Mechanical Sensors

The class of mechanical sensors used for the measurement of mechanical

variables is of special significance because of the extensive use of these
devices throughout the process – control industry.

The measurement of displacement, position or location is an important topic

in the process– industry. Examples of industrial measurements to measure
these variables are:
- Position and location of objects on conveyor )‫ (نقل الحركة‬systems.
- Liquid/solid level measurements.
- Conversion of pressure to a physical displacement that is measured to
indicate pressure ….. etc.

Potentiometric Sensors:
- The simplest type of displacement sensor involves the action of
displacement in moving the wiper of a potentiometer as shown in Figure.
- This device converts linear or angular motion into a changing resistance that
may be converted directly to voltage and / or current signals.
- Such potentiometric devices often suffer from the obvious problems of
mechanical wear, friction in the wiper action and high electronic noise.

Potentiometric displacement sensor.

Example:
A Potentiometric displacement sensor is to be used to measure an object
motion from 0 to 10m. The resistance changes linearly over this range from 0
to 1k. Develop signal conditioning to provide a linear 0 to 10volt output.
Solution:
57
The output voltage is:
R  R 
Vout = −  2  Vin =  D  (−10 V) = 0.01R D
 R1   1 k 
As RD varies from 0 to 1k, Vout changes linearly from 0 to 10volts.
The circuit of Figure will satisfy this problem.

Capacitance and Inductive Sensors:

A second class of sensors for displacement measurement involves changes in
capacitance or inductance.

1. Capacitive sensor:
The basic operation of a capacitive sensor can be seen from the familiar
equation of a parallel – plate capacitor:
A
C = K o (1)
d
where:
K = dielectric constant,
o = permittivity of vacuum,
A = plate area,
d = plates separation,
o = 8.85 pF/m.

There are three ways to change the capacitance as shown in Figure:

58
- Variation of the distance between the plates (d).
- Variation of the shared area of the plates (A).
- Variation of the dielectric constant (K).

The former two methods are shown in Figure.

Capacitance varies with the distance between the plates and the common
area. Both effects are used in sensors.

Example:
Figure shows a capacitive-displacement sensor designed to monitor small
changes in work-piece position. The two metal cylinders are separated by a
plastic sheath of thickness 1mm and dielectric constant at 1kHz of 2.5. If the
radius is 2.5cm, find the sensitivity in pF/m as the upper cylinder slides in
and out of the lower cylinder. What is the range of capacitance if h varies
from 1.0 to 2.0cm?

59
Solution:
The capacitance is given by Equation (1). The net area is the area of the shared
cylindrical area, which has a radius, r, and height, h.

Thus, A = 2πrh, so the capacitance can be expressed as:

rh
C = 2 K  o
d
The sensitivity with respect to the height, h, is defined by how C changes
with h; that is, it is given by the derivative:
dC r
= 2 K  o
dh d
Substituting for the given values, we get:
dC r 2.5 10 − 2 m
= 2 K  o = 2 (2.5) (8.5 pF / m) 
dh d 10 − 2 m
= 3475 pF / m
Since the function is linear with respect to h, we find the capacitance
range as:
Cmin = (3475 pF/m)(10-2m) = 34.75 pF
to

Cmax = (3475 pF/m)(210-2m) = 69.50 pF

2. Inductive Sensor:
- If a permeable core is inserted into an inductor as shown in Figure, the net
inductance is increased.
- Every new position of the core produces a different inductance. In this case,
the inductor and movable core assembly may be used as a displacement
sensor.

60
- An AC bridge or other active electronic circuit sensitive to inductance then
may be employed for signal conditioning.

This variable-reluctance displacement sensor changes the inductance in a coil in

response to core motion.

Variable - Reluctance Sensors:

In variable-reluctance displacement sensors, a moving core is used to vary
the magnetic flux coupling between two or more coils. Such devices find
application for the measure of angular displacements. Many configurations of
this device exist, but the most common used is called a Linear Variable
Differential Transformer (LVDT).

LVDT:
- Fig.6 shows that an LVDT consists of three coils
- The inner coil is the primary, which is excited by some AC source as
shown and the outer two coils are the secondary coils.
- Flux formed by the primary is linked to the two secondary coils, inducing an
AC voltage in each coil.
- When the core is centrally located, the net voltage is zero.
- If the core moves to one side or the other, an AC voltage will be induced
between the two secondary coils.
61
- If the two secondary coils are connected in series opposition, as shown in
Figure 6, then the two voltages will subtract; that is, the differential voltage
is formed.
- When the core is moved to one side, the net voltage amplitude will increase.
one side or the other.

