Power management for energy harvesting wireless sensors (original) (raw)

Abstract

The objective of this work was to demonstrate smart wireless sensing nodes capable of operation at extremely low power levels. These systems were designed to be compatible with energy harvesting systems using piezoelectric materials and/or solar cells. The wireless sensing nodes included a microprocessor, on-board memory, sensing means (1000 ohm foil strain gauge), sensor signal conditioning, 2.4 GHz IEEE 802.15.4 radio transceiver, and rechargeable battery. Extremely low power consumption sleep currents combined with periodic, timed wake-up was used to minimize the average power consumption. Furthermore, we deployed pulsed sensor excitation and microprocessor power control of the signal conditioning elements to minimize the sensors' average contribution to power draw. By sleeping in between samples, we were able to demonstrate extremely low average power consumption. At 10 Hz, current consumption was 300 microamps at 3 VDC (900 microwatts); at 5 Hz: 400 microwatts, at 1 Hz: 90 microwatts. When the RF stage was not used, but data were logged to memory, consumption was further reduced. Piezoelectric strain energy harvesting systems delivered ~2000 microwatts under low level vibration conditions. Output power levels were also measured from two miniature solar cells; which provided a wide range of output power (~100 to 1400 microwatts), depending on the light type & distance from the source. In summary, system power consumption may be reduced by: 1) removing the load from the energy harvesting & storage elements while charging, 2) by using sleep modes in between samples, 3) pulsing excitation to the sensing and signal conditioning elements in between samples, and 4) by recording and/or averaging, rather than frequently transmitting, sensor data.

Figures (10)

Figure 1. System block diagram for energy harvesting wireless sensing node with data logging and bi- directional RF communications capabilities. The processor manages power to the sensors and data acquisition elements, as well as responds to commands from the base station. A system level block diagram of the energy harvesting wireless sensing system is provided in Figure 1. A modular design approach provides a flexible and versatile platform to address the needs of a variety of applications ”.

Figure 1. System block diagram for energy harvesting wireless sensing node with data logging and bi- directional RF communications capabilities. The processor manages power to the sensors and data acquisition elements, as well as responds to commands from the base station. A system level block diagram of the energy harvesting wireless sensing system is provided in Figure 1. A modular design approach provides a flexible and versatile platform to address the needs of a variety of applications ”.

Figure 2. SG-Link ™ wireless strain gauge node (MicroStrain. Inc) Figure 2 provides a photo of a miniature, strain gauge module as developed by the authors and used in our energy harvesting demonstrations. Accompanying base stations are commercially available with serial (USB) and analog output interfaces. The microprocessor (MicroChip, Chandler, Arizona) is capable of micropower sleep modes. During sleep, the timer function is maintained, but the quiescent power level is reduced to only 20 microwatts. Therefore, for energy harvesting wireless sensors, it is a great advantage to place the processor in sleep mode as frequently as the application will allow, in order to reduce the system’s average power consumption.

Figure 2. SG-Link ™ wireless strain gauge node (MicroStrain. Inc) Figure 2 provides a photo of a miniature, strain gauge module as developed by the authors and used in our energy harvesting demonstrations. Accompanying base stations are commercially available with serial (USB) and analog output interfaces. The microprocessor (MicroChip, Chandler, Arizona) is capable of micropower sleep modes. During sleep, the timer function is maintained, but the quiescent power level is reduced to only 20 microwatts. Therefore, for energy harvesting wireless sensors, it is a great advantage to place the processor in sleep mode as frequently as the application will allow, in order to reduce the system’s average power consumption.

Electrochemical thin film rechargeable batteries were used to store energy for these demonstrations. These batteries were chosen because they can be continuously trickle charged, they exhibit very low leakage, they suffer no memory effects, and they maintain their capacity after repeated charge/recharge cycles.

Electrochemical thin film rechargeable batteries were used to store energy for these demonstrations. These batteries were chosen because they can be continuously trickle charged, they exhibit very low leakage, they suffer no memory effects, and they maintain their capacity after repeated charge/recharge cycles.

mplementation of strict power management resulted in a significant reduction in the average power consumption of our wireless sensor nodes. Figure 6 below plots the average current (in microamperes) for our wireless strain sensor nodes as a function of the sensor’s sampling rate, from 0 to 10 Hz. The supply voltage for these nodes is +3 volts DC. Note that the radio link dominates the current consumption, and is particularly comsumptive at the higher update rates. At sampling rates of 10 Hz, these systems consume ~275 microamps, (less than 900 microwatts at 3 VDC). At update rates of 1 Hz, the power consumption dropped to only 30 microamps (90 microwatts at 3 VDC). Note that wireless sensing systems that do not sleep between samples would draw significantly more current (45,000 microwatts while streaming over the air, and 5000 microwatts while logging only).

mplementation of strict power management resulted in a significant reduction in the average power consumption of our wireless sensor nodes. Figure 6 below plots the average current (in microamperes) for our wireless strain sensor nodes as a function of the sensor’s sampling rate, from 0 to 10 Hz. The supply voltage for these nodes is +3 volts DC. Note that the radio link dominates the current consumption, and is particularly comsumptive at the higher update rates. At sampling rates of 10 Hz, these systems consume ~275 microamps, (less than 900 microwatts at 3 VDC). At update rates of 1 Hz, the power consumption dropped to only 30 microamps (90 microwatts at 3 VDC). Note that wireless sensing systems that do not sleep between samples would draw significantly more current (45,000 microwatts while streaming over the air, and 5000 microwatts while logging only).

