PDHengineer.

com
Course № IC-3004

Industrial Sensors and How They Work

To receive credit for this course

This document is the course text. You may review this material at
your leisure either before or after you purchase the course. To
purchase this course, click on the course overview page:

http://www.pdhengineer.com/pages/IC-3004.htm

or type the link into your browser. Next, click on the Take Quiz button
at the bottom of the course overview page. If you already have an
account, log in to purchase the course. If you do not have a
PDHengineer.com account, click the New User Sign Up link to create
your account.

After logging in and purchasing the course, you can take the online
quiz immediately or you can wait until another day if you have not yet
reviewed the course text. When you complete the online quiz, your
score will automatically be calculated. If you receive a passing score,
you may instantly download your certificate of completion. If you do
not pass on your first try, you can retake the quiz as many times as
needed by simply logging into your PDHengineer.com account and
clicking on the link Courses Purchased But Not Completed.

If you have any questions, please call us toll-free at 877 500-7145.

PDHengineer.com
5870 Highway 6 North, Suite 310
Houston, TX 77084
Toll Free: 877 500-7145
administrator@PDHengineer.com
 Industrial Sensors, And How They Work
Robert J. Scoff, PE

Copyright © December, 2007, Robert J. Scoff, PE

Copyright © Robert J. Scoff, PE December 2007 1

Table of Contents Page

1. Introduction 4
2. Thermocouples 4
3. Resistance Temperature Detector (RTD) 6
4. Thermistors 9
5. Bourdon Tubes 11
6. Radiation Thermometer 11
7. Bi-Metal Strips 11
8. Strain Gages 12
9. Oxygen Sensors 15
10. Ph Measurement 16
11. ORP Sensors 18
12. Flow Measuring 19
13. Conclusions 24

Copyright © Robert J. Scoff, PE December 2007 2

List of Illustrations
Figure Title Page

Figure 2.1 Commonly Used Symbol for a Thermocouple 4

Figure 2.2 Electrical Model of a Thermocouple 4
Figure 2.3 One Way to Temperature Compensate a Thermocouple 5
Figure 3.1 Typical RTD Measuring Circuit Showing Line Resistance 7
Figure 3.2 A Typical Circuit for Measuring Temperature With a Three Wire RTD 8
Figure 3.3 Typical 4 Wire RTD Measuring Technique 9
Figure 4.1 Electrical Model of a Thermistor 9
Figure 4.2 Circuit Showing PTC Used as a Resettable Fuse 10
Figure 5.1 How a Bourdon Tube Is Constructed 11
Figure 7.1 Construction of a Bi-Metal Strip 12
Figure 8.1 A Simple Round Resistor 12
Figure 8.2 Simple Load Cell 13
Figure 8.3 Wheatstone Bridge Load Cell Circuit 13
Figure 8.4 Load Cell with All Four Resistors Strain Gages 14
Figure 8.5 S Beam Load Cell 14
Figure 8.6 Resistance Values on a Fully Loaded Load Cell 15
Figure 8.7 Some Pictures of Actual Load Cells 15
Figure 9.1 Pictures of Actual Oxygen Sensors 16
Figure 10.1 Model of a Typical Ph Probe 16
Figure 10.2 Ph Probe Response as a Function of Temperature 17
Figure 10.3 Another way to look at Ph 17
Figure 11.1 Electrical Model of an ORP Probe 18
Figure 12.1 Simplified Drawing of a Magnetic Flow Meter 19
Figure 12.2 Electrical Model of a Magnetic Flow Meter Sensor 20
Figure 12.3 Pictures of Magnetic Flow Meters 20
Figure 12.4 Construction of a Electronic Mass Flow Meter 21
Figure 12.5 Construction Technique of a Paddle Wheel Flow Meter 21
Figure 12.6 Electrical Model of a Paddle Wheel Flow Meter 21
Figure 12.7 Electrical Model of an Encoder 22
Figure 12.8 Ultra Sonic Flow Meter Construction 22
Figure 12.9 Construction of a Differential Pressure Flow Meter 23
Figure 12.10 Picture of a Typical Weir 23

Copyright © Robert J. Scoff, PE December 2007 3

 1. Introduction

In today’s industrial and commercial world, there are many processes that must be measured and
controlled. The first part of this process is sensing. There are many sensors that that are used to do this.
This course covers a few of the more common ones and describes how they work. With that in mind,
let’s get started with temperature measuring sensors. Remember, this course is only concerned with how
these devices work.

2. Thermocouples

Thermocouples are one of the most common measuring devices used in the industrial world. A
commonly used symbol for a thermocouple is shown in Figure 2.1.