- A variety of LVDTs are available with linear ranges at least from ±25 cm
down to ±1 mm.
- The static transfer function is typically given in (mV/mm) for a given
primary amplitude. Also specified are the range of linearity and the extent of
linearity.
62
Example:
An LVDT has a maximum core motion of ±1.5cm with a linearity of ±0.3%
over that range. The transfer function is 23.8mV/mm. If it is used to track
work-piece motion from -1.2 to +1.4cm, what is the expected output voltage?
What is the uncertainty in position determination due to nonlinearity?

Solution
Using the known transfer function, the output voltages can easily be found,
V(-1.2 cm) = (23.8 mV/mm)(-12 mm) = -285.6 mV

and
V(1.4 cm) = (23.8 mV/mm)(14 mm) = 333 mV

The linearity deviation shows up in deviations of the transfer function. Thus,

the transfer function has an uncertainty of
(±0.003)(23.8 mV/mm) = ±0.0714 mV/mm

Strain Sensors:
The measurement of strain in solid objects is common in process control.
Strain sensors are used to measure many other process variables, including
flow, pressure, weight, and acceleration.

Strain and Stress:

Strain is the result of the application of forces to solid objects. The forces are
defined in a special way described by the general term stress. In this section,
we will define stress and the resulting strain.

Definition
- A special case exists for the relation between forces applied to a solid
object and the resulting deformation of that object.
- The effect of applied force is referred to as a stress, and the resulting
deformation as a strain.

63
Tensile Stress-Strain In Figure (a), the nature of a tensile force is shown as a
force applied to a sample of material so as to elongate or pull apart the sample.
In this case, the stress is defined as:
F
Tensile Stress = (2)
A
where
F = applied force in N.
A = cross-sectional area of the sample in m2.

We see that the units of stress are N/m2 in SI units (or lb/in.2 in English
units), and they are like a pressure.

Tensile and compressional stress can be defined in terms of forces applied to a

uniform rod.

The strain in this case is defined as the fractional change in length of the
sample.

where

Tensile Strain = (3)

where
 = Change in length in m (in.)
 = Original length in m (in.)

64
Compressional Stress-Strain The only differences between
compressional and tensile stress are the direction of the applied force and the
polarity of the change in length. Thus, in a compressional stress, the force
presses in on the sample, as shown in Figure (b). The compressional stress is
defined as in Equation (2):
F
Compressio nal Stress =
A

The resulting strain is also defined as the fractional change in length as in

Equation (3), but the sample will now decrease in length:

Compressio nal Strain =

Stress-Strain Curve:
If a specific sample is exposed to a range of applied stress and the resulting
strain is measured, a graph similar to Figure results.

A typical stress-strain curve

In tensile and compressional stress, this constant is called the modulus of

elasticity, or Young’s modulus, as given by
Stress F/ A
E= =
Strain  / 

65
where
stress = F/A in N/m2 (or lb/in2)
strain =  /  (Uunitless)
E = modulus of elasticity in N/m2

Example:
Find the strain that results from a tensile force of 1000 N applied to a 10 m
aluminum beam having a 410-4 m2 cross-sectional area.

Solution
The modulus of elasticity of aluminum is found from Table 1 to be
E = 6.891010 N/m2. Now we have:

Stress F/ A
E= =
Strain  / 

So that:
Stress F/ A F
E= =   Strain =
Strain Strain EA
103
=
(6.89 1010 ) (4 10− 4 )
= 3.63 10−5 or 36.3 m / m

66
Strain Units Although strain is a unitless quantity, it is common practice to
express the strain as the ratio of two length units, for example, as m/m or
in./in; also, because the strain is usually a very small number, a micro (m)
prefix is often included. In this sense, a strain of 0.001 would be expressed as
1000 min./in., or 1000 mm/m.

3.2 Strain Gauge Principles

Earlier, we saw that the resistance of a metal sample is given by
0
R0 = 
A0
Where:
R0 = Sample resistance Ω
 = Sample resistivity Ω · m
 0 = Length in m
A0 = Cross-sectional area in m2

Suppose this sample is now stressed by the application of a force, F, as shown

in Figure (a). Then we know that the material elongates by some amount,
 , so that the new length is  = 0 +  , but its volume will remain nearly
constant. Because the volume unstressed is V = l0A0, it follows that if the
volume remains constant and the length increases, then the area must decrease
by some amount,  A :

V = A 0  0 = ( 0 + )(A 0 − A) (9)

Because both length and area have changed, we find that the resistance of the
sample will have also changed:
 0 + 
R = (10)
A 0 − A

67
Using Equations (9) and (10), the reader can verify that the new resistance is
approximately given by:

       
R = 0 1 + 2  = R 0 1 + 2  = R 0 + 2 R0 (11)
A0  0   0  0
from which we conclude that the change in resistance is

R = 2 R 0 (12)
0
Equation (12) is the basic equation that underlies the use of metal strain
gauges because it shows that the strain  /  converts directly into a
resistance change.

