7th Meeting

Physical phenomenon

Measurement Output

Measurement output:• interaction between a sensor and the environment surrounding the sensor• compound response of multiple inputs

Measurement errors:• System errors: imperfect design of the measurement setup and the approximation, can be corrected by calibration• Random errors: variations due to uncontrolled variables. Can be reduced by averaging.

Sensor technology; Sensor data collection topologies; Data communication; Power supply; Data synchronization; Environmental parameters and

influence; Remote data analysis.

Transducer is a device which transforms energy from one type to another, even if both energy types are in the same domain.• Typical energy domains are mechanical, electrical,

chemical, magnetic, optical and thermal. Transducer can be further divided into Sensors,

which monitors a system and Actuators, which impose an action on the system.• Sensors are devices which monitor a parameter of a

system, hopefully without disturbing that parameter.

Classification based on physical phenomena• Mechanical: strain gage, displacement (LVDT), velocity

(laser vibrometer), accelerometer, tilt meter, viscometer, pressure, etc.

• Thermal: thermal couple• Optical: camera, infrared sensor• Others …

Classification based on measuring mechanism• Resistance sensing, capacitance sensing, inductance

sensing, piezoelectricity, etc. Materials capable of converting of one form of

energy to another are at the heart of many sensors. • Invention of new materials, e.g., “smart” materials, would

permit the design of new types of sensors.

Zhang & Aktan, 2005

Definition: a device for sensing a physical variable of a physical system or an environment

Classification of Sensors Mechanical quantities: displacement, Strain, rotation velocity, acceleration, pressure, force/torque, twisting, weight, flow Thermal quantities: temperature, heat. Electromagnetic/optical quantities: voltage, current, frequency phase; visual/images, light; magnetism. Chemical quantities: moisture, pH value

Accuracy: error between the result of a measurement and the true value being measured.

Resolution: the smallest increment of measure that a device can make.

Sensitivity: the ratio between the change in the output signal to a small change in input physical signal. Slope of the input-output fit line.

Repeatability/Precision: the ability of the sensor to output the same value for the same input over a number of trials

True value

measurement

Precision without accuracy

Accuracy without precision

Precision and accuracy

Dynamic Range: the ratio of maximum recordable input amplitude to minimum input amplitude, i.e. D.R. = 20 log (Max. Input Ampl./Min. Input Ampl.) dB

Linearity: the deviation of the output from a best-fit straight line for a given range of the sensor

Transfer Function (Frequency Response): The relationship between physical input signal and electrical output signal, which may constitute a complete description of the sensor characteristics.

Bandwidth: the frequency range between the lower and upper cutoff frequencies, within which the sensor transfer function is constant gain or linear.

Noise: random fluctuation in the value of input that causes random fluctuation in the output value

Operating Principle: Embedded technologies that make sensors function, such as electro-optics, electromagnetic, piezoelectricity, active and passive ultraviolet.

Dimension of Variables: The number of dimensions of physical variables.

Size: The physical volume of sensors. Data Format: The measuring feature of data in time;

continuous or discrete/analog or digital. Intelligence: Capabilities of on-board data processing and

decision-making. Active versus Passive Sensors: Capability of generating

vs. just receiving signals. Physical Contact: The way sensors observe the

disturbance in environment. Environmental durability: will the sensor robust enough

for its operation conditions

Foil strain gauge• Least expensive• Widely used• Not suitable for long distance• Electromagnetic Interference• Sensitive to moisture & humidity

Vibration wire strain gauge• Determine strain from freq. of AC signal• Bulky

Fiber optic gauge• Immune to EM and electrostatic noise• Compact size• High cost• Fragile

Resistive Foil Strain Gage• Technology well developed; Low cost • High response speed & broad frequency

bandwidth• A wide assortment of foil strain gages

commercially available• Subject to electromagnetic (EM) noise,

interference, offset drift in signal. • Long-term performance of adhesives

used for bonding strain gages is questionable

Vibrating wire strain gages can NOT be used for dynamic application because of their low response speed.

