ET ZC341-L5

8/10/2019 ET ZC341-L5

http://slidepdf.com/reader/full/et-zc341-l5 1/87

BITS PilaniPilani Campus

Instrumentation and

control

ET ZC 341

1

Swapna Kulkarni

Lecturer,

8/10/2019 ET ZC341-L5



Digital Signal

Conditioning

8/10/2019 ET ZC341-L5


BITS Pilani, Pilani Campus

Data Acquisition Systems

• Microprocessor- based personal computers(PCs)are used extensively to implement direst digital

control in the process industries.

• These desktops are designed using a bus thatconsists of the data lines, address lines, and

control lines.

• All communication with the processor is viathese bus lines . This includes essential

equipment such as RAM, ROM , disk, and CD-

ROM.

8/10/2019 ET ZC341-L5



•

The PC also connects the bus lines to anumber of printed circuit board (PCB) sockets,

using an industry standard configuration of

how the bus lines are connected to the socket.

These sockets are referred to as expansion

slots.

• Many special types of peripheral equipment

such as fax/modem boards, game boards, and

network connection boards are designed on

PCBs that plug into these expansion slots.

8/10/2019 ET ZC341-L5



• Special PCBs called data acquisition systems(DACs) have been developed for the purpose

of providing for input and output of analog

data.• These are used when the PC is to be used in a

control system.

8/10/2019 ET ZC341-L5



DAS hardware

Typical layout of a data-acquisition board for use in a personal computerexpansion slot.

Copyright ©2006 by Pearson Education, Inc.

Upper Saddle River, New Jersey 07458

All rights reserved.

8/10/2019 ET ZC341-L5

http://slidepdf.com/reader/full/et-zc341-l5 7/87BITS Pilani, Pilani Campus

ADC and S/H: The DAS typically has a high-speed, successive approximation type ADC

and a fast S/H circuit.

• Whenever the DAS is requested to obtain adata sample, the S/H is automatically

incorporated into the process.

•The ADC conversion time constitute the majorpart of the data sample acquisition time, but

the S/H acquisition time must also be

considered to establish maximum throughput.

8/10/2019 ET ZC341-L5


Analog Multiplexer: The analog multiplexer(MUX) allows the DAS to select data from a

number of different sources.

• The MUX has a number of input channels,

each of which is connected to a different

analog input voltage source.

• The MUX acts like a multiple set of switches,

arranged in such a fashion that any one of the

input channels can be selected to provide its

voltage to the S/H.

8/10/2019 ET ZC341-L5


FIGURE 3.30 An analog multiplexer acts as a multiposition switch for selecting

particular inputs to the ADC.




8/10/2019 ET ZC341-L5


Address Decoder/command processor:

•The computer can select to input a sample from agiven channel by sending an appropriate selection on

the address lines and control lines of the computer

bus.

• These are decoded to initiate the proper sequence of

commands to the MUX, ADC, and S/H.

• Another common feature is the ability to program

the DAS to take a number of samples from a channelwith a specified time between samples.

• In this case, the computer is notified by interrupt

when a sample is ready for input.

8/10/2019 ET ZC341-L5


DAC and Latch: For output purposes, the DASoften includes a latch and DAC.

• The address decoder/command processor is

used to latch data written to the DAS, which isthen converted to an appropriate analog

signal by the DAC.

8/10/2019 ET ZC341-L5


DAS Software

• The process of selecting a channel andinitiating a data input from that channel

involves some interface between the

computer and the DAS.• This interface is facilitated by software that

the computer executes. The software can be

written by the user, but is often also providedby the DAS manufacturer in the form of

programs on disk.

8/10/2019 ET ZC341-L5


• Software for data acquisition

involves operations to start the

ADC, test the EOC, and input the

data.

• Fig. is a flowchart of the basis

sequence of operations that

must occur when a sample isrequired from the DAS.

• Generally, the DAS is mapped

into a base port address location

in the PC system.

• In the PC, this address can be

000H to FFFH, but many

addresses are reserved for use

by the processor and other

peripherals.

