Lecture 2: Sensor characteristics

g g g g
Transducers, sensors and measurements Calibration, interfering and modifying inputs Static sensor characteristics Dynamic sensor characteristics

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

Transducers: sensors and actuators

g Transducer
n A device that converts a signal from one physical form to a corresponding signal having a different physical form
g Physical form: mechanical, thermal, magnetic, electric, optical, chemical

n Transducers are ENERGY CONVERTERS or MODIFIERS

g Sensor
n A device that receives and responds to a signal or stimulus
g This is a broader concept that includes the extension of our perception capabilities to acquire information about physical quantities

g Transducers: sensors and actuators

n Sensor: an input transducer (i.e., a microphone) n Actuator: an output transducer (i.e., a loudspeaker)
Input signal (measurand) Output signal (measurand)

Sensor
Input transducer

Processor

Actuator
Output transducer

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

Measurements
g A simple instrument model
Physical measurement variable Signal variable

Measurand

Sensor

Display

Measurement M

PHYSICAL PROCESS

n A observable variable X is obtained from the measurand

g X is related to the measurand in some KNOWN way (i.e., measuring mass)

n The sensor generates a signal variable that can be manipulated:

g Processed, transmitted or displayed

n In the example above the signal is passed to a display, where a measurement can be taken

g Measurement
n The process of comparing an unknown quantity with a standard of the same quantity (measuring length) or standards of two or more related quantities (measuring velocity)
Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University 3

Calibration
g The relationship between the physical measurement variable (X) and the signal variable (S)
n A sensor or instrument is calibrated by applying a number of KNOWN physical inputs and recording the response of the system

Signal output (Y)

Physical input (X)

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

Additional inputs
g Interfering inputs (Y)
n Those that the sensor to respond as the linear superposition with the measurand variable X
g Linear superposition assumption: S(aX+bY)=aS(X)+bS(Y)
Modifying input Z Physical variable X Signal variable

Measurand

Interfering input Y

Sensor

Signal output (Y)

Z=Z1 Z=Z2

g Modifying inputs (Z)

n Those that change the behavior of the sensor and, hence, the calibration curve
g Temperature is a typical modifying input

Physical input (X)

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University 5

Sensor characteristics [PAW91, Web99]

g Static characteristics
n The properties of the system after all transient effects have settled to their final or steady state
g g g g g g g g

Accuracy Discrimination Precision Errors Drift Sensitivity Linearity Hystheresis (backslash)

g Dynamic characteristics
n The properties of the system transient response to an input
g Zero order systems g First order systems g Second order systems

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

Accuracy, discrimination and precision

g Accuracy is the capacity of a measuring instrument to give RESULTS close to the TRUE VALUE of the measured quantity
n Accuracy is related to the bias of a set of measurements n (IN)Accuracy is measured by the absolute and relative errors
ABSOLUTE ERROR = RESULT - TRUE VALUE RELATIVE ERROR = ABSOLUTE ERROR TRUE VALUE

g More on errors in a later slide

g Discrimination is the minimal change of the input necessary to produce a detectable change at the output
n Discrimination is also known as RESOLUTION n When the increment is from zero, it is called THRESHOLD

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

Precision
g The capacity of a measuring instrument to give the same reading when repetitively measuring the same quantity under the same prescribed conditions
n Precision implies agreement between successive readings, NOT closeness to the true value
g Precision is related to the variance of a set of measurements

n Precision is a necessary but not sufficient condition for accuracy

g Two terms closely related to precision

n Repeatability
g The precision of a set of measurements taken over a short time interval

n Reproducibility
g The precision of a set of measurements BUT
n n n n
taken over a long time interval or Performed by different operators or with different instruments or in different laboratories

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

Example
g Shooting darts
n Discrimination
g The size of the hole produced by a dart

n Which shooter is more accurate? n Which shooter is more precise?