Optical fiber strain sensor

Piezoelectric Strain Sensor• Piezoelectric ceramic-based or Piezoelectric polymer-based

(e.g., PVDF)• Very high resolution (able to measure nanostrain)• Excellent performance in ultrasonic frequency range, very high

frequency bandwidth; therefore very popular in ultrasonic applications, such as measuring signals due to surface wave propagation

• When used for measuring plane strain, can not distinguish the strain in X, Y direction

• Piezoelectric ceramic is a brittle material (can not measure large deformation)

Courtesy of PCB Piezotronics

Piezoelectric accelerometer• Nonzero lower cutoff frequency (0.1 – 1 Hz for

5%)• Light, compact size (miniature accelerometer

weighing 0.7 g is available)• Measurement range up to +/- 500 g• Less expensive than capacitive accelerometer• Sensitivity typically from 5 – 100 mv/g• Broad frequency bandwidth (typically 0.2 – 5 kHz)• Operating temperature: -70 – 150 C

Photo courtesy of PCB Piezotronics

Capacitive accelerometer• Good performance over low frequency range, can

measure gravity!• Heavier (~ 100 g) and bigger size than piezoelectric

accelerometer• Measurement range up to +/- 200 g• More expensive than piezoelectric accelerometer• Sensitivity typically from 10 – 1000 mV/g• Frequency bandwidth typically from 0 to 800 Hz• Operating temperature: -65 – 120 C

Photo courtesy of PCB Piezotronics

Metal foil strain-gage based (load cell)• Good in low frequency response• High load rating• Resolution lower than piezoelectricity-based• Rugged, typically big size, heavy weight

Courtesy of Davidson Measurement

Piezoelectricity based (force sensor)• lower cutoff frequency at 0.01 Hz

can NOT be used for static load measurement• Good in high frequency• High resolution• Limited operating temperature (can not be used for

high temperature applications)• Compact size, light

Courtesy of PCB Piezotronics

LVDT (Linear Variable Differential Transformer):• Inductance-based ctromechanical

sensor• “Infinite” resolution

limited by external electronics• Limited frequency bandwidth (250 Hz

typical for DC-LVDT, 500 Hz for AC-LVDT)

• No contact between the moving core and coil structure no friction, no wear, very long operating

lifetime• Accuracy limited mostly by linearity

0.1%-1% typical• Models with strokes from mm’s to 1 m

available

Photo courtesy of MSI

Linear Potentiometer • Resolution (infinite), depends on?• High frequency bandwidth (> 10 kHz)• Fast response speed• Velocity (up to 2.5 m/s)• Low cost• Finite operating life (2 million cycles) due to

contact wear• Accuracy: +/- 0.01 % - 3 % FSO• Operating temperature: -55 ~ 125 C

Photo courtesy of Duncan Electronics

Magnetostrictive Linear Displacement Transducer• Exceptional performance for long stroke position

measurement up to 3 m• Operation is based on accurately measuring the distance

from a predetermined point to a magnetic field produced by a movable permanent magnet.

• Repeatability up to 0.002% of the measurement range. • Resolution up to 0.002% of full scale range (FSR)• Relatively low frequency bandwidth (-3dB at 100 Hz)• Very expensive• Operating temperature: 0 – 70 C

Photo courtesy of Schaevitz

Differential Variable Reluctance Transducers • Relatively short stroke• High resolution• Non-contact between the measured object and sensor

Type of Construction Standard

tubular

Fixing Modeby 8mm

diameter

Total Measuring Range

2(+/-1)mm

Pneumatic Retraction No

Repeatability 0.1um

Operating Temperature Limits

-10 to +65 degrees C

Courtesy of Microstrain, Inc.

