Energy harvesting applications are everywhere

Background and History
The concept of energy harvesting has been around for more than 10 years, but in real-world environments, systems powered by ambient energy have been bulky, complex, and expensive. However, some markets have successfully adopted energy harvesting methods, such as transportation infrastructure, wireless medical devices, tire pressure monitoring, and building automation markets. In particular, in building automation systems, such as occupancy sensors, thermostats, and even light-operated switches, which were previously installed without power or control wiring, now use local energy harvesting systems instead.

One major application for energy harvesting systems is wireless sensors in building automation systems. To illustrate, consider the distribution of energy use in the United States. Buildings are the number one user of energy production each year, accounting for approximately 38% of total energy consumption, followed by transportation and industry, each accounting for 28% of total energy consumption. Furthermore, buildings can be further divided into commercial buildings and residential buildings, which account for 17% and 21% of the 38% of energy consumption, respectively. The 21% figure for residential buildings can be further divided, with heating, ventilation and air conditioning (HVAC) accounting for approximately three-quarters of the total energy consumption for residential buildings. With energy use currently expected to double between 2003 and 2030, the use of building automation systems could save up to 30% of energy (source: "World Energy, Technology and Climate Policy Outlook (WETO)", jointly written by several EU research institutions).

Similarly, a wireless network using energy harvesting can connect any number of sensors in a building to adjust the temperature or turn off lights in non-critical areas when the building or rooms are unoccupied, thereby reducing HVAC and electricity costs. In addition, the cost of energy harvesting electronics is often lower than the cost of running power lines or the routine maintenance costs required to replace batteries, so there is a clear economic benefit to using harvested energy for powering.

However, if each node requires its own external power source, many wireless sensor networks lose their advantage. While power management technology does continue to advance and has enabled electronic circuits to operate longer on a given power source, this is limited, and powering with harvested energy provides a complementary approach. Therefore, energy harvesting is a method to power wireless sensor nodes by converting local ambient energy into usable electrical energy. Ambient energy includes light, temperature differences, vibrating beams, transmitted RF signals, or any energy that can generate an electric charge through a transducer. These energy sources are all around us and can be converted into electrical energy using suitable transducers, such as thermoelectric generators (TEGs) for temperature differences, piezoelectric components for vibrations, photovoltaic cells for sunlight (or indoor lighting), and even electricity generated by humid gases. These so-called "free" energy sources can be used to autonomously power electronic components and systems.

All wireless sensor nodes can now operate at microwatt average power, so it is feasible to power them with non-traditional power sources. This has led to the emergence of energy harvesting, which can be used to charge, supplement or replace batteries in systems where using batteries is inconvenient, impractical, expensive or dangerous. Powering with harvested energy can also eliminate the need for wires to provide power or transmit data. In addition, energy generated by industrial processes, solar panels or internal combustion engines can be harvested and used, which would otherwise be wasted.

However, existing implementations of energy harvesting power conversion circuits typically consist of low-performance and complex discrete configurations, often including 30 or more components. Such designs have low conversion efficiency and high quiescent current. Both of these drawbacks result in compromised performance of the final system. Low conversion efficiency increases the time required to power up the system, which in turn increases the time interval between taking a sensor reading and transmitting that data. High quiescent current limits the lowest value that the output of the energy harvesting power supply can reach, as it must first provide the current required for its own operation before it can provide additional power to the output.

Problems and Characteristics of Energy Harvesting Applications
A typical energy harvesting configuration or wireless sensor node (WSN) consists of four blocks (see Figure 1). They are: 1) an ambient energy source; 2) a transducer component and power conversion circuitry to power downstream electronics; 3) a sensing component that connects the node to the real world and a computing component (consisting of a microprocessor or microcontroller that processes the measurements and stores the data in memory); and 4) a communication component consisting of a short-range radio unit that enables wireless communication with neighboring nodes and the outside world.

Examples of ambient energy sources include a thermoelectric generator (TEG) or thermopile connected to a heat source such as an HVAC duct, or a piezoelectric transducer connected to a mechanical vibration source such as a window pane. In the case of a heat source, a compact thermoelectric device (often called a transducer) converts a small temperature difference into electrical energy. In the presence of mechanical vibration or pressure, a piezoelectric device can be used to convert mechanical energy into electrical energy.

Once generated, the energy can be converted and conditioned by energy harvesting circuits into a suitable form to power downstream electronics. Thus, a microprocessor can wake up a sensor to take a reading or measurement, which can then be processed by an analog-to-digital converter for transmission via an ultra-low power wireless transceiver.

Figure 1: Block diagram of the main components of a typical energy harvesting system or wireless sensor node

There are several factors that affect the power consumption characteristics of energy harvesting systems in wireless sensor nodes. Table 1 summarizes these factors.