Measured Measuring
Junction Junction

Figure 2.1 Commonly Used Symbol for a Thermocouple

All thermocouples are made of two different metals connected together at a point which can be called The
Measured Junction. Almost any two dissimilar metals will work to make a thermocouple, but in the
industrial world there are only a dozen or so pairs of metals that are used. Now, when there is a
difference in temperature between the two ends of the thermocouple, a voltage is generated that
depends upon the magnitude of that difference in temperature. For instance, if the measured junction is
held at 100 degrees Centigrade, and the measuring junction is held at 0 degrees Centigrade, a small
voltage can be measured between the two ends of the measuring junction. A model of a thermocouple is
shown in Figure 2.2.

Figure 2.2 Electrical Model of a Thermocouple

For common thermocouple types, there are tables available that give that voltage versus temperature over
the useful range of that thermocouple. The tables are usually given with the measuring junction at 0
degrees Centigrade or 32 degrees Fahrenheit. One of the reasons that this reference temperature is used is
that the temperature of melting ice is one of the most stable and consistent constants of nature. It’s also
easy to replicate in a testing situation. The voltage that is generated is from 10 to 50 micro-volts per
degree Centigrade. The voltage is also very exact. The thermocouple tables are often given with 4 or 5
significant figures.

Have you noticed that the measuring junction, often called the Cold Junction, needs to be kept at the
temperature of melting ice to get accurate readings? For instance, if the measured junction was at 100

Copyright © Robert J. Scoff, PE December 2007 4

degrees Centigrade, and the cold junction was held at 10 degrees Centigrade, the thermocouple would
generate a voltage that would equate to 90 degrees Centigrade. A meter that should be reading 100
degrees and is reading 90 degrees is not very useful. This error becomes larger as the measuring junction
gets warmer. At the extreme, if both junctions were at 100 degrees, the actual temperature on the meter
would be 0 degrees. If you realize that 0 degrees Centigrade is freezing water, and 100 degrees
Centigrade is boiling water, you know that this is not a very accurate measuring method. Something
needs to be done to cancel this Cold Junction effect. What is done is something called Cold Junction
Compensation. What this means is that a temperature is added to the measured thermocouple
temperature that is equal to the temperature of the cold junction. There are several ways to do this, but
they all do the same thing. A voltage is added to the thermocouple generated voltage that is proportional
to the temperature of the measuring junction. Figure 2.3 shows one way that this is done. In this method,
a 4 resistor bridge circuit is added to the thermocouple. One of the resistors is temperature dependent.
That makes the voltage across the bridge variable with temperature. The resistors are chosen so that, for a
limited range, the voltage added in series with the thermocouple voltage is the same as the drop in voltage
because the measuring junction is not at zero degrees C.

Figure 2.3 One Way to Temperature Compensate a Thermocouple

An important thing to note is that the temperature dependent resistor is mounted physically close to the
measuring junction. This method has the advantage of not using any active components, just 4 resistors.
There are other techniques for doing temperature compensation, but this is one of the most common. In
all of them, the idea is the same. A voltage is added to the thermocouple voltage to make the voltage at
the measuring junction really proportional to the temperature at the measured junction.

There are no more than a dozen commonly used thermocouple types. Following is a list of 9 of the most
commonly used thermocouple types, including the temperature range over which they are useful. The
color associated with each thermocouple is the color of the insulation on the positive wire. The color of
the negative lead is always red.

Copyright © Robert J. Scoff, PE December 2007 5

 o o
Type Materials C Range F Range Color
• T Copper/Constantan -200 to 350oC - -328 to 662oF Blue
• J Iron/Constantan -40 to 750oC -40 to 1382oF White
• K Chromal/Alumel -200 to 1200oC -328 to 2282oF Yellow
• E Chromal/Constantan - 200 to 900oC -328 to 1652oF Purple
• N Nicrosil/Nisil -270 to 1300oC -450 to 2372oF Orange
• C Tungsten5%Rhenium/ 0 to 2320oC 32 to 4208oF White
Tungsten26%Rhenium
• B Platinum30%Rhodium/ 0 to 1700oC 32 to 3092oF Gray
Platinum6%Rhodium
• R Platinum13%Rhodium/ 0 to 1450oC 32 to 2642oF Black
Platinum
• S Platinum19%Rhodium/ 0 to 1450oC 32 to 2642oF Black
Platinum

Because of the physical characteristics of the materials used, there are certain applications where certain
types of thermocouples are used. For instance, Type T Thermocouples are commonly used where high
magnetic fields are present because both materials copper and constantan are non magnetic. Type J,
because one of the leads is iron, is not used where high magnetic fields are present. So, we might find a
Type T measuring the temperature of a motor or transformer, but never a Type J, which is iron
constantan. Type J is commonly found on plastic extrusion machines. Type K, which has a higher
operating temperature, is often used in high temperature metal annealing processes. Types B, R, and S
are often found measuring temperatures in ceramics heat treating processes.