• A common address for

input/output(I/O) systems such

as the DAS is port 300H.Copyright ©2006 by Pearson Education, Inc.

Upper Saddle River, New Jersey 07458 All rights reserved.

8/10/2019 ET ZC341-L5


• The sequence starts with selection of a channel for

input.• This is accomplished by a write to the DAS decoder

that identifies the required channel.

• The MUX then places that channel input voltage at

the S/H input.

• The software then issues a start convert (SC)

command according to the specifications of the DAS.

• This is often accomplished by a write to some base +offset address.

• The DAS internally activates the hold mode of the

S/H and starts the converter.

8/10/2019 ET ZC341-L5


• The end –of-converter (EOC) is provided in a statusregister in the DAS. The contents of this status register

can be read by the processor by a port input of a base +

offset address.

• The appropriate bit is then tested by the software to

deduce whether the EOC has been issued.

• Once the EOC has been issued, the software can input

the data itself by a read of an appropriate address, againa base + offset, which enables tri states, placing the ADC

output on the data bus.

8/10/2019 ET ZC341-L5


• If the DAS fails, the computer will be locked in the loop

waiting for the EOC to be issued.• One way to resolve this is to add an additional timer loop

for a time greater than the conversion time of the ADC.

• If the EOC is not detected prior to time-out, an error is

announced, and the computer is returned to an errorhandling routine.

• In some cases, the EOC detection is handled by an interrupt

service routine. In this way the computer is free to execute

other software until the interrupt occurs.

• Then the data is input. Again, there needs to be a system to

detect that an EOC was not provided to protect against DAS

failure.

8/10/2019 ET ZC341-L5


Summary

1. The use of digital words enables the

encoding of analog information into a digital

format.

2. It is possible to encode fractional decimalnumbers as binary, and vice versa, using

3. Boolean algebraic techniques can be applied

to the development of process alarms and

elementary control functions.

1 2

10 1 22 2 ...... 2

m

m N b b b

8/10/2019 ET ZC341-L5


4. Digital electronic gates and comparators allow

the implementation of process Boolean

equations.5. DACs are used to convert digital words into

analog numbers using a fractional-number

representative. The resolution is

6. The data acquisition system(DAS) is a modular

device that interfaces many analog signals tocomputer. Signal address decoding,

multiplexing, and ADC operations are included

in the device.

2 n

RV V

8/10/2019 ET ZC341-L5


7. The sampling rate of a signal must be highenough to assure the signal can be

reconstructed from the samples.Generally,

we must sample about 10 times themaximum signal frequency.

8. One of the great advantages of digitizing data

and feeding it into a computer is thatnonlinearities can then be removed by

software. This is done by either an equation

or by a table look-up process.

8/10/2019 ET ZC341-L5



Chapter 4Thermal Sensors

20

8/10/2019 ET ZC341-L5


21

Instructional Objectives

• Define thermal energy, the relation oftemperature scales to thermal energy, andtemperature scale calibrations.

•

Transform a temperature reading among theKelvin, Rankine, Celsius, and Fahrenheittemperature scales.

• Design the application of an RTD temperature

sensor to specific problems in temperaturemeasurement.

8/10/2019 ET ZC341-L5


22

• Design the application of a thermistor tospecific temperature measurement problems.

• Design the application of a thermocouple to

specific temperature measurement problems.

• Explain the operation of a bimetal strip for

temperature measurement.

• Explain the operation of a gas thermometer anda vapor pressure thermometer.

8/10/2019 ET ZC341-L5



23

Introduction

• Process Control is a term used to describe,

natural or artificial, by which a physical

quantity is regulated.

8/10/2019 ET ZC341-L5



24

Thermal Energy

• SolidIn any solid material, the individual atoms ormolecules are strongly attracted and bonded to

each other, so that no atom is able to move farfrom its particular location, or equilibriumposition. Each atom, however, is capable ofvibration about its particular location. We

introduce the concept of thermal energy byconsidering the molecules’ vibration.