Table 1: Factors affecting power consumption of wireless sensor nodes

Of course, the amount of energy a harvested source delivers depends on how long it can operate. Therefore, the primary metric for comparing harvested sources is power density, not energy density. Energy harvesting typically encounters low, variable, and unpredictable available power, so a hybrid architecture is used that connects to a harvester and a secondary power bank. The harvester becomes the system energy source due to its unlimited energy supply and power shortage. The secondary power bank (either a battery or a capacitor) produces higher output power, but stores less energy, providing power when needed and otherwise receiving charge from the harvester periodically.

The most advanced and readily available energy harvesting technologies, such as vibration energy harvesting and indoor photovoltaics, produce milliwatts of power under typical operating conditions. Although such low powers may seem limiting, several years of work with harvesting components have shown that these technologies are generally similar to long-life primary batteries in terms of both energy supply and cost per unit of energy provided. In addition, systems using energy harvesting can generally be recharged after depletion, which is not possible with primary battery-powered systems.

As already discussed, ambient energy sources include light, temperature differences, vibrating beams, transmitted RF signals, or any other energy source that can generate an electrical charge through a transducer. Table 2 below illustrates how much energy can be generated from different energy sources.

Table 2: Energy sources and how much energy they produce

To successfully design a complete, self-contained wireless sensor system, a steady supply of power-saving microcontrollers and transducers are required, and these devices are required to consume the lowest possible power and come from a low-energy environment. Fortunately, low-cost, low-power sensors and microcontrollers have been on the market for about two or three years, but only recently have ultra-low-power transceivers become commercially available. However, the laggard in this series of links has always been energy harvesters. It is in this area of ​​energy harvesters that Linear Technology's recent products, the LTC3109, LTC3588-1 and LTC3105, have brought a new level of performance and simplicity.

A real-world example: Aircraft health monitoring

Structural fatigue in today’s large fleet of aircraft is a real problem that could lead to catastrophic consequences if ignored. Currently, aircraft structural health is monitored through a variety of inspection methods, such as through improved structural analysis and tracking methods, through the use of innovative concepts to assess structural integrity… and more. These methods are sometimes collectively referred to as “aircraft health monitoring” methods. In the aircraft health monitoring process, sensors, artificial intelligence and advanced analytical methods are used to conduct continuous health assessments in real time.

Acoustic emission testing is the leading method for locating and monitoring cracks in metallic structures. This method can be easily used to diagnose damage to composite aircraft structures. An obvious requirement is to indicate structural integrity in a simple "pass", "fail" form, or to take immediate repair action. This testing method uses a flat test sensor consisting of a piezoelectric chip and an optical sensor, which is sealed with a polymer film. The sensor is firmly mounted to the surface of the structure, and triangulation can be used to locate the acoustic activity of the structure on which the sensor is mounted. The sensor data is then captured by the instrument and parameterized in a form suitable for narrowband storage and transmission.

As a result, wireless sensor modules are often embedded in various parts of aircraft, such as wings or fuselages, for structural analysis, but powering these sensors can be complex. Therefore, these sensor modules can be more convenient and efficient if they can be powered wirelessly or even self-powered. In the aircraft environment, there are many "free" energy sources that can be used to power such sensors. Two obvious and convenient methods are thermal energy harvesting and/or piezoelectric energy harvesting.

In the case of a typical aircraft engine, the temperature can range from a few hundred ºC to 1,000 ºC or even 2,000 ºC. Although most of this energy is lost as mechanical energy (combustion and engine thrust), some is still consumed purely as heat. Since the Seebeck effect is the fundamental thermodynamic phenomenon for converting heat into electrical power, the main equation to consider is:

P = ηQ

Where P is electrical power, Q is heat, and η is efficiency.

Larger thermoelectric generators (TEGs) use more heat (Q) to produce more power (P). Similarly, using twice as many power converters naturally produces twice as much power because they can capture twice as much heat. Larger TEGs are formed by connecting more PN junctions in series, but while this produces a larger voltage (mV/dT) when the temperature changes, it also increases the series resistance of the TEG. This increase in series resistance limits the power that can be delivered to the load. Therefore, depending on the application requirements, it is sometimes better to use smaller parallel TEGs rather than using larger TEGs. Regardless of which TEG is chosen, there are many manufacturers that provide commercial TEG products.

By applying pressure to a component, piezoelectricity can be generated, which in turn generates an electric potential. The piezoelectric effect is reversible, and materials that exhibit the direct piezoelectric effect (when pressure is applied, an electric potential is generated) also exhibit the inverse piezoelectric effect (when an electric field is applied, pressure and/or stress is generated).