All thermocouples are somewhat non-linear. This characteristic, along with the cold junction
compensation, make knowing the exact temperature somewhat of an art form. Older thermocouple meters
used non linear analog meter scales. Then an approximate cold junction temperature was chosen as the
temperature where the non linear scale tracked the actual temperature. With the advent of micro-
processors, look up tables were used to determine actual temperatures. Greater precision is now possible.
The new digital meters are able to give temperatures to degrees and fractions of a degree, but this
precision does not necessarily mean accuracy. When using thermocouples it is always important to
look at both the thermocouple characteristics and the operating temperature of the measuring junction.

3. Resistance Temperature Detector (RTD)

Another device for measuring temperature is a Resistance Temperature Detector. A RTD is simply a
resistor that changes resistance with temperature in a predictable way. Then if the RTD is heated, its
resistance can be measured, and its temperature known with a high degree of accuracy. Some materials
that are used to construct RTD’s are Platinum, Copper, Nickel, Balco (a 70% Nickel, 30% Iron alloy), and
Tungsten. The most common material from which RTD’s are made is Platinum.

Platinum RTD’s are made of 99.99% pure platinum (PT) and have a temperature coefficient (α) of
0.00385 Ohms/Ohm/oC or 0.385% per degree Centigrade. The most common resistances of Platinum
RTD’s are 100, 500, and 1000 Ohms (Ω’s) at 0o Centigrade. A 100 Ohm PT RTD will have a resistance
of 138.5 Ohms at 100 oC. The simplified linear equation of resistance versus temperature is:

R = Ro (1 + α {T – To} )

Copyright © Robert J. Scoff, PE December 2007 6

Where R is the resistance at the new temperature T, Ro is the resistance at temperature To (Usually 0o
Centigrade), and α is the temperature coefficient ( 0.00385 Ω / Ω / oC. The above equation is satisfactory
for most applications above 0 degrees centigrade. For temperatures below 0 degrees Centigrade, there is a
forth order polynomial that is used. This is because RTD’s are non linear below 0o C. Also, for more
accuracy, there is a second order polynomial that is sometimes used above 0o C.

To measure temperature with a RTD, the resistance must be measured. The only reasonable way to do
that is to force a current through the device and measure the voltage drop across it. Two problems occur.
The first is that the device being measured (the RTD), will heat up if a current is forced to flow through it.
From basic power theory, Power equals Volts Times Amps. So the RTD will get heated up by the
process that is used to measure its resistance and thus its temperature.

That means that the temperature that is read is always higher than the actual temperature. The fortunate
thing here is that the current can be made low enough so that this effect is not a significant source of error.

The second problem that occurs is that the wires leading from the RTD to the measuring instrument have
resistance. Not only that, the resistance of these wires changes with temperature. This can cause a small
error in measurement. To give some idea of the magnitude of this error, consider connecting a 100 ohm
RTD to an instrument that is 100 feet away. That means that there is 200 feet of wire from the sensor to
the meter. We have to count 100 feet going to the sensor and 100 feet returning to the meter. Number 18
wire, a typical size of wire used in this application, has a resistance of about 7 ohms per 1000 feet. That
means that there is an extra 1.4 ohms of resistance in series with the RTD. Figure 3.1shows how this line
resistance affects the RTD resistance.

Figure 3.1 Typical RTD Measuring Circuit Showing Line Resistance

Instead of measuring 100 ohms at 0 degrees Centigrade, the meter would measure 101.4 ohms. The
voltmeter, V, represents the measuring meter. The current source, Iref, is also a part of the measuring
meter. Obviously, the shorter the distance from the meter to the RTD, the lower the error. Also, the
bigger the wire, the lower the error. If # 10 wire, with a resistance of 1 ohm per 1000 feet, were used, the
line resistance for a distance of 100 feet between the sensor and the meter would only be 0.2 ohms. This
means that the total resistance would only be 100.2 ohms. Again, with bigger wire, the error would be
much smaller. Using the equation:

R = Ro (1 + α {T – To} )
For a 100 Ω PT RTD, there is an error of 3.6o C with # 18 wire and an error of 0.52o C with # 10 wire at
0o C. Using α of 0.00385 and Ro of 100 Ω, prove that this correct.

Copyright © Robert J. Scoff, PE December 2007 7

There are three techniques used when applying RTD’s. They are the two wire method, the three wire
method, and the four wire method. Figure .3.1 shows the two wire method of measurement. Figure 3.2
shows a way that a three wire RTD is used. Notice that one end of the RTD has two connections. The
voltage across the ba terminals can then be shown to be:

Circuits like this typical bridge use a precision voltage reference. If a 1.000 volt reference were used, the
power dissipation of the 100 ohm RTD at 0o C would be about 2500 microwatts. This could cause a self
heating error. Using a smaller reference voltage will cause a smaller self heating error. However, using
a smaller reference voltage will make more amplification necessary. This could lead to amplifier
instabilities. Not only that, but the above equation is non-linear.

