Passive, proximity-based electric current sensors: Passive, proximity-based electric current sensors Energy and environment Energy efficiency Design MEMS A research topic related to my primary academic interests: Eli Leland’s Qualifying Exam May 4th, 2007


Research and publication history: Research and publication history TinyTemp: A wireless temperature sensor powered by vibrations from a wooden staircase Publication: E. Leland, E. Lai, P. Wright “A Self-Powered Wireless Sensor for Indoor Environmental Monitoring,” WNCG Wireless Networking Symposium, Austin, Texas, October 2004 Resonance tuning: Adjust the resonance frequency of a vibration scavenger to broaden deployment possibilities Publication: E. Leland, P. Wright “Resonance tuning of piezoelectric vibration energy scavenging generators using compressive axial preload,” Smart Materials and Structures, vol. 15 (2006) pp. 1413-1420 Resonance frequency vs. preload


PhD research: current sensing for Demand Response: PhD research: current sensing for Demand Response Design passive, proximity-based current sensors for homes and buildings Passive sensors require no external power, dramatically extend life of wireless sensor node Proximity-based doesn’t require electrical connection or wraparound, doesn’t require precise alignment to conductor, potentially integrated with CPU and radio on a single piece of silicon Demand Response: mechanisms that encourage electricity consumers to shift their usage away from periods of peak demand Focus on currents in and around homes and other buildings


Existing technology: Existing technology No MEMS current sensor on the market No passive (self-powered) sensor on the market that doesn’t require wraparound Kill-A-Watt shunt resistor-based in-line Current Transformer self-powered requires wraparound – impractical for many applications Hall Sensor proximity-based requires 10s of mW of power Rogowski Coil voltage scales with square of linear dimension so small scales = small voltages difficult to microfabricate a coil of many turns


Design concept: Permanent magnets and piezoelectric Materials: Design concept: Permanent magnets and piezoelectric Materials Permanent magnets can couple to the magnetic fields surrounding AC current carriers Piezoelectric materials can transduce the forces on the permanent magnet to an output voltage


Literature review: Literature review Measuring current C. Xiao, et al., “An Overview of Integratable Current Sensor Technologies,” Proc. 38th IAS Annual Meeting (2003) pp. 1251-1258 P. Ripka, “Current Sensors using Magnetic Materials,” Journal of Optoelectronics and Advanced Materials, vol. 6 no. 2 (2004) pp. 587-592 MEMS magnetometers Beroulle, et al., “Monolithic piezoresistive CMOS magnetic field sensors,” Sensors and Actuators A, vol. 103 (2003) pp. 23-32 H.H. Yang, et al., “Ferromagnetic micromechanical magnetometer,” Sensors and Actuators A, vol. 97-98 (2002) pp. 88-97 J. Liu, X. Li, “A piezoresistive microcantilever magnetic field sensor with on-chip self-calibration function integrated,” Microelectronics Journal, vol. 38 (2007) pp. 210-215 Beroulle, et al. (2003)


Literature review, continued: Literature review, continued Piezoelectric cantilever output S. Roundy, P. Wright, “A piezoelectric vibration based generator for wireless electronics,” Smart Materials and Structures vol. 13 (2004) pp.1131-1142 S.N. Chen, et al., “Analytical modeling of piezoelectric vibration-induced micro power generator,” Mechatronics vol. 16 (2006) pp. 379-387 Magnetic microactuators/sensors B. Wagner, W. Benecke, “Microfabricated actuator with moving permanent magnet,” Proc. MEMS ’91 (1991) pp. 27-32 Lagorce, et al., “Magnetic Microactuators based on Polymer Magnets,” J. of MEMS vol. 8, no. 1 (1999) pp. 2-9


Initial prototypes show promising behavior: Initial prototypes show promising behavior Sensor mounted on a single-conductor power cable Current sensor response – varying distance from power cable


