental Investigation of the Tradeoff between Switching Losses and EMI Generation With Hard-Switched All-Si, Si-SiC, and All-SiC Device Combinations

Silicon carbide (SiC) switching power devices (MOSFETs, JFETs) of 1200 V rating are now commercially available, and in conjunction with SiC diodes, they offer substantially reduced switching losses relative to silicon (Si) insulated gate bipolar transistors (IGBTs) paired with fast-recovery diodes. Low-voltage industrial variable-speed drives are a key application for 1200 V devices, and there is great interest in the replacement of the Si IGBTs and diodes that presently dominate in this application with SiC-based devices. However, much of the performance benefit of SiC-based devices is due to their increased switching speeds ( di/dt, dv/ dt), which raises the issues of increased electromagnetic interference (EMI) generation and detrimental effects on the reliability of inverter-fed electrical machines. In this paper, the tradeoff between switching losses and the high-frequency spectral amplitude of the device switching waveforms is quantified experimentally for all-Si, Si-SiC, and all-SiC device combinations. While exploiting the full switching-speed capability of SiC-based devices results in significantly increased EMI generation, the all-SiC combination provides a 70% reduction in switching losses relative to all-Si when operated at comparable dv/dt. It is also shown that the loss-EMI tradeoff obtained with the Si-SiC device combination can be significantly improved by driving the IGBT with a modified gate voltage profile.

Analysis of Shaped Pulse Transitions in Power Electronic Switching Waveforms for Reduced EMI Generation

Consideration of the higher order time derivatives of voltage and current transitions in power semiconductor devices enables the specification of “S-shaped” switching waveforms which offer an improved tradeoff between high-frequency EMI generation and switching losses. In comparison with the widely used first-order derivative trapezoidal switching waveform approximation, Fourier analysis of the proposed “S-shaped” waveform shows that it exhibits a 20 dB/dec steeper spectral gradient at high frequencies, resulting in a 20 dB greater reduction in high-frequency spectral content per decade increase in rise time. Numerical analysis of the proposed waveform shows that both peak and total RF power, employed as indicative EMI metrics, are reduced significantly with no increase in overall switching time. Experimental investigation of the effect of introducing a frequency-selective EMI transmission path shows that the overall trends in the relationships between time-domain waveform parameters and high-frequency spectral content are maintained, while the values of the waveform timing parameters which minimize the two EMI metrics are changed.

A comparison of power output from linear and nonlinear kinetic energy harvesters using real vibration data

The design of vibration energy harvesters (VEHs) is highly dependent upon the characteristics of the environmental vibrations present in the intended application. VEHs can be linear resonant systems tuned to particular frequencies or non-linear systems with either bi-stable operation or a Duffing-type response. This paper provides detailed vibration data from a range of applications, which has been made freely available for download through the Energy Harvesting Network’s online data repository. In particular, this research shows that simulation is essential in designing and selecting the most suitable vibration energy harvester for particular applications. This is illustrated through C-based simulations of different types of VEHs, using real vibration data from a diesel ferry engine, a combined heat and power pump, a petrol car engine and a helicopter. The analysis shows that a bistable energy harvester only has a higher output power than a linear or Duffing-type nonlinear energy harvester with the same Q-factor when it is subjected to white noise vibration. The analysis also indicates that piezoelectric transduction mechanisms are more suitable for bistable energy harvesters than electromagnetic transduction. Furthermore, the linear energy harvester has a higher output power compared to the Duffing-type nonlinear energy harvester with the same Q factor in most cases. The Duffing-type nonlinear energy harvester can generate more power than the linear energy harvester only when it is excited at vibrations with multiple peaks and the frequencies of these peaks are within its bandwidth. Through these new observations, this paper illustrates the importance of simulation in the design of energy harvesting systems, with particular emphasis on the need to incorporate real vibration data.

Ultralow Power, Fully Autonomous Boost Rectifier for Electromagnetic Energy Harvesters

In this paper, a complete power conditioning system for a vibration energy harvester is presented that operates at ultralow power levels. The power conditioning system, implemented with discrete components, is self-starting and fully autonomous, and based upon a full-wave boost rectifier topology. The design utilizes the stray inductance of the harvesters coil, eliminating the need for separate inductors, and employs open-loop control that reduces the quiescent power overhead to just 21 μW, while still extracting 84% of the maximum available power from the harvester. The design of the subsystems, including self-start circuitry, is described in detail, and it is shown that careful active device selection is required to minimize losses. It is experimentally demonstrated that the power converter achieves conversion efficiencies of up to 76% at submilliwatt power levels, including quiescent losses. The overall system efficiency peaks at 65% at 0.9 mW, while still achieving 51% at 200 μW. The ability of this system to operate efficiently at ultralow average power levels opens up new possibilities to further miniaturize vibration harvesters and deploy them into environments with lower vibration levels than is currently possible.