Figure 3.2 A Typical Circuit for Measuring Temperature With a Three Wire RTD

Figure 3.3 shows a four wire RTD and a typical measuring technique. Each end of the RTD has 2
connections. One of the connections carries the Reference Current, which is a constant, and the other is
only used to measure the voltage at that connection. Iline1 and Iline4 carry the excitation current, and Rline2
and Rline3 carry almost zero current and are used only to send a voltage to the measuring instrument.

If the measuring instrument that measures Vab has a high input impedance, the current in Rline2 and Rline3 is
very low. This makes the error caused by that current very small. The Vab voltage is also linear with
change of RTD resistance and thus temperature. The only non-linearity that can enter in is any actual
RTD non linearity.

Copyright © Robert J. Scoff, PE December 2007 8

 Rline1

Rline2
a

Vab
RTD Iref

Rline3
b

Rline4

Figure 3.3 Typical 4 Wire RTD Measuring Technique

4. Thermistors

Thermistors are devices that change resistance, just as RTD’s change resistance. However, the similarity
ends there. Thermistors can have a positive or negative temperature coefficient. In other words, their
resistance can either increase or decrease with temperature.

An electrical model of a thermistor is very simple. It is simply a resistor that changes resistance with
temperature. The model should indicate whether the thermistor is a positive or negative temperature
coefficient device. Figure 4.1 shows an electrical model of a thermistor.

P or N
The P or N
indicates the
Thermistor type

Figure 4.1 Electrical Model of a Thermistor

The temperature versus resistance characteristic of most thermistors is very specific for each type of
thermistor. This characteristic can also be very precise and linear over limited temperature ranges. When
used to measure temperature care has to be taken to use them over limited temperature ranges or to use an
electronic circuit to linearize the temperature reading. One place where these devices are used is
thermostats for HVAC systems. They take the place of mercury and bimetal switches that were used in
older devices. With the mercury removed, they are environmentally friendly.

Copyright © Robert J. Scoff, PE December 2007 9

Positive temperature coefficient (PTC) Thermistors can be used as resettable fuses. Figure 4.2 shows a
circuit where a PTC thermistor is used to limit current flow to a circuit if the current increases and causes
the PTC thermistor to increase in resistance.
PTC

Load

Figure 4.2 Circuit Showing PTC Used as a Resettable Fuse

When PTC’s are used as resettable fuses they have a large temperature coefficient of resistance and a low
resistance at the operating temperature. They also have a point where the resistance increases rapidly
with temperature (Remember that they are non-linear). These thermistors are said to be of the ‘Switching
Type’. The high resistance when hot limits, but does not turn off, the current to the load. Devices like
this might be useful in places where it is difficult to replace fuses, such as satellites.

Negative Temperature Coefficient Thermistors (NTC) have certain applications in the temperature
measurement world. In spite of being non-linear, they are very accurate. Over small temperature ranges
they can be very linear. One application for thermistors can be as the temperature compensating resistor
in a thermocouple circuit. They are also used for very low temperature measurements.

Thermistors can also be used as temperature detecting devices in motors and transformers. A large
resistance change at a certain temperature can be detected to either shut down a device or send an alarm to
a Supervisory Control and Data Acquisition (SCADA) System. There are many different types of
thermistors whose characteristics can be used in different places. Some manufacturers make devices that
are linear over a certain range. If wider temperature ranges are used the Steinhart-Hart equation is
widely used. The Steinhart-Hart equation is:

1/T = a + b ln (R) + c ln3 (R)

Where a, b, and c are constants depending upon the individual device. T is the temperature in Kelvin, and
R is the resistance in Ohms. If a, b, and c are accurately known, there can be very small measurement
errors. For thermistors, the resistance is usually given at 25 degrees Centigrade (25o C = 298.15oK).
Since thermistors are used for measuring low temperatures, they are often specified in degrees Kelvin. In
case that it isn’t obvious, that equation is not easy to use.

Copyright © Robert J. Scoff, PE December 2007 10

 5. Bourdon Tubes

Bourdon tubes are gas or liquid filled tubes which are connected by a small tube to a diaphragm. When
the tube is heated, the gas or liquid inside expands causing the diaphragm to move. The movement of the
diaphragm can be used to move a dial pointer or cause a switch to change state or both. Figure 5.1 shows
how a bourdon tube device is constructed
Switch
Pointer

Anchor Place
for Spring

Diaphram

Small Hollow Tube

Figure 5.1 How a Bourdon Tube Is Constructed

The pointer is connected to the diaphragm so that it moves as the diaphragm expands or contracts. There
is a scale that will tell what the temperature of the tube is. The spring will hold the pointer and cause it to
return when the Bourdon Tube temperature drops. The switch is a snap action switch that will change
state at a certain position of the pointer. The switch can be used to turn heating or cooling off and on.
Bourdon tubes can operate satisfactorily for years. The bourdon tube or the small hollow tube can be
damaged or disappear in corrosive conditions. When a bourdon tube fails, it usually fails in a low
temperature position. If it were being used to bring a process up to temperature, the heater would never
turn off. This would cause the system to overheat. If the system did not have an over temperature
shutdown of some kind, damage or fires could result.