Scanning Laser Vibrometry • No physical contact with the test object; facilitate remote,

mass-loading-free vibration measurements on targets • measuring velocity (translational or angular)• automated scanning measurements with fast scanning

speed • However, very expensive (> $120K)

Photo courtesy of Bruel & Kjaer

Photo courtesy of Polytec

References• Structural health monitoring using scanning laser

vibrometry,” by L. Mallet, Smart Materials & Structures, vol. 13, 2004, pg. 261

• the technical note entitled “Principle of Vibrometry” from Polytec

Shock Pressure Sensor• Measurement range up to 69 MPa (10

ksi)• High response speed (rise time < 2

sec.)• High frequency bandwidth (resonant

frequency up to > 500 kHz) • Operating temperature: -70 to 130 C• Light (typically weighs ~ 10 g)

Shock Accelerometer• Measurement range up to +/- 70,000 g• Frequency bandwidth typically from 0.5

– 30 kHz at -3 dB• Operating temperature: -40 to 80 C• Light (weighs ~ 5 g)

Photo courtesy of PCB Piezotronics

Inertial Gyroscope (e.g., http://www.xbow.com)• used to measure angular rates and X, Y, and Z acceleration.

Tilt Sensor/Inclinometer (e.g., http://www.microstrain.com)• Tilt sensors and inclinometers generate an artificial horizon

and measure angular tilt with respect to this horizon.

Rotary Position Sensor (e.g., http://www.msiusa.com)• includes potentiometers and a variety of magnetic and

capacitive technologies. Sensors are designed for angular displacement less than one turn or for multi-turn displacement. Photo courtesy of MSI and Crossbow

What is MEMS?• Acronym for Microelectromechanical Systems• “MEMS is the name given to the practice of making and

combining miniaturized mechanical and electrical components.” – K. Gabriel, SciAm, Sept 1995.

Synonym to:• Micromachines (in Japan)• Microsystems technology (in Europe)

Leverage on existing IC-based fabrication techniques (but now extend to other non IC techniques)• Potential for low cost through batch fabrication• Thousands of MEMS devices (scale from ~ 0.2 m to 1

mm) could be made simultaneously on a single silicon wafer

Co-location of sensing, computing, actuating, control, communication & power on a small chip-size device

High spatial functionality and fast response speed• Very high precision in

manufacture• miniaturized components

improve response speed and reduce power consumption

Courtesy of A.P. Pisano, DARPA

Miniaturization• micromachines (sensors and actuators) can

handle microobjects and move freely in small spaces

Multiplicity• cooperative work from many small

micromachines may be best way to perform a large task

• inexpensive to make many machines in parallel Microelectronics

• integrate microelectronic control devices with sensors and actuators

Fujita, Proc. IEEE, Vol. 86, No 8

Capacitive MEMS accelerometer• High precision dual axis

accelerometer with signal conditioned voltage outputs, all on a single monolithic IC

• Sensitivity from 20 to 1000 mV/g

• High accuracy• High temperature stability • Low power (less than 700

uA typical) • 5 mm x 5 mm x 2 mm LCC

package • Low cost ($5 ~ $14/pc. in

Yr. 2004)Courtesy of Analog Devices, Inc.

Piezoresistive MEMS accelerometer• Operating Principle: a proof mass attached to a silicon

housing through a short flexural element. The implantation of a piezoresistive material on the upper surface of the flexural element. The strain experienced by a piezoresistive material causes a position change of its internal atoms, resulting in the change of its electrical resistance

• low-noise property at high frequencies

Courtesy of JP Lynch, U Mich.

MEMS dust here has the same scale as a single dandelion seed - something so small and light that it literally floats in the air.

Source: Distributed MEMS: New Challenges for Computation, by A.A. BERLIN and K.J. GABRIEL, IEEE Comp. Sci. Eng., 1997

ReferenceZhang, R. and Aktan, E., “Design consideration for sensing systems to ensure data quality”, Sensing issues in Civil Structural Health Monitoring, Eded by Ansari, F., Springer, 2005, P281-290