mean

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

Accuracy and errors

g Systematic errors
n Result from a variety of factors
g g g g g
Interfering or modifying variables (i.e., temperature) Drift (i.e., changes in chemical structure or mechanical stresses) The measurement process changes the measurand (i.e., loading errors) The transmission process changes the signal (i.e., attenuation) Human observers (i.e., parallax errors)

n Systematic errors can be corrected with COMPENSATION methods (i.e., feedback, filtering)

g Random errors
n Also called NOISE: a signal that carries no information n True random errors (white noise) follow a Gaussian distribution n Sources of randomness:
g Repeatability of the measurand itself (i.e., height of a rough surface) g Environmental noise (i.e., background noise picked by a microphone) g Transmission noise (i.e., 60Hz hum)

n Signal to noise ratio (SNR) should be >>1

g With knowledge of the signal characteristics it may be possible to interpret a signal with a low SNR (i.e., understanding speech in a loud environment)
Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University 10

Example: systematic and random errors

Systematic error (Bias) Random error (Precision)

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

11

More static characteristics

g Input range
n The maximum and minimum value of the physical variable that can be measured (i.e., -40F/100F in a thermometer) n Output range can be defined similarly

g Sensitivity
n The slope of the calibration curve y=f(x)
g An ideal sensor will have a large and constant sensitivity

n Sensitivity-related errors: saturation and dead-bands

g Linearity
n The closeness of the calibration curve to a specified straight line (i.e., theoretical behavior, least-squares fit)

g Monotonicity
n A monotonic curve is one in which the dependent variable always increases or decreases as the independent variable increases

g Hystheresis
n The difference between two output values that correspond to the same input depending on the trajectory followed by the sensor (i.e., magnetization in ferromagnetic materials)
g Backslash: hystheresis caused by looseness in a mechanical joint

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

12

Dynamic characteristics
g The sensor response to a variable input is different from that exhibited when the input signals are constant (the latter is described by the static characteristics) g The reason for dynamic characteristics is the presence of energy-storing elements
n Inertial: masses, inductances n Capacitances: electrical, thermal

g Dynamic characteristics are determined by analyzing the response of the sensor to a family of variable input waveforms:
n Impulse, step, ramp, sinusoidal, white noise
Amplitude Amplitude Amplitude Amplitude Amplitude time A 1

1/A

time

time

time

time

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

13

Dynamic models
g The dynamic response of the sensor is (typically) assumed to be linear
n Therefore, it can be modeled by a constant-coefficient linear differential equation dk y(t) d2 y(t) dy(t) +L a 2 + a1 + a0 y(t) = x(t) ak 2 k dt dt dt n In practice, these models are confined to zero, first and second order. Higher order models are rarely applied

g These dynamic models are typically analyzed with the Laplace transform, which converts the differential equation into a polynomial expression
n Think of the Laplace domain as an extension of the Fourier transform
g Fourier analysis is restricted to sinusoidal signals
n x(t) = sin(t) = e-jt

g Laplace analysis can also handle exponential behavior

n x(t) = e-tsin(t) = e-( +j)t

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

14

The Laplace Transform (review)

g The Laplace transform of a time signal y(t) is denoted by
n L[y(t)] = Y(s)
g The s variable is a complex number s=+j
n The real component defines the real exponential behavior n The imaginary component defines the frequency of oscillatory behavior

g The fundamental relationship is the one that concerns the transformation of differentiation
d L y(t) = sY(s) - f (0 ) dt

g Other useful relationships are

Impulse : L[ W] = 1 Step : Ramp : 1 s 1 L[r(t)] = 2 s L[u(t)] = Decay : Sine : Cosine : L[exp(at )] = (s - a ) L[sin( W )] = L[cos( s2 + s W )] = 2 s +
-1

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

15

The Laplace Transform (review)

g Applying the Laplace transform to the sensor model yields
dk y d2 y dy + ao y(t) = x(t) L ak k + L a 2 2 + a1 dt dt dt

(a s
k

+ L a 2s2 + a1s + a o Y(s) = X(s) Y(s) 1 = X(s) ak sk + L a 2s2 + a1s + ao

G(s) =

n G(s) is called the transfer function of the sensor

g The position of the poles of G(s) -zeros of the denominator- in the s-plane determines the dynamic behavior of the sensor such as
n Oscillating components n Exponential decays n Instability
Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University 16