Can we power the sensor node without a battery?: Can we power the sensor node without a battery? Device scavenged 350 microwatts from a standard 1500 W space heater appliance cord, sufficient to power a commercially-available wireless sensor node at a 1% duty cycle Tuned to 60 Hz resonance frequency for maximum coupling and power output Energy scavenging power output – space heater power cord Publication: E. S. Leland, R. M. White, P. K. Wright “Energy scavenging power sources for household electrical monitoring,” PowerMEMS 2006, November 29–December 1, 2006, Berkeley, California


Current sensor design questions: Current sensor design questions How sensitive will the current sensor be? (in V/A) How do I calculate the forces on a permanent magnet in a magnetic field? How does force on the magnet translate to voltage out of the piezoelectric bimorph transducer? How well will this sensor’s performance scale downward to smaller sizes?


Wires and Magnetic Fields: Wires and Magnetic Fields Electric power is 60 Hz AC in the Americas, 50 Hz in Europe Voltage and current are sinusoidal – rated value is root-mean-square (rms) Magnetic field surrounding a current-carrying wire is circumferential (right-hand rule) and alternating B


What do these fields look like?: What do these fields look like? One wire Circumferentially directed, right hand rule Strength decreases with inverse of distance to wire Two-wire (zip-cord) Currents 180° out of phase Fields add along vertical line at center Fields cancel as distance increases H-field magnitude units in A/m (color bars)


What are the forces on the magnet?: What are the forces on the magnet? Force equals the magnetization multiplied by the volume integral of the gradient of the H-field over the magnet’s volume For a permanent magnet, the magnetization M can be replaced with the residual flux density Br Br and V are constant. Is the field gradient proportional to the current? B. Wagner, W. Benecke, “Magnetically driven microactuators: Design considerations,” Microsystem Technologies ’90, H. Reichl (Ed.), Springer-Verlag 1990


Where will the force on the magnet be largest?: Where will the force on the magnet be largest? Force is proportional to magnetic field gradient To generate maximum vertical force, magnetization vector should make ±45° angle with radial vector Gradient magnitude along line varies with 1/y2 H-field gradient magnitude units in A/m2 (color bars)


Empirical tests confirm predictions about the forces around a wire: Empirical tests confirm predictions about the forces around a wire Magnet centered above wire: ~0.5 Vrms Magnet along diagonal: 11-12 Vrms Magnet beside wire: ~0.5 Vrms Simple experiment measuring voltage output of a 60 Hz sensor placed near a single wire carrying 13.4 Arms Theory predicts the greatest force when magnetization vector makes 45° angle with radial vector


What about the forces around a zip-cord?: What about the forces around a zip-cord? Conveniently, optimal magnet placement for vertical force generation is on vertical line at center H-field gradient magnitude units in A/m2 (color bars)


Where is the force above the heater cord largest?: Where is the force above the heater cord largest? Gradient has maximum absolute value at y = ±d/(√3) where d is half the distance between the wires For a 16AWG heater cord this is about y ≈ 1.05 mm (still inside insulation) Maximum absolute value of gradient at y = ±d/(√3) is roughly equal to 0.207(i/d2) At twice this height, gradient has about 65% of its maximum value At three times this height, gradient has about 33% of its maximum value insulation


More empirical confirmation of where the biggest forces are: More empirical confirmation of where the biggest forces are Centered above heater cord: 12 Vrms Beside heater cord: ~0.5 Vrms Centered above one wire: ~0.5 Vrms Above and to the side: 10 Vrms


So how big are these forces?: So how big are these forces? You can feel them!