Example:
Find the approximate change in a metal wire of resistance 120 Ω that results
from a strain of 1000 m/m.

Solution
We can find the change in gauge resistance from

R = 2 R 0 = (2) (120) (103  10 −6 ) = 0.24 
0

Temperature Effects If not for temperature compensation effects, the

method of SG measurement would be useless. Change in resistance would
have resulted. Thus,
R (T) = R (To ) (1 +  0 T) , T1  T  T2
Or

 R = R 0  0 T

68
Metal Strain Gauges:
Metal SGs are devices that operate on the principles discussed earlier. The
following items are important to understanding SG applications.

Gauge Factor The relation between strain and resistance change [Equation
(12)] is only approximately true. Impurities in the metal, the type of metal,
and other factors lead to slight corrections. An SG specification always
indicates the correct relation through statement of a gauge factor (GF), which
is defined as
R / R
GF =
Strain
Where:
 R / R = Fractional change in gauge resistance because of
strain strain
 /  = Fractional change in length

For metal gauges, this number is always close to 2. For some special alloys
and carbon gauges, the GF may be as large as 10. A high gauge factor is
desirable because it indicates a larger change in resistance for a given strain
and is easier to measure.

A metal strain gauge is composed of thin metal deposited in a pattern on a

backing or carrier material.

71
Signal Conditioning
Two effects are critical in the signal-conditioning techniques used for SGs.
The first is the small, fractional changes in resistance that require carefully
designed resistance measurement circuits. A good SG system might
require a resolution of 2 mm/m strain. From Equation (12), this would
result in a ∆R of only4.8  10-4 Ω for a nominal gauge resistance of 120 Ω.

The second effect is the need to provide some compensation for temperature
effects to eliminate masking changes in strain.

The bridge circuit provides the answer to both effects. The sensitivity of the
bridge circuit for detecting small changes in resistance is well known.
Furthermore, by using a dummy gauge as shown in Figure 16a, we can
provide the required temperature compensation. In particular, the dummy is
mounted in an insensitive orientation (Figure 16b), but in the same proximity
as the active SG. Then, both gauges change in resistance from temperature
effects, but the bridge does not respond to a change in both strain gauges.
Only the active SG responds to strain effects. This is called a one-arm bridge.

Strain gauges are used in pairs to provide temperature compensation. In

some cases, such as this, only one gauge actually deforms during stress.

72
The sensitivity of this bridge to strain can be found by considering the
equation for bridge offset voltage. Suppose R1 = R2 = RD = R, which is the
nominal (unstrained) gauge resistance. Then the active strain gauge
resistance will be given by
 R 
R A = R 1 + 
 R 
and the bridge off-null voltage will be given by:
 RD RA 
V = Vs  − 
 R D + R1 R A + R 2 
If substitutions are made as defined previously, this voltage can be shown to
be
R
V R V R
V = − s − s
4 1 + R 4 R
R
where the approximation is good for (∆R/R) << 1. Substituting from
Equation (13) allows the expression for ∆V in terms of strain:
R
V R V 
V = − s  − s GF
4 1 + R 4 
R
Example:
A strain gauge with GF = 2.03 and R = 350 Ω is used in the bridge of Figure
16a. The bridge resistors are R1 = R2 = 350 Ω, and the dummy gauge has R = 350
Ω. If a tensile strain of 1450 mm/m is applied, find the bridge offset voltage if V s
= 10 V. Find the relation between bridge off-null voltage and strain. How
much voltage results from a strain of 1 micro?

73
Solution:
With no strain, the bridge is balanced. When the strain is applied, the gauge
resistance will change by a value given by
R / R
GF =
Strain
Then:

R = (GF) (Strain ) (R ) = (2) (1.4510 −3 ) (350) =1.03 

Since its tensile strain the resistance will increase to R = 351 Ω. The bridge offset
voltage is

RV R AV (351) (10)
V = − = 5− = −0.007 V
R1 + R R A + R 2 701

So that a 7mV offset results.

The sensitivity is found from Equation (14):

Vs  10  
V = − GF =− (2,03) = −10.15
4  4  
Thus, every micro of strain will supply only 10.15 mV.

74