8/10/2019 ET ZC341-L5



25

Liquid

• If more and more energy is added to the material, thevibrations because more and more violent as thethermal energy increases. Finally, a condition is reachedwhere the bonding attractions that hold the molecules intheir equilibrium positions are overcome and the

molecules “break away” and move about in the material.When this occurs, we say the material has melted andbecome a liquid. Now, even though the molecules arestill attracted to one another, the thermal energy issufficient to cause the molecules to move about and nolonger to maintain the rigid structure of the solid. Weinstead of vibrating, one consider the molecules asrandomly sliding about each other, and the averagespeed with which they move is a measure of the thermalenergy imparted to the material.

8/10/2019 ET ZC341-L5



26

Gas

• Further increases in thermal energy of the

material intensify the velocity of the molecules

until finally, the molecules gain sufficient energy

to escape completely from the attraction ofother molecules. Such a condition is manifested

by boiling of the liquid. When the material

consists of such unattached molecules movingrandomly throughout a containing volume, we

say the material has become a gas.

8/10/2019 ET ZC341-L5



27

Metal Resistance Versus Temperature

Devices

• Time response becomes very important in thesecases because the measurement must wait until

the device comes into thermal equilibrium with

the environment. The two basic devices used arethe resistance-temperature detector (RTD) based

on the variation of metal resistance with

temperature, and the thermistor, based on thevariation of semiconductor resistance with

temperature.

8/10/2019 ET ZC341-L5



28

Figure Metal resistance increases almost linearly with

temperature, but the slope is small.

8/10/2019 ET ZC341-L5



29

Constant T A

l R

Where

R = Sample resistance ()

L = Length (m)

A = Cross – sectional area (m2)

= Resitivity (.m)

8/10/2019 ET ZC341-L5



30

Resistance-Temperature Detectors

• A resistance-temperature detector (RTD) is atemperature sensor that is based on theprinciples discussed in the preceding section;that is metal resistance increasing withtemperature.

• Metals used in these devices vary from platinum,which is very repeatable, quite sensitive, and

very expensive, to nickel, which is not quite asrepeatable, more sensitive and less expensive.

8/10/2019 ET ZC341-L5



31

Sensitivity

• An estimate or RTD sensitivity can be notedfrom typical values of the linear fractional

change in resistance with temperature. For

platinum, this number is typically on the orderof 0.004/ºC and for a nickel a typical value is

0.005/ºC. Thus, with the platinum, for example,

a change of only 0.4 would be expected for a

100- RTD if the temperature is changed by

1ºC.

8/10/2019 ET ZC341-L5



32

Response Time

• RTD has a response time of 0.5 to 5 s or more.

• The slowness of response is due principally to

the slowness of thermal conductivity in

bringing the device into thermal equilibrium

with its environment.

8/10/2019 ET ZC341-L5



33

Construction

• A length of wire whose resistance is to bemonitored as a function of temperature. The

construction is typically such that the wire is

wound on a form (in a coil) to achieve small size

and improve thermal conductivity to decrease

response time.

Signal Conditioning • In view of the small fractional changes of

resistance with temperature (0.4%), the RTD is

generally used in a bridge circuit.

8/10/2019 ET ZC341-L5



34

Figure : Note the compensation lines in this typical

RTD signal – conditioning circuit.

8/10/2019 ET ZC341-L5



35

Dissipation Constant

• The dissipation constant is usually specified

under two condition

• Free air and a well stirred oil bath. This is

because of the difference in capacity of the

medium to carry heat away from the device.

The self heating temperature rise can be

found from the power dissipated by the RTD,and the dissipation constant from

8/10/2019 ET ZC341-L5



36

D P

P T

Where

T = Temperature rise because of self-heating

in ºC

P = Power dissipated in the RTD from thecircuit in W.

PD= Dissipation constant of the RTD in W/ºC

8/10/2019 ET ZC341-L5



37

Range

• The effective range of RTDs principally

depends on the type of wire used as the active

element. Thus, a typical platinum RTD may

have a range of -100º to 650ºC. Whereas an

RTD constructed from nickel might typically

have a specified range of -180º to 300ºC.