What is the voltage out of the piezoelectric sensing element?: What is the voltage out of the piezoelectric sensing element? Constitutive Equations S = strain (non-dimensional) T = stress (Pa) cp= Young’s modulus (Pa) d = piezoelectric coefficient (m/V) D = dielectric displacement (C/m2) e = dielectric permittivity (F/m) E = electric field (V/m) Usable Modes of PZT


Force on the magnet generates voltage in piezoelectric element: Force on the magnet generates voltage in piezoelectric element S. Roundy, P. Wright, “A piezoelectric vibration based generator for wireless electronics,” Smart Materials and Structures, vol. 13 (2004) pp. 1131-1142 Using electro-mechanical equivalents, applying Kirchhoff’s Voltage Law to the left and Kirchhoff’s Current Law to the right: To determine equations of motion, we need to find equivalent expressions for Tin, Lm, Rb, Ck, n, i, Cb KVL: KCL: magnet force on magnet


Voltage output is proportional to force input: Voltage output is proportional to force input Voltage frequency response function: The absolute value of this complex function gives the output magnitude, and the angle gives the phase shift between magnetic force and voltage out For a constant frequency input of 60 Hz (w = 2p×60), the voltage output is a linear function of magnetic input force


Voltage model experiments: Voltage model experiments A series of known input forces were applied using the shake table (F=ma) and output voltage was measured Using d31 = -138 pm/V (19-27% lower than datasheet values for PZT-5A*), model closely matches experimental data photo of sensor on shaker *sample values of d31 for PZT-5A: -190x10-12 m/V from Piezo Systems datasheet, -170x10-12 m/V from Morgan Electroceramics datasheet


Using the voltage model to validate the force/current model: Using the voltage model to validate the force/current model The force modeling predicted a value for force/current, let’s call it b This value of b was substituted into the voltage-force model along with the same value of d31 = -138 pm/V determined previously As the relationship between force and voltage was already established, these experimental results support the force/current model


How will this all scale downward: How will this all scale downward Trimmer’s “S” analysis to determine how output will scale with decrease in linear dimension W. S. N. Trimmer, “Microrobots and Micromechanical Systems,” Sensors and Actuators A vol. 19 (1989) pp. 267-287 So the sensor’s output will scale as [S2], or with the square of linear dimension.


Simulations indicate microscale versions should produce measurable voltages: Simulations indicate microscale versions should produce measurable voltages With sensitivities 10x those of PZT, model results suggest AlN is worth a closer look as an active material. Notes: Prototype bimorph had two PZT layers, all others have only one active layer. PZT properties: d31 = -138 pm/V, er = 1800, density = 7800 kg/m3, cp = 66 GPa. AlN properties: d31 = -7.5 pm/V, er = 9, density = 3200 kg/m3, cp = 135 GPa. NdFeB properties: density = 7500 kg/m3, Br = 1.3 T for the leftmost two beams, 0.5 T for the rest. Steel properties: cp = 200 GPa, density = 7800 kg/m3. Pt properties: cp = 171 GPa, density = 21450 kg/m3.


Major design take-aways: Major design take-aways Current sensor output voltage is linearly proportional to current, as predicted by theory and demonstrated by experiment Magnetic force is proportional to current Voltage out proportional to force Magnetic force proportional to magnet volume, remanent magnetization Theoretical models suggest microscale devices should be feasible


Research Plan Going Forward: Research Plan Going Forward Fabricate AlN cantilever devices in the Microlab Characterize AlN cantilevers to verify voltage/force model at the microscale Identify most promising method to fabricate micromagnets


Special thanks: Special thanks ASEE/NDSEG My committee Profs. Paul Wright and Dick White BMI/Ford lab past and present, especially Mike Koplow and Dr. Shad Roundy


Additional slides: Additional slides


Course list: Course list Design ME 134: Controls ME 221: High-tech Product Design ME 224: Mechanics of Materials ME 273: Vibrations IEOR 170: Industrial Design and Human Factors MEMS ME 119: Introduction to MEMS EE 245: MEMS Design ME 219: Optimal MEMS design Energy and Environment ER 100: Intro. to Energy and Resources MSE 226: Photovoltaic Materials ER 291: Wind Power ARCH 240: Energy Efficient Building Design MBA 212: Corporate Environmental Policy