To optimize a piezoelectric transducer, the vibration frequency and displacement characteristics of the piezoelectric source need to be determined. Once these levels are determined, the piezoelectric component manufacturer can design a piezoelectric component that is mechanically tuned to a specific vibration frequency and sized to deliver the required amount of power. Vibrations in the piezoelectric material trigger the positive piezoelectric effect, which causes charge to accumulate on the output capacitance of the device. The accumulated charge is usually quite small, so the AC open circuit voltage is high, in many cases on the order of 200V. Since the amount of charge generated by each deflection is relatively small, it is necessary to full-wave rectify this AC signal and accumulate charge on an input capacitor cycle by cycle.

In terms of energy selection, there are trade-offs between thermal and piezoelectric sources. However, regardless of which method is chosen, both methods are viable and realistic solutions that can be easily used with existing technology. The following table summarizes the advantages and disadvantages of these two methods:

Note 1: The best way to obtain the temperature difference in an aircraft is to take the difference between the "skin" temperature of the aircraft cabin and the temperature inside the cabin.

The LTC3109 energy harvesting power conversion IC
is a highly integrated DC-DC converter and power manager. It can collect and manage excess energy from very low input voltage sources such as TEGs (thermoelectric generators), thermopiles, and even small solar cells. Its unique proprietary automatic polarity topology allows the device to operate with input supplies as low as 30mV, regardless of the supply polarity.

Figure 2: Typical application schematic of LTC3109

The circuit above uses two compact step-up transformers to increase the voltage of the LTC3109 input voltage source, and the device then provides a complete power management solution for wireless sensing and data acquisition. It can harvest small temperature differences and generate system power without the traditional battery power supply.

For input voltages as low as 30mV, a transformer with a primary-to-secondary turns ratio of approximately 1:100 is recommended. For higher input voltages, lower turns ratios can be used to achieve greater output power. These transformers are standard, off-the-shelf components and are readily available from magnetics suppliers such as Coilcraft.

The LTC3109 takes a “system-level” approach to solving complex problems. It converts low voltage sources and manages energy between multiple outputs. The AC voltage generated on each transformer secondary winding is boosted and rectified using the LTC3109’s external charge pump capacitors and internal rectifiers. The rectifier circuit feeds current into the V _AUX pin, which in turn supplies power to the external V _AUX

capacitors and then the other outputs.

_{The internal 2.2V LDO can support low-power processors or other low-power ICs. The LDO is powered by the higher of VAUX and VOUT. This allows it to operate efficiently as soon as VAUX charges to 2.3V while the VOUT storage capacitor is still charging. If there is a step load on the LDO output, current can be drawn from the main VOUT capacitor if VAUX drops below VOUT. The LDO can provide 3mA output current.}

_{The VSTORE capacitor may be very large (thousands of microfarads or even several farads) to maintain power when the input power may be lost. Once power is up, the main output, backup output, and switching output are available. If the input power fails, operation can still be continued using the power from the VSTORE capacitor.}

_{The LTC3588-1 is a complete energy harvesting solution optimized for low energy sources including piezoelectric transducers. Piezoelectric devices generate energy through the squeezing or flexing of the device. Depending on the size and construction, these piezoelectric elements can generate hundreds of uW/cm2 ^of energy.}

_{Figure 3: Typical application schematic of LTC3588}

_{It should be mentioned that the piezoelectric effect is reversible, ie a material that exhibits the direct piezoelectric effect (electric potential generated upon application of pressure) also exhibits the reverse piezoelectric effect (pressure and/or stress, ie deflection, generated upon application of voltage).}

_{The LTC3588-1 operates over an input voltage range of 2.7V to 20V, making it ideal for a variety of piezoelectric transducers and other high output impedance energy sources. Its high efficiency step-down DC/DC converter provides up to 100mA of continuous output current or even higher pulse loads. Its output can be set to one of four fixed voltages (1.8V, 2.5V, 3.3V or 3.6V) to power wireless transmitters or sensors. When the output is in regulation (no load), the quiescent current is only 950nA, maximizing overall efficiency.}

_{The LTC3588-1 is designed to interface directly with a piezoelectric or alternative high impedance AC source, rectifying the voltage waveform and storing the harvested energy in an external storage capacitor while dissipating excess power through an internal shunt regulator. An ultralow quiescent current (450nA) undervoltage lockout (ULVO) mode with a 1V to 1.4V hysteresis window allows charge to accumulate on the storage capacitor until the buck converter can efficiently transfer a portion of the stored charge to the output.}

_{The LTC3105 is an ultralow voltage step-up converter and LDO designed to greatly simplify the task of harvesting and managing energy from low voltage, high impedance alternative power sources such as photovoltaic cells, thermoelectric generators (TEGs), fuel cells, and more. Its synchronous step-up design starts up with an input voltage as low as 250mV, making it ideal for harvesting energy from even the smallest photovoltaic cells in less than ideal lighting conditions. Its wide input voltage range of 0.2V to 5V makes it ideal for a variety of applications. An integrated maximum power point controller (MPPC) enables the LTC3105 to extract the maximum available power that the source can provide. Without MPPC, the source can only produce a fraction of the theoretical maximum. Peak current limit automatically adjusts to maximize power conversion efficiency, while Burst Mode® operation reduces quiescent current to only 22uA, minimizing leakage current from energy storage elements. The ultralow IQ LDO can directly power popular low power microcontrollers or sensor circuits.}