6. Radiation Thermometers

There are several types of non contact temperature measuring devices. Describing them in much detail is
beyond the scope of this course. What they do is measure temperature from the amount of thermal
electromagnetic radiation received from an object. The radiation level is converted to a voltage which is
correlated to a temperature. Industrial uses include applications where a moving object needs to have its
temperature controlled. An example might be a bread baking conveyor. There are also high temperature
applications where most other sensors fail due to the high temperature. There are also radiation type
flame detectors used industrial ovens and furnaces. This is a critical application, as a failure to detect a
flame out condition can lead to serious explosions.

7. Bi-Metal Strips

When two metals that have different coefficients of thermal expansion, strips of the two metals can be
connected together in such a way that when the assembly is heated, it will bend. Figure 7.1 shows how
this can be done.

Copyright © Robert J. Scoff, PE December 2007 11

 Metal A
Rivets

Metal B

Rivets

Figure 7.1 Construction of a Bi-Metal Strip

In operation, one end of the strip is firmly connected to a base, and the other end is allowed to move. The
end that moves can be connected to a pointer to indicate temperature. It can also cause a switch to change
state at a certain temperature. One industrial application of these devices is as overload heaters in motor
starters. Another application is controlling the fans of gas furnaces.

8. Strain Gages

Another common industrial sensor is a strain gage. A strain gage is simply a resistor that changes
resistance as it is stretched. To understand this, consider a simple resistor shown as a round cylinder, as
shown in Figure 8.1.

ρ *l
R=
A

ρ = Resistivity of Material
l

Figure 8.1 A Simple Round Resistor

As seen by the equation in Figure 8.1, the resistance of a resistor depends upon its length and cross
sectional area. In The English System of units, length is given in feet and cross sectional area is given in
Circular Mills. A Circular Mill (CM) is the area of a circle 0.001 inches in diameter. Rho (ρ) is a
constant that depends upon the material. Now, if we take that piece of material and stretch it, it becomes
longer and thinner. If we do not stretch it beyond its limits of elasticity, it will return to its original length
and cross sectional area when the stretching force is removed. However, when it is being stretched, its

Copyright © Robert J. Scoff, PE December 2007 12

resistance increases. We know that it will increase because the length, l, is longer and the cross section
area, A, is smaller. Both terms will cause the resistance to increase. This change of resistance can be
measured and converted into a force measurement.

Instead of being round wires, strain gages are usually thin strips of metal deposited on a substrate. Figure
8.2 shows how a strain gage can be glued to a metal substrate. When this is done, the device is called a
load cell.

Figure 8.2 Simple Load Cell

The measurement circuits for load cells are usually circuits called Wheatstone Bridges. Figure 8.3
shows a Wheatstone Bridge circuit with one leg of the bridge being a variable resistance. That variable
resistance would be the load cell.

Figure 8.3 Wheatstone Bridge Load Cell Circuit

The Wheatstone Bridge excitation voltage is a precision voltage source, as indicated by the 3 zeros
following the decimal point. This is because the Load Cell does not change in resistance very much.
That means the voltage from A to B does not change very much. That voltage, VAB, is used by an
electronic circuit to indicate the force on the load cell. To make the change of voltage bigger, and thus
easier to use, all four resistors in the Wheatstone Bridge circuit are made a part of the strain gage. The
circuit will then look like Figure 8.4.

Copyright © Robert J. Scoff, PE December 2007 13

 Ω Ω

Ω Ω

Figure 8.4 Load Cell with All Four Resistors Strain Gages

There are a number of physical constructions for Load Cells. Figure 8.5 shows one of the more common
ones. It is called an S Beam load cell because of its S shape.

Mounting
Stud

Four Strain
Gages
In Red

Mounting
Stud

Figure 8.5 S Beam Load Cell

Load cells can be designed to operate in tension or compression, or both. S Beam load cells can operate
under tension or compression. They are commonly used in weighing operations such as hoppers and
truck scales. Load Cells are typically rated for output voltages of 1 to 3 milli volts per volt of input for
full scale output. That means that if 10.000 volts is used for the excitation voltage, the full scale output
voltage will be 10 to 30 milli volts. Figure 8.6 shows an example of what could happen to the resistances
in a load cell under full load conditions.

Copyright © Robert J. Scoff, PE December 2007 14

 Ω Ω

Ω Ω

Figure 8.6 Resistance Values on a Fully Loaded Load Cell

Notice that two of the resistors increased by 1 ohm, and two of them decreased by 1 ohm. Under these
conditions, VAB equals 0.028 volts or 28 milli volts. With 10 volt excitation, that would make the
sensitivity of this load cell be 2.8 milli volts per Volt of excitation.