Research History: TinyTemp: Research History: TinyTemp First self-powered temperature sensor Harvested vibrations from wooden staircase Piezoelectric vibration energy scavenging 450mw power output 42% circuit efficiency (non-optimized) E. Leland, E. Lai, P. Wright “A Self-Powered Wireless Sensor for Indoor Environmental Monitoring,” WNCG Wireless Networking Symposium, Austin, Texas, October 2004 Power transfer of 108mW


Resonance Tuning for Energy Scavenging: Motivation: Resonance Tuning for Energy Scavenging: Motivation Cantilever-mount energy scavengers only work effectively at a single frequency dictated by design parameters. A device that produces usable power across a range of driving frequencies will greatly enhance the viability of vibration energy scavenging.


Common sources of vibrations: Common sources of vibrations Vibration sources abound, but they comprise a diverse set of frequencies and magnitudes Roundy, et al, “A study of low-level vibrations as a power source for wireless sensor nodes,” Computer Communications, vol. 26 (11) 2003, pp. 1131-1144


Resonance tuning for vibration energy scavenging: Theory: Resonance tuning for vibration energy scavenging: Theory Resonance frequency of a simply-supported (“pin-pin”) piezoelectric bimorph can be adjusted downward using compressive axial preload. Theoretical resonance frequency vs. preload, normalized


Resonance Tuning: Results: Resonance Tuning: Results Experiments demonstrated that this design can reduce resonance frequency up to 24% while still providing 65-90% of the nominal “unloaded” power output. E. Leland, P. Wright “Resonance tuning of piezoelectric vibration energy scavenging generators using compressive axial preload,” Smart Materials and Structures vol. 15 (2006) pp. 1413-1420


Other publications: Other publications S. Roundy, E. Leland, J. Baker, E. Carleton, E. Reilly, E. Lai, B. Otis, J. Rabaey, V. Sundararajan and P.K. Wright "Improving Power Output for Vibration-Based Energy Scavengers" IEEE Pervasive Computing Journal on Mobile and Ubiquitous Computing, Volume 4, Number 1, January - March 2005, pp. 28-36. Kate Hammond, Elaine Lai, Eli Leland, Sue Mellers, Dan Steingart, Eric Carleton, Beth Reilly, Jessy Baker, Brian Otis, Jan Rabaey, David Culler, and Paul Wright. "An integrated node for energy-scavenging, sensing, and data-transmission: applications in medical diagnostics ." The Second International Workshop on Body Sensor Networks." Imperial College London April 11-13th 2005. Chris R. Baker, Kenneth Armijo, Simon Belka, Merwan Benhabib, Vikas Bhargava, Nathan Burkhart,Artin Der Minassians, Gunes Dervisoglu, Lilia Gutnik, M. Brent Haick, Christine Ho, Mike Koplow,Jennifer Mangold, Stefanie Robinson, Matt Rosa, Miclas Schwartz, Christo Sims, Hanns Stoffregen, Andrew Waterbury, Eli S. Leland, Trevor Pering, and Paul K. Wright, “Wireless Sensor Networks for Home Health Care,” First International Workshop on Smart Homes for Tele-Health, Niagara Falls, Canada, May 21-23, 2007


More on current sensing, demand response, etc.: More on current sensing, demand response, etc. Use content from citris/lab funding proposals


Derivation of magnetic intensity surrounding a wire: Derivation of magnetic intensity surrounding a wire dl R Rdq q


Heater Cord Magnetic Fields: Heater Cord Magnetic Fields Above the center of the cord is a good place for a sensor Wraparound sensors won’t work on a 2-wire appliance cord


Empirical results: Sensor response around a heater cord: Empirical results: Sensor response around a heater cord Centered above heater cord: 11.5 Vrms Beside heater cord: ~0.5 Vrms Centered above one wire: ~0.5 Vrms Above and to the side: 10 Vrms


Heater cord B-field measurements: Heater cord B-field measurements Measured B-field at top-center of heater cord using Allegro 1392 Hall sensors Excellent agreement between theory and measured results Hall sensors seem to be a good way to measure current, but… Theory: B = 0.94 x Current (gauss)