Start-up circuit with low minimum operating power for microwatt energy harvesters

Remote sensors powered by energy harvesting need to restart successfully after long periods of no available energy, during which all stored energy may have been depleted. This start-up is affected by known phenomena such as `lock-up` and `voltage collapse`. In this study, the authors address these phenomena in the context of energy harvesting where the energy required for a sense and transmit cycle is accumulated over long time periods at low power. A voltage-detecting switch with very low power consumption is proposed, which avoids system lock-up. This start-up circuit uses an array of discrete MOSFETs operated in their sub-threshold regions. A performance metric for start-up circuits is proposed. `Minimum operating power` is the harvester power level below which the load does not start up. Reducing this minimum operating power thus reduces the required size of harvester or increases the application range. For the proposed circuit, a minimum operating power of 3.5 μW is derived experimentally, and recommendations are provided to further reduce this. The minimum operating power of the proposed circuit is shown to be a strong function of capacitive surge current, and surge current associated with undefined logic states. The transient contribution of components to the quiescent power is analysed through experiment. Simulation shows that increased temperature not only reduces the minimum operating power, but also reduces the energy available to the load.

Maximum Power Transfer Tracking for Ultralow-Power Electromagnetic Energy Harvesters

This paper describes the design and operation of power conditioning system with maximum power transfer tracking (MPTT) for low-power electromagnetic energy harvesters. The system is fully autonomous, starts up from zero stored energy, and actively rectifies and boosts the harvester voltage. The power conditioning system is able to operate the harvester at the maximum power point against varying excitation and load conditions, resulting in significantly increased power generation when the load current waveform has a high peak-to-mean ratio. First, the paper sets out the argument for MPTT, alongside the discussion on the dynamic effects of varying electrical damping on the mechanical structure. With sources featuring stored energy, such as a resonant harvester, maximum power point control can become unstable in certain conditions, and thus, a method to determine the maximum rate of change of electrical damping is presented. The complete power conditioning circuit is tested with an electromagnetic energy harvester that generates 600 mV rms ac output at 870 μW under optimum load conditions, at 3.75 m·s-2 excitation. The digital MPTT control circuit is shown to successfully track the optimum operating conditions, responding to changes in both excitation and the load conditions. At 2 V dc output, the total current consumption of the combined ancillary and control circuits is just 22 μA. The power conditioning system is capable of transferring up to 70% of the potentially extractable power to the energy storage.

Energy harvesting study on single and multilayer ferroelectret foams under compressive force

Cellular polypropylene (PP) ferro electret is a thin and flexible cellular polymer foam that generates electrical power under mechanical force. This work investigates single and multilayer ferro electret PP foams and their potential to supply energy for human-body-worn sensors. Human foot-fall is emulated using an electrodynamic instrument, allowing applied compressive force and momentum to be correlated with energy output. Peak power, output pulse duration, and energy per strike is derived experimentally as a function of force and momentum, and shown to be a strong function of external load resistance, thus providing a clear maximum energy point. The possibility of increasing pulse time and reducing voltage to CMOS compatible levels at some expense of peak power is shown. To further increase the output power, multilayer ferro electret is presented. The synchronized power generation of each layer is studied and illustrated using simulation, and results are supported by experiments. Finally, the energy output of single-layer and multi-layer ferro electrets are compared by charging a capacitor via a rectifier. A ten-layer ferro electret is shown to have charging ability 29.1 times better than that of the single-layer ferro electret. It demonstrates energy output that is capable of powering the start-up and transmission of a typical low-power wireless sensor chipset.

A new circuit topology for floating High Voltage level shifters

A novel and simple circuit topology is presented for high-speed, floating, high voltage level shifters. It uses a current mirror plus latch circuit composed of two inverters. Simulations based on AMS 0.18 μm High Voltage (HV) CMOS Technology show this circuit to combine high speed, low power dissipation, and small layout area. The simulation results show the propagation delay to be below 150 ps for a transition from 1.8 V to 13.8 V.

Experimental investigation of inductorless, single-stage boost rectification for sub-mW electromagnetic energy harvesters

This paper demonstrates single-stage boost rectification for electromagnetic energy harvesters down to approximately 100 μW using practical low-power techniques. The circuits exploit the inductance of the generator, and operate without a discrete inductor, which facilitates integration. Experimental results demonstrate the importance of switching device selection, and the compound effect of the duty ratio on energy harvester output power and converter efficiency, as a function of load current. The circuits demonstrate up to 84.1% harvester utilization at the maximum extractable harvester power of 141 μW, and conversion efficiencies of 73.3% and 59.4% for half- and full-wave operation respectively, neglecting gate drive losses.