Load cells come in several resistance values. Some typical values are 120 ohm, 350 ohm, and 1000 ohm.
Our example was a 350 ohm load cell. Something to notice here is that each resistor in our bridge circuit
is 350 ohms, and the total resistance of the circuit is also 350 ohms.

Figure 8.7 shows several pictures of actual load cells. The strain gages themselves are very delicate, but
they are attached to the load cells in such a way to protect them from damage.

http://www.loadcells.in/single-point-loadcells-load-cells.htm

Figure 8.7 Some Pictures of Actual Load Cells

9. Oxygen Sensors

The measurement of oxygen content of air in industrial processes has historically been a difficult process.
Modern technology has improved oxygen measurement, but problems still exist. Many sensors are prone
to short expected lifetimes, calibration difficulties, and drift. With this in mind, let’s look at several of the
techniques used to measure oxygen content.

One of the techniques used in oxygen sensors is Electro Chemical. With this technique, O2 that is present
combines with water and extra electrons that are generated by a chemical process to form OH- ions. The
number of extra electrons that are formed and converted to OH- ions depends upon the O2 concentration
of the air flowing past the probe. That electron flow, or current, is measured and converted to percentage
oxygen in the air flow past the probe. This is a difficult technique to use in practice.

Copyright © Robert J. Scoff, PE December 2007 15

Another method to measure oxygen concentration is called a Zirconia Oxygen sensor. They must be
heated to 600o F to operate. These probes generate a voltage that depends on the oxygen content of the air
flowing past them. Many O2 sensors that are part of the combustion systems of modern car motors use
this type of sensor. They are not very linear and often just switch between a low and high reading as the
O2 content of the air flowing through them varies.

http://www.amio2.com/percent.html

Figure 9.1 Pictures of Actual Oxygen Sensors

Figure 9.1 shows some pictures of actual oxygen sensors. Note that they are designed to have air flow
through them. My feeling is that the velocity of the air flowing through them is important, as well as
temperature and humidity.

10. Ph Measurement

Ph is a common industrial measurement that tells an operator how basic or acidic a liquid is. It is a very
important measurement because many industrial processes depend on accurately knowing how acid or
basic the process solutions are. Ph varies from 0 to 14, with 0 being extremely acidic and 14 being
extremely basic. A Ph of 7 is neutral and is the Ph of distilled water.

Ph probes are essentially small batteries. Without going into the construction techniques of this battery,
we can say that it looks like a small voltage source in series with an internal resistance of about 400
million ohms. Figure 10.1 shows a model of a Ph Probe.

Figure 10.1 Model of a Typical Ph Probe

Copyright © Robert J. Scoff, PE December 2007 16

The voltage of this battery (the Ph Probe) depends upon the Ph of the liquid that it is immersed in. The
voltage of this battery is approximately 59 milli volts per Ph unit at 25o Centigrade. Along with the 400
Meg Ohm internal battery resistance, the battery voltage is temperature dependent. Fortunately, the
generated voltage versus Ph is linear. This makes it somewhat easier to measure Ph. Figure 10.2 shows
the temperature dependence of a typical Ph Probe. When the voltage is 0 Volts DC, the measured Ph is 7.
This is true for a wide range of temperatures. What this all means is that temperature compensation has to
be considered when measuring Ph. Many Ph probes come with a built in RTD to measure temperature, as
well as Ph. The electronic instrument then gives a temperature compensated reading of Ph of the liquid in
which it is immersed.

http://www.kittiwake.com/Default.aspx/Page/KB/KB/236

Figure 10.2 Ph Probe Response as a Function of Temperature

Because of the high internal resistance of the Ph Probe, the measuring instrument has to have a very high
input resistance. Newer electronic instruments are able to do this.

Figure 10.3 Another way to look at Ph

Copyright © Robert J. Scoff, PE December 2007 17

Figure 10.3 shows the Ph of a number of relatively common liquids. To measure Ph the probe has to be
immersed in a liquid.

Calibration of Ph probes is necessary on a regular schedule. Some sources say that Ph Probes should be
calibrated every time that they are used. Other sources say daily, weekly, or monthly, depending on the
application. The calibration process is relatively easy. Three buffer solutions are usually required with
known Ph’s of 4, 7, and 10. After allowing the probe and the three solutions to stabilize at a set
temperature, the probe is immersed in the neutral, Ph = 7, solution. There are usually two adjustments.
One is called zero or offset. While the probe is in the neutral solution, the zero adjustment is used to set
the meter to 7.0. Then the probe is washed and put in the 4 or 10 solution. The other adjustment, called
gain or slope, is used to set the meter to either 4 or 10, depending upon which solution was used. The
probe is then washed and checked at the other end, 4 or 10. The process is then repeated until the three
points, 4, 7, and 10, give accurate readings. As a side note, Ph probes should never be let to dry out. Ph
Probes are often stored with the tip immersed in buffer solution with a Ph of 4 or 7. Tap water is also
acceptable.