Why not use Hall sensors?: Why not use Hall sensors? Allegro 1392 Hall sensor 9.6 mW average power during operation, 75 mW sleep Assume duty cycling of 120 samples/sec, 1 millisecond/sample Average power 1.2 mW, ten times the target average power for a sensor node Pico radio paper reference for 100 microwatt node


Derivation of Force on a Magnet: Derivation of Force on a Magnet Go from this: To this (it’s in my notes, just typeset eqns): (for a magnet uniformly magnetized in the y-direction)


More piezo model derivation: More piezo model derivation More typesetting include electro-mechanical analogue table


Piezoelectric bimorph equations of motion: Piezoelectric bimorph equations of motion Fin is the force applied to the tip of the beam ksp is the equivalent spring constant of the beam, relating tip deflection to applied force k2 is a geometric constant relating tip deflection to average strain in the piezoelectric layer(s) a1, a2, and a3 vary depending on whether the beam is a unimorph, a series-poled bimorph, or a parallel-poled bimorph


Improving the model: Improving the model a more compliant pin-pin mounting better approximates the resonance behavior of the piezoelectric beam


How accurate is a lumped parameter model?: How accurate is a lumped parameter model? I’ve developed this continuous model already. just plug in numbers for no tip mass and for tip mass models, lumped parameter models and continuous and find resonance frequencies. compare numbers. no way it’s closer than the pin-pin model on the previous slide


What is the parasitic effect of the current sensor?: What is the parasitic effect of the current sensor? consider closed system containing length of cable and sensor with no sensor, power flow in = flow out, all in cable (steady state) with sensor, sensor is excited to some energy state (kinetic or potential – whichever is easier), with some total energy extraction (electric and mechanical) through damping I measured zeta = 0.024, (be ready to show this data, along with log decrement expression), Q = 1/(2*zeta) and equals total system energy/energy loss per cycle. from that I can get energy loss per cycle and I know cycle time is 1/60th sec this gives an order of magnitude at least


Why does AlN have good sensor potential in spite of lower coupling d31?: Why does AlN have good sensor potential in spite of lower coupling d31? Remember:


Continuous beam model equation: Continuous beam model equation solving free vibration equation for a cantilever with a tip mass (add full solution from rao): wxt = 16*(cos(B*x)-cosh(B*x)-(cos(B*l)+cosh(B*l))/(sin(B*l)+sinh(B*l))*(sin(B*x)-sinh(B*x)))/rho/A/(-4*cosh(B*l)^2*cos(B*l)*exp(B*l)-8*sin(B*l)^2*cosh(B*l)*exp(B*l)-8*sin(B*l)^2*sinh(B*l)*exp(B*l)-4*sin(B*l)*cos(B*l)^2*exp(B*l)-4*sin(B*l)^2*cos(B*l)*exp(B*l)+4*sinh(B*l)^2*cos(B*l)*exp(B*l)-4*sinh(B*l)^2*exp(B*l)*sin(B*l)-4*cosh(B*l)^2*exp(B*l)*sin(B*l)-8*cos(B*l)^2*cosh(B*l)*exp(B*l)-2*sin(B*l)*cosh(B*l)-2*sinh(B*l)*cos(B*l)-2*cos(B*l)*cosh(B*l)-2*sin(B*l)*sinh(B*l)-4*sin(B*l)^3*exp(B*l)-4*cos(B*l)^3*exp(B*l)-2*exp(B*l)^4*sin(B*l)*cos(B*l)+2*exp(B*l)^4*sin(B*l)*sinh(B*l)+2*cos(B*l)*exp(B*l)^4*cosh(B*l)-4*exp(B*l)^2*sin(B*l)*cos(B*l)-4*exp(B*l)^2*sin(B*l)*cosh(B*l)-4*exp(B*l)^2*sinh(B*l)*cos(B*l)-4*exp(B*l)^2*sinh(B*l)*cosh(B*l)+8*cos(B*l)^2*cosh(B*l)*exp(B*l)^3+4*sin(B*l)^2*cos(B*l)*exp(B*l)^3-8*sin(B*l)^2*sinh(B*l)*exp(B*l)^3+8*sin(B*l)^2*cosh(B*l)*exp(B*l)^3+8*sinh(B*l)*cos(B*l)^3*exp(B*l)^2+4*sin(B*l)*cos(B*l)^3*exp(B*l)^2+4*sin(B*l)^3*cos(B*l)*exp(B*l)^2-4*cosh(B*l)^2*exp(B*l)^3*sin(B*l)+4*cosh(B*l)^2*cos(B*l)*exp(B*l)^3-2*exp(B*l)^4*sinh(B*l)*cosh(B*l)-2*exp(B*l)^4*sinh(B*l)*cos(B*l)-2*exp(B*l)^4*sin(B*l)*cosh(B*l)-4*sin(B*l)*cos(B*l)^2*exp(B*l)^3-4*sinh(B*l)^2*cos(B*l)*exp(B*l)^3-4*sinh(B*l)^2*exp(B*l)^3*sin(B*l)-sin(B*l)^2-2*sin(B*l)*cos(B*l)-sinh(B*l)^2-2*sinh(B*l)*cosh(B*l)-cos(B*l)^2-cosh(B*l)^2-8*sinh(B*l)*cosh(B*l)*exp(B*l)*sin(B*l)-8*cos(B*l)*cosh(B*l)*exp(B*l)*sin(B*l)+16*sin(B*l)*sinh(B*l)*B*l*exp(B*l)^2-4*sin(B*l)^3*exp(B*l)^3+exp(B*l)^4*sinh(B*l)^2+cosh(B*l)^2*exp(B*l)^4+exp(B*l)^4*sin(B*l)^2+cos(B*l)^2*exp(B*l)^4+4*sinh(B*l)^2*sin(B*l)*cos(B*l)*exp(B*l)^2-8*cos(B*l)*cosh(B*l)*exp(B*l)^3*sin(B*l)+8*sinh(B*l)*cosh(B*l)*exp(B*l)^3*sin(B*l)+8*sinh(B*l)^2*B*l*exp(B*l)^2+8*sinh(B*l)*cosh(B*l)*cos(B*l)^2*exp(B*l)^2-4*cosh(B*l)^2*sin(B*l)*cos(B*l)*exp(B*l)^2+8*sin(B*l)^2*sinh(B*l)*cos(B*l)*exp(B*l)^2+8*sin(B*l)^2*l*B*exp(B*l)^2+4*cos(B*l)^3*exp(B*l)^3)*B*(sin(B*l)+sinh(B*l))^2*exp(B*l)^2/wn*F0*(sinh(B*l)*cos(B*l)-sin(B*l)*cosh(B*l))*(-sin(w*t)*wn+sin(wn*t)*w)/(w^2*sin(B*l)+w^2*sinh(B*l)-wn^2*sin(B*l)-wn^2*sinh(B*l)) Yuck!


What about using a coil of wire to measure current?: What about using a coil of wire to measure current? I built two coil “sensors” 200-turn air/plastic core, 3.2 mm diameter, 7 mm long 400-turn magnetic steel core (from a bolt), 6.3 mm diameter, 9 mm long using 38-gauge copper wire (0.0040” diameter) On the heater cord: no signal from 200-turn coil with 400-turn coil: 10 mV pk-pk from 9 A current, 16 mV pk-pk from 13.4 A current (~0.4 mV/A) my piezo/magnet sensors are 100-1000x better! Coil voltage scales with number of turns planar coils get big fast A 400 turn coil of 5 mm wires with 5 mm spaces is at least 8 mm x 8 mm!