8/10/2019 ET ZC341-L5



Thermistors

• The thermistor represents another class of

temperature sensor that measures

temperature through changes of material

resistance.

• The characteristics of these devices are very

different from those of RTDs and depend on

the peculiar behavior of semiconductorresistance versus temperature.

8/10/2019 ET ZC341-L5



Semiconductor Resistance versus

Temperature• In contrast to metals, electrons in semiconductor materials

are bound to each molecule with sufficient strength that noconduction electrons are contributed from the valence bandto the conduction band.

• A gap of energy , ∆Wg exists between valence and conduction

electrons.• Such a material behaves as an insulator because there are no

conduction electrons to carry current through the material.

• This is true only when no thermal energy is present in thesample-i.e., at a temperature of 0K.

• When the temperature of the material is increased, themolecules begin to vibrate.

• In the case of semiconductor, such vibration providesadditional energy to the valence electrons.

8/10/2019 ET ZC341-L5



• When such energy equals or exceeds the gapenergy,∆ Wg, some of these electrons becomefree of the molecules.

• Thus, the electron is now in the conduction band

and is free to carry current through the bulk ofthe material.

• As the temperature is further increased, moreand more electrons gain sufficient energy to

enter the conduction band.• It is then clear that the semiconductor becomes a

better conductor of current as its temperature isincreased-that is, as its resistance decreases.

Thermistor resistance versus temperature is highly

8/10/2019 ET ZC341-L5



Thermistor resistance versus temperature is highly

nonlinear and usually has a negative slope.




8/10/2019 ET ZC341-L5



•

Resistance of semiconductor material is decreasingfrom very large values at low temperature to smallerresistance at high temperature(which is opposite of ametal)

• An important distinction is that the change in the

semiconductor resistance is highly nonlinear.• The reason for semiconductors behave is the energy

gap between conduction and valence bands is smallenough to allow thermal excitation of electrons acrossthe gap.

• Note: the effect just described requires that thethermal energy provide sufficient energy to overcomethe band gap energy, ∆Wg .

8/10/2019 ET ZC341-L5



•

In general, a material is classified as a semiconductorwhen the gap energy is typically 0.01-4eV(1ev=1.6* 10-

19J)

• A semiconductor has a band gap of ∆Wg =1.107eV.

•

When heated, this material passes from insulator toconductor.

• The corresponding thermal energies that bring thisabout can be found and joules to eV conversion, thus:

For T=0K WTH=0.0eVFor T=100K WTH=0.013eV

For T=300K WTH=0.039eV

8/10/2019 ET ZC341-L5



•

With average thermal energies as high as0.039eV, sufficient numbers of electrons are

raised to the conduction level for the material

to become a conductor.

• In true insulators, the gap energy is so large

that temperatures less than destructive to the

material cannot provide sufficient energy to

overcome the gap energy.

h h

8/10/2019 ET ZC341-L5



Thermistor Characteristics

• A thermistor is a temperature sensor that has been

developed from the principles regarding semiconductor

resistance change with temperature.

• The particular semiconductor material used varies

widely to accommodate temperature ranges, sensitivity,resistance ranges, and other factors.

• The devices are usually mass-produced for a particular

configuration, and tables or graphs of resistance versus

temperature are provided for calibration.

• Variation of individual units from these nominal values

is indicated as a net percentage deviation or a

percentage deviation as a function of temperature.

8/10/2019 ET ZC341-L5



Sensitivity: The sensitivity of the thermistorsis a significant factor in their application.

• Changes in resistance of 10% per ⁰C are not

uncommon.• Thus, a thermistor with a nominal resistance

of 10kΩ at some temperature may change by

1kΩ for a 1 ⁰C change in temperature.

• When used in null-detecting bridge circuits,

sensitivity this large can provide for control, in

principle, to less than 1 ⁰C in temperature.

8/10/2019 ET ZC341-L5



Construction: Because the thermistor is a bulk

semiconductor, it can be fabricated in manyforms.