Pole location and dynamic behavior

j -1 -2 j +j1 e 1t sin( -j1
1

t) t

2t

1t

-1

j +j1 +j2 +j2

sin(

t)
t

+j1

sin(

t)

e
-1

1t

j +j1 t -j1 +1 e + 1t sin( t) t

1

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

17

Zero-order sensors
g Input and output are related by an equation of the type

Y(s) y(t) = k x(t) =k X(s)

n Zero-order is the desirable response of a sensor
g No delays g Infinite bandwidth g The sensor only changes the amplitude of the input signal

n Zero-order systems do not include energy-storing elements n Example of a zero-order sensor

g A potentiometer used to measure linear and rotary displacements
n This model would not work for fast-varying displacements

VCC

X Y

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

18

First-order sensors
g Inputs and outputs related by a first-order differential equation

dy Y(s) 1 k a1 + a 0 y(t) = x(t) = = dt X(s) a1s + a0 s + 1

n First-order sensors have one element that stores energy and one that dissipates it n Step response
g y(t) = Ak(1-e-t/)
n A is the amplitude of the step n k (=1/a0) is the static gain, which determines the static response n (=a1/a0) is the time constant, which determines the dynamic response

n Ramp response
g y(t) = Akt - Aku(t) + Ake-t/

n Frequency response
g Better described by the amplitude and phase shift plots

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

19

First-order sensor response

g Step response g Frequency response
n Corner frequency c=1/ n Bandwidth

From [PAW91]

g Ramp response

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

20

Example of a first-order sensor

g A mercury thermometer immersed into a fluid
n What type of input was applied to the sensor? n Parameters
g g g g
C: thermal capacitance of the mercury R: thermal resistance of the glass to heat transfer F: temperature of the fluid (t): temperature of the thermometer

n The equivalent circuit is an RC network

g Derivation

n Heat flow through the glass ( F W)/R n Temperature of the thermometer rises as n Taking the Laplace transform

dt

W RC

(s) V (RCs + 1) V = F (s) RC F (s) V = W = F 1 e t/RC (RCs + 1) s

V =
F

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

21

Second-order sensors
g Inputs and outputs are related by a second-order differential equation

d2 y dy Y(s) 1 a 2 2 + a1 + a0 y(t) = x(t) = dt dt X(s) a 2s 2 + a1s + a0

n We can express this second-order transfer function as
Y(s) k = 2 X(s) s + 2 with k =
2 n n

s+

2 n

n Where

1 , a0

a1 , 2 a0a1

a0 a2

g k is the static gain g is known as the damping coefficient g n is known as the natural frequency

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

22

Second-order step response

g Response types
n Underdamped (<1) n Critically damped (=1) n Overdamped (>1)

g Response parameters
n n n n
Rise time (tr) Peak overshoot (Mp) Time to peak (tp) Settling time (ts)

From [PAW91]

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

23

Second-order response (cont)

g Ramp response g Frequency response

From [PAW91]

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

24

Example of second-order sensors

g A thermometer covered for protection
n Adding the heat capacity and thermal resistance of the protection yields a second-order system with two real poles (overdamped)

g Spring-mass-dampen accelerometer
n The armature suffers an acceleration
g We will assume that this acceleration is orthogonal to the direction of gravity

M
K B

x0

n x0 is the displacement of the mass M with respect to the armature n The equilibrium equation is:

&& x

& M(&&i &&0 ) = Kx 0 + Bx 0 x x Ms2 Xi (s) = X0 (s) K + Bs + Ms2 X0 (s) M K/M = s 2 Xi (s) K s 2 + s(B/M) + K/M

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

25

References
[PAW91] [Web99] [Tay97] [Fdn97] [BW96] R. Pallas-Areny and J. G. Webster, 1991, Sensors and Signal Conditioning, Wiley, New York J. G. Webster, 1999, The Measurement, Instrumentation and Sensors Handbook, CRC/IEEE Press , Boca Raton, FL. H. R. Taylor, 1997, Data Acquisition for Sensor Systems, Chapman and Hall, London, UK. J. Fraden, 1997, Handbook of Modern Sensors. Physics, Designs and Applications, AIP, Woodbury, NY J. Brignell and N. White, 1996, Intelligent Sensor Systems, 2nd Ed., IOP, Bristol, UK

Intelligent Sensor Systems Ricardo Gutierrez-Osuna Wright State University

26