Since Ph Probes are really small batteries, they degrade with usage. Depending upon the application, they
may need to be replaced on a regular basis. If a facility has a Ph meter or controller, it is a good idea to
have a spare probe in case the one in operation fails. Whenever a Ph probe is replaced, it is necessary to
re-calibrate the measuring meter.

11. ORP Sensors

ORP stands for Oxidation Reduction Potential and is commonly called REDOX. It is a measure of the
oxidation and reduction potential of a liquid. It is commonly used to measure the percentage of chlorine
in water. This is useful in water and waste water plants. In water plants, the amount of residual chlorine
in the water shipped to customers has to be below certain levels. This is also true of the water discharged
from waste water plants. Chlorine is such a good oxidizer that a high concentration of it in water will kill
almost anything. It kills by causing the outside membranes of whatever it touches to dissolve. It actually
oxidizes the outer shell of micro organisms. In the case of water shipped to customers, the residual
chlorine has to be kept low enough to keep the customers alive and kill the harmful bacteria. In the case
of treated waste water discharged to streams, we want to make sure that the wildlife in the streams is not
adversely affected by the residual chlorine in the discharge.

ORP measurement is similar to Ph measurement. In fact, they look very much alike. Some companies
sell convertible Ph ORP probes. The electrical model of an ORP probe is similar to the electrical model
of a Ph Probe. This model is shown in Figure 11.1

Large Ohmic Resistor

Voltage from
ORP Probe Figure 11.1 Electrical Model of an ORP Probe

Copyright © Robert J. Scoff, PE December 2007 18

The voltage generated by an ORP Probe varies from about -800 milli volts, which is a reducing solution
to about +1200 milli volts, which is a strong oxidizing solution. A reading of 0 ORP is neutral. For
swimming pools and hot tubs, it is desired to keep the ORP reading above 700 milli volts. Adding
chlorine or other oxidizing substances when ORP readings fall below 700 is strongly recommended.
Most organisms will be killed by water with an ORP reading of above 700. High ORP readings will also
cause many metals to oxidize rapidly. In Germany, the ORP readings of public swimming pools is
required to be kept above 750. As a matter of interest, tap water has an ORP reading of between +400
to +500 milli volts. It has a slight oxidizing potential.

When waste water is treated, one of the last steps is to treat the water with chlorine to kill any unwanted
organisms. This raises the ORP reading to at least 700, which will quickly kill almost 100% of the
bacteria that remain in the treated effluent. When treated water is released into a stream, the ORP reading
has to be brought down so as to not kill the native wildlife and micro organisms in the stream into which
the effluent is released. The generally accepted upper limit for effluent is 400 milli volts. That means
that the chlorine treated effluent has to be treated again to lower the ORP to an acceptable level.

ORP readings are relatively slow to stabilize. While Ph readings will stabilize in seconds, ORP readings
can take minutes or even hours to stabilize. However, this is still a very useful measurement to tell us
how safe our water supply is.

Another way to look at ORP readings is a high ORP readings means that the liquid has a shortage of
electrons, giving it the ability to oxidize and sterilize. Low ORP readings (less than 0) has extra free
electrons that can be donated to free radicals. There is also a strong relationship between Ph and ORP.
Higher ORP readings generally indicate lower Ph readings. .

12. Flow Measuring

There are a number of flow measuring techniques, each with their own sensor. We will now look at
some of them.

Magnetic flow meters work by having the liquid flow through a magnetic field. The liquid must be a
conductor for this technique to work. The liquid (conductor) moving through a magnetic field causes a
voltage to be induced in the liquid. There are two probes in the pipe that measure that voltage. As the
liquid (conductor) moves faster, the induced voltage increases. The voltage level is converted to a flow
by an electronic circuit. Figure 12.1 shows how a Magnetic Flow Meter (Mag Meter) is constructed.
Liquid
Flow

Magnetic Figure 12.1 Simplified Drawing of a

Field Magnetic Flow Meter
4-20 ma
Generator 100 Output
Signal to
Measuring Recorder,
Probes Valve, or
Pump
Meter
Liquid Controller
Flow

Copyright © Robert J. Scoff, PE December 2007 19

The pipe through which the liquid conductor moves needs to be an insulator so that the pipe doesn’t affect
the voltage being generated by the liquid movement. Since most water is a conductor, this works for
almost any water flow in a pipe. When Magnetic Flow Meters are sold, they are usually sold as a pre-
calibrated unit. The meter is connected directly to the flanged section of pipe which includes the
magnetic generating coil and the probes. The meter is connected directly to terminals on the measuring
pipe. The sensor will then look like a voltage source in series with a resistor. The value of the resistor
will depend upon the liquid being measured. Figure 12.2 shows a model of a Magnetic Flow Meter
sensor.