Figure 6. Average current consumed by first generation wireless strain sensing node as a function of sensor update rate. First Generation Power Consumption Chart - 1000 Ohm Gauge Additional effort was made to decrease the power consumption at data rates of 5 Hz and higher. This was accomplished by using faster components in the signal conditioning elements of the sensor circuit. In order to bette

Figure 6. Average current consumed by first generation wireless strain sensing node as a function of sensor update rate. First Generation Power Consumption Chart - 1000 Ohm Gauge Additional effort was made to decrease the power consumption at data rates of 5 Hz and higher. This was accomplished by using faster components in the signal conditioning elements of the sensor circuit. In order to bette

Figure 7. Average current consumed by first generation wireless strain sensing node as a function of sensor update rate, with RF section eliminated for clarity. First Generation Power Consumption Chart - 1000 Ohm Gauge

Figure 7. Average current consumed by first generation wireless strain sensing node as a function of sensor update rate, with RF section eliminated for clarity. First Generation Power Consumption Chart - 1000 Ohm Gauge

Figure 8. Average current consumed by second generation wireless strain sensing node as a function « sensor update rate, with RF section eliminated for clarity.

Figure 8. Average current consumed by second generation wireless strain sensing node as a function « sensor update rate, with RF section eliminated for clarity.

Figure 9. Output power for piezoelectric energy harvesting demonstration systems as a function of strain level Sutput Power vs. Frequency and Applied Strain Energy Harvesting Demonstration. The results for output power developed by the piezoelectric energy harvester are plotted below in Figure 9. Output power is plotted as a function of maximum strain on the piezoelectric harvesting element. The resonant tapered beam structure delivered significantly more power than a strip of material mounted on a uniform composite beam loaded in three point bending. This underscores the importance of insuring that the piezoelectric material is subject to a uniform strain field. Another important feature of the resonant cantilever flexure element harvester is that it generated a relatively high amount of out put power at low input vibration levels (100 to 130 milliG’s) and modest strain levels (+/- 200 microstrain).

Figure 9. Output power for piezoelectric energy harvesting demonstration systems as a function of strain level Sutput Power vs. Frequency and Applied Strain Energy Harvesting Demonstration. The results for output power developed by the piezoelectric energy harvester are plotted below in Figure 9. Output power is plotted as a function of maximum strain on the piezoelectric harvesting element. The resonant tapered beam structure delivered significantly more power than a strip of material mounted on a uniform composite beam loaded in three point bending. This underscores the importance of insuring that the piezoelectric material is subject to a uniform strain field. Another important feature of the resonant cantilever flexure element harvester is that it generated a relatively high amount of out put power at low input vibration levels (100 to 130 milliG’s) and modest strain levels (+/- 200 microstrain).

Figure 10. Typical output power and charge current curves for photovoltaic energy harvesting demonstration system as a function of photovoltaic cell output voltage The solar panels produced optimization charge curves typical of the plot provided in Figure 10, below. Table II summarizes the output power generated by the pair of solar panels we tested (Panasonic model BP-243318) under various proximity conditions and under various light sources.

Figure 10. Typical output power and charge current curves for photovoltaic energy harvesting demonstration system as a function of photovoltaic cell output voltage The solar panels produced optimization charge curves typical of the plot provided in Figure 10, below. Table II summarizes the output power generated by the pair of solar panels we tested (Panasonic model BP-243318) under various proximity conditions and under various light sources.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (10)

  1. REFERENCES
  2. C.P. Townsend, S.W. Arms, Wireless Sensor Networks: Principles and Applications, Sensor Technology Handbook, Editor: Jon S. Wilson, publisher: Elsevier Newnes, ISBN: 0-7506-7729-5, Chapter 22, pp. 575-589, 2005
  3. C.P. Townsend, S.W. Arms, M.J. Hamel.; "Remotely Powered, Multichannel, Microprocessor Based Telemetry systems for Smart Implantable devices and Smart Structures", Proc. SPIE Vol. 3673, p. 150-156, Smart Structures and Materials 1999: Smart Electronics and MEMS; Vijay K. Varadan; Ed
  4. D.L. Churchill et al., Strain Energy Harvesting for Wireless Sensor Networks, Smart Structures and Materials, SPIE, vol. 5005, pp. 319-327, 2003
  5. M.J. Hamel et al., Energy Harvesting for Wireless Sensor Operation and Data Transmission, US Patent Appl. Publ. US 2004/0078662A1, filed March 2003
  6. S.W. Arms et al., Shaft Mounted Energy Harvesting System for Wireless Sensor Operation and Data Transmission, US Patent Appl. Publ. US 2005/0017602A1, filed Jan 2004
  7. S. Roundy et al., Energy Scavenging for Wireless Sensor Networks with Special Focus on Vibrations, Kluwer Academic Press, 2004.
  8. N.E. DuToit, B.L. Wardle, Sang-Gook Kim; Design Considerations for MEMS-Scale Piezoelectric Mechanical Vibration Energy Harvesters, XIII International Mat'ls Res. Congress (IMRC) Symp. 9 -Ferroelectricity & Piezoelectricity, 2004
  9. J.A. Paradiso & T. Starner, Energy Scavenging for Mobile and Wireless Electronics, Pervasive Computing, IEEE CS and IEEE ComSoc, Vol 1536-1268, pp 18-26, 2005
  10. Galbreath J.H, Townsend, C.P., Mundell, S.W., Hamel M.J., Esser B., Huston, D., Arms, S.W. "Civil Structure Strain Monitoring with Power-Efficient High-Speed Wireless Sensor Networks", International Workshop for Structural Health Monitoring, Stanford, CA, September 2003