Max stress/strain: Max stress/strain Calculate von Mises stress as a function of input force max measurable current before it breaks? max stress at base of beam: shear is F/A, only tensile stress is My/I = [F(lm+0.5lb)y]/I


How might this device be fabricated?: How might this device be fabricated? Cantilever SOI Bulk micromachining Polysilicon LSN Piezoelectric AlN using LPCVD machine in microlab PZT using Beth Reilly’s process – pulsed laser deposition PZT Sol-gel – sg kim Direct printing of PZT – j lewis, steingart ho Magnet/Ferromagnetic material Electroplated - judy Screen-printed (lagorce) Pick-and place – NdFeB micromagnet paper


4-mask AlN process flow: 4-mask AlN process flow LPCVD low-stress nitride Sputter deposition, pattern, and liftoff of Pt electrode Sputter deposition of AlN film Deposition, patterning, and Cl2 RIE of Al top electrode Open via to bottom electrode using hot phosphoric acid AlN patterning using LTO “hard mask” and Cl2 RIE XeF2 release etch P. Stephanou, “Piezoelectric Aluminum Nitride MEMS Resonators for RF Signal Processing,” PhD Dissertation, University of California, Berkeley (2006) I’ll have to include an elastic support layer, as I’ll be fabricating a “31” mode cantilever. Can the electrode layers be made thicker to serve this function?


AlN recipe in detail: AlN recipe in detail (G. Piazza, 2004)


PZT cantilevers using PLD: PZT cantilevers using PLD Single crystal silicon wafer coated with 10 nm SrTiO3 (STO, from Motorola, Inc.) Deposit SrRuO3 (SRO) bottom electrode using pulsed laser deposition (PLD) Deposit PZT (PbZr0.47Ti0.53O3 )using PLD Deposit top electrode/elastic layer (Pt with Ti adhesion layer) using e-beam/thermal evaporation Define cantilever structures using photolithography Etch down to Si substrate using ion mill Release cantilever structures using isotropic XeF2 etch E. Reilly, E. Carleton, P. Wright, “Thin Film Piezoelectric Energy Scavenging Systems for Long Term Medical Monitoring,” Proc. IEEE Body Sensor Networks 2006 (2006)


Possibilities for magnet fabrication: Possibilities for magnet fabrication Screen/direct printing of magnetic powder/polymer composites L. Lagorce, M. Allen, “Magnetic and Mechanical Properties of Micromachined Strontium Ferrite/Polyimide Composites,” Journal of MEMS vol. 6 no. 4 (1997) pp. 307-312 ho steingart PowerMEMS 2006 Electroplating CoNiMnP, NiFe, Permalloy T. Liakopoulos, W. Zhang, C. Ahn, “Electroplated Thick Film CoNiMnP Permanent Magnet Arrays for Micromachined Magnetic Device Applications,” Proc. MEMS ’96 (1996) pp. 79-84 J. Judy, R. Muller, H. Zappe, “Magnetic Microactuation of Polysilicon Flexure Structures,” Journal of MEMS vol. 4 no. 4 (1995) pp. 162-169 Is NdFeB possible? Is it worth the trouble? B. Pawlowski, et al., “NdFeB thick films prepared by tape casting,” J. Magnetism and Mag. Mat’ls vol. 265 (2003) pp. 337-344 P. McGuiness, et al., “100-mm-thick Nd-Fe-B magnets for MEMS applications produced via a low-temperature sintering route,” J. Magnetism and Mag. Mat’ls vol. 305 (2006) pp. 177-181


Microlab training: Microlab training Qualified on Ion mill Edwards e-beam evaporation XeF2 etcher Trained on Tystar 17 low stress nitride oven Canon stepper Quintel Spinner SVGcoat


Heater cord anatomy: Heater cord anatomy 16AWGx2C: two-conductors, each 16 gauge 15 A, 300 V maximum rating per conductor 65 strands of 34AWG copper wire per conductor HPN: heater cord, neoprene insulation, parallel construction VW-1: UL UL flame resistant/retardant 105°C temperature rating 3.6 mm 7.6 mm 3.9 mm 1.3 mm


cut and paste: cut and paste