Figure 12.2 Electrical Model of a Magnetic Flow Meter Sensor

Magnetic Flow Meters can be used for most conductive liquids. The variable series resistance shown in
the model will not usually affect the measurement because the input measurement circuit will be a high
input impedance amplifier. These meters do not work well with insulating liquids such as most oils.
Figure 12.3 shows several pictures of Magnetic Flow Meters.

Figure 12.3 Pictures of Magnetic Flow Meters

Another type of flow meter is an Electronic Mass Flow Meter. It is used primarily for measuring gas
flow. It works by heating a section of pipe that gas is flowing through. The temperature of the pipe is
measured both before and after the section that is heated. The difference in temperature is then used to
calculate the mass that is flowing through the pipe. That difference in temperature is actually proportional

Copyright © Robert J. Scoff, PE December 2007 20

to the mass of the gas passing that point. Figure 12.4 shows the construction of an Electronic Mass Flow
Meter. Some manufacturers also make mass flow meters for liquids.
Gas
Flow

Temperature
Sensor # 2

Heater

Temperature
Gas Sensor # 1
Flow
Figure 12.4 Construction of a Electronic Mass Flow Meter

Another type of flow meter is the Paddle Wheel Flow Meter. A paddle wheel is inserted into the flow and
turns in proportion to the flow rate. Figure 12.5 shows how a Paddle Wheel Flow Meter is made. They
are relatively inexpensive, but have some drawbacks. They are subject to clogging and are not as
accurate as other techniques.

Figure 12.5 Construction Technique of a Paddle Wheel Flow Meter

The voltage generator is similar to the tachometer that is used for motor RPM feedback. Figure 12.6
shows the electrical model of a Paddle Wheel Flow Meter.

Figure 12.6 Electrical Model of a Paddle Wheel Flow Meter

Copyright © Robert J. Scoff, PE December 2007 21

It is possible for the Paddle Wheel Flow Meter to have an encoder output instead of a DC Voltage. In this
case the output would be a variable frequency square wave. It would require an auxiliary power supply
and typically give a 5 volt square wave output. Figure 12.7 shows the electrical model of a Paddle Wheel
Flow Meter with an encoder output.

Figure 12.7 Electrical Model of an Encoder

An Ultra Sonic Flow Meter works by bouncing a signal off of suspended particles or air bubbles in a
fluid. It uses the Doppler Effect to measure the velocity of a fluid in a pipe, and thus its flow rate. This is
the same technique used in radar speed detectors. Note that the fluid being measured must be dirty or
have air bubbles in it. Figure 12.8 shows how a Ultra Sonic Flow Meter is constructed.

Figure 12.8 Ultra Sonic Flow Meter Construction

Differential Pressure (DP) Cells are another method of measuring flow. A Differential Pressure Cell has
two pressure sensors on either side of a orifice plate of some sort. This orifice plate can be simply a plate
with a hole in it smaller than the pipe diameter in which it is placed. Or it can be a venturi where the flow
is constricted. In either case, there is a pressure drop across the flow constriction. The pressures can be
measured by strain gages, or other pressure measuring methods. The differential pressure is then
converted to flow rate. Many DP cells have the flow rate proportional to the square of the pressure
difference. Then, to get flow rate the square root of the pressure difference is taken. Of course, the
pressure difference is a voltage level, since the output of strain gage circuits are voltages. Figure 12.9
shows how a Differential Pressure Flow Meter is constructed. In practice, there are valves to turn the
pressure lines off and on and to bypass the measuring sensors. These valves are used to test and calibrate
the flow meter.

Copyright © Robert J. Scoff, PE December 2007 22

 Differential 4-20 ma
Pressure signal to
Load Cell Control
Cabinet
Connection
to Input Connection
Pressure to Output
Pressure
Input Output
Flow Narrowing Flow
or
Orifice

Figure 12.9 Construction of a Differential Pressure Flow Meter

Although there are other types of flow meters, we will only cover one more general type. Open Channel
Flow Meters are used wherever liquid, usually water, flows through open waterways. They all work by
measuring the depth of liquid flowing in a specially constructed channel. The flow rate versus depth is
not usually a linear relationship. Electronic instruments are able to measure the depth of the liquid flow
and convert it into a flow rate. These specially constructed channels are called Flumes and Weirs. The
depth of the water in the flume can be measured by a number of techniques. Figure 12.10 is a picture of a
typical Weir.

Figure 12.10 Picture of a Typical Weir

Copyright © Robert J. Scoff, PE December 2007 23

An example of an unusual weir is a dam built across a stream where the water flows over the top of the
dam. This is a rectangular weir, and tables can be used to determine the flow rate of a stream if the depth
of the water flowing over the top of the dam can be measured.

13. Conclusions

This course does not cover the entire list of sensors that are used in the commercial and industrial world.
It was my intention to show that most, if not all, sensors can be modeled by a simple electrical circuit.
This understanding will help us whenever we need to apply a sensor to determine some physical quantity
or quality in the industrial world.

Copyright © Robert J. Scoff, PE December 2007 24