_{The circuit shown in Figure 4 uses the LTC3105 to charge a single lithium-ion battery from a single photovoltaic cell. The circuit allows the battery to be continuously charged while solar energy is available, and the battery can use the stored energy to power the application when solar energy is no longer available.}

_{Figure 4: Li-ion battery trickle charger using a single photovoltaic cell}

_{The LTC3105 can start up with voltages as low as 250mV. At startup, the AUX output is initially charged with the synchronous rectifiers disabled. Once VAUX reaches approximately 1.4V, the converter leaves startup mode and enters normal operation. Maximum power point control is not enabled at startup, however, current is internally limited to a low enough level to allow startup from very low current input supplies. While the converter is in startup mode, the internal switch between AUX and VOUT remains disabled and the LDO is not used. See Figure 5 for an example of a typical startup sequence.}

_{When VIN or VAUX is above 1.4V, the converter enters normal operation. The converter continues to charge the AUX output until the LDO output enters regulation. Once the LDO output enters regulation, the converter begins charging the VOUT pin. VAUX remains high enough to ensure that the LDO is in regulation. If VAUX is higher than required to keep the LDO stable, the current switches from charging the AUX output to charging the VOUT output. If VAUX drops too much, the current flows back to the AUX output instead of being used to charge the VOUT output. Once VOUT rises above VAUX, an internal switch is activated to connect the two outputs together.}

_{If VIN is higher than the voltage on the driven output (VOUT or VAUX), or the driven output is less than 1.2V, the synchronous rectifiers are disabled and operate in critical conduction mode, allowing regulation even when VIN > VOUT.}

_{If the output voltage is higher than the input voltage and is above 1.2V, the synchronous rectifier is enabled. In this mode, the N-channel MOSFET between SW and GND is enabled until the inductor current reaches the peak current limit. Once the current limit is reached, the N-channel MOSFET is turned off and the P-channel MOSFET between SW and the driven output is enabled. The switch remains on until the inductor current drops below the valley current limit, and the cycle repeats. When VOUT reaches the regulation point, both the N-channel and P-channel MOSFETs connected to the SW pin are disabled and the converter enters a sleep state.}

_{Figure 5: Typical LTC3105 startup sequence}

_{To power the microcontroller and external sensors, an integrated LDO provides a regulated 6mA rail. The LDO is powered by the AUX output, allowing the LDO to reach regulation while the main output is still charging. The output voltage of the LDO can be fixed at 2.2V or adjustable via a resistor divider.}

The integrated maximum power point control circuit allows the user to set the optimal input voltage operating point for a given power supply, see Figure 6. The MPPC circuit dynamically adjusts the average inductor current to prevent the input voltage from falling below the MPPC threshold. When VIN is above the MPPC voltage, the inductor current increases until VIN is pulled down to the MPPC set point. If VIN is below the MPPC voltage, the inductor current decreases until VIN increases to the MPPC set point.

Figure 6: Typical maximum power point control points for a single photovoltaic cell

The LTC3105 incorporates features that maximize efficiency at light loads, while also enhancing the ability to deliver power at heavy loads by adjusting the inductor peak and valley currents as a function of load. At light loads, reducing the inductor peak current to 100mA reduces conduction losses, thereby optimizing efficiency. As the load increases, the inductor peak current automatically increases to 400mA (maximum). At medium loads, the inductor peak current may vary between 100mA and 400mA. These features have a lower priority than the MPPC feature and only come into play when the power provided by the power supply exceeds the load requirements.

In applications such as photovoltaic conversion, the input power may not be present for long periods of time. To prevent output discharge in such situations, the LTC3105 incorporates an undervoltage lockout (UVLO) function that forces the converter into shutdown mode if the input voltage drops below 90mV (typical). In shutdown mode, the switch connecting AUX and VOUT is enabled, the LDO is placed in reverse isolation mode, and the current flowing into VOUT is reduced to 4uA (typical). In shutdown mode, the reverse current through the LDO is limited to 1uA to minimize output discharge.

Conclusion
Power management is a critical aspect of implementing remote wireless sensing. However, power management must be implemented from the design concept. Therefore, system designers and system planners must prioritize power management needs from the beginning to ensure efficient design and successful long-term deployment. Fortunately, a growing number of energy harvesting power management ICs are now available from leading high-performance analog IC manufacturers, greatly simplifying this task.