• Thus, common forms include discs, beads, and

rods, varying in size from a bead 1mm in

diameter to a disc several centimeters in

diameter and several centimeters thick.

• By variation of doping and use of different

semiconducting materials, a manufacturer can

provide a wide range of resistance values at

any particular temperature.

8/10/2019 ET ZC341-L5



Range: The temperature of thermistor depend s

upon on the materials used to construct thesensor.

• In general, there are three range limitationeffects: (1) melting or deterioration of the

semiconductor, (2) deterioration of encapsulationmaterial, and (3) insensitivity at highertemperatures.

• The semiconductor material may melt or

otherwise deteriorate as the temperature israised.

• This condition generally limits upper temperature

to less than 300 ⁰C.

8/10/2019 ET ZC341-L5



• At the low end, the principal limitation is that

the thermistor resistance becomes very high,into the MΩs, making practical applications

difficult.

•

For the thermistor shown in fig., if extended,the lower limit is about -80 ⁰C, where its

resistance has risen to over 3MΩ. Generally

the lower limit is -50 to -100 ⁰C.

• In most cases, the thermistor is encapsulated

in plastic, epoxy, Teflon, or some other inert

material.

8/10/2019 ET ZC341-L5



•

This protects the thermistor itself from theenvironment.

• This material may place an upper limit on the

temperature at which the sensor can be used.• At higher temperatures, the slope of the R-T

curve of the thermistor goes to zero.

•

The device is then is unable to measuretemperature effectively because very little

change in resistance occurs.

8/10/2019 ET ZC341-L5



Response time: The response time ofthermistor depends upon principally on the

quantity of material present and the

environment.• Thus, for the smallest bead thermistors in an

oil bath(good thermal contact), a response of

½ is typical.• The same thermistor in still air will respond

with a typical response time of 10s.

8/10/2019 ET ZC341-L5



• When encapsulated, as in Teflon or othermaterials, for protection against a hostile

environment, the time response is increased

by the poor thermal contact with theenvironment.

• Large disc or rod thermistors may have

response times of 10s or more, even withgood thermal contact.

8/10/2019 ET ZC341-L5



Signal Conditioning: Because a thermistorexhibits such a large change in resistance with

temperature, there are many possible circuit

applications.• In many cases, however, a bridge circuit is

used because the nonlinear features of the

thermistor make its use difficult as an actual

measurement device.

8/10/2019 ET ZC341-L5



• Because these devices are resistances, care must betaken to ensure that power dissipation in the

thermistor does not exceed the limits specified or

even interface with the environment for which the

temperature is being measured.

• Dissipation constant are quoted for thermistors as

the power in mill watts required to raise a

thermistor’s temperature 1⁰C above its environment. • Typical values vary from 1mW/⁰C in free air to

10mW/⁰C or more in an oil bath.

8/10/2019 ET ZC341-L5



•

Ex. A thermistor is to monitor roomtemperature. It has a resistance of 3.5 kΩ at

20⁰c with a slope of -10%/⁰c. the dissipation

constant is pd= 5mw/⁰c. it is proposed to use

the thermistor in the divider of fig. to provide

a voltage of 5.0 v at 20 ⁰c. evaluate the effects

of self heating.

8/10/2019 ET ZC341-L5



• It is easy to see that the design seems to work.

At 20⁰C, the thermistor resistance will be3.5kΩ, and the divider voltage will be

VD=3.5kΩ/(3.5kΩ+3.5kΩ)*10=5V

Let us now consider the effect of self heating.The power dissipation in the thermistor will be

given by P= V2/RTH=(5)2/3.5kΩ=7.1mW

The temperature rise of the thermistor can befound as

∆T=P/PD=7.1 mW/(5mW/⁰C)=1.42 ⁰C

8/10/2019 ET ZC341-L5



• But this means the thermistor resistance is reallygiven by

RTH= 3.5kΩ- 1.42 ⁰C(0.1/ ⁰C)(3.5kΩ)=3.0 kΩ

And so the divider voltage is actually VD=4.6 V. Theactual temperature of the environment is 20⁰C,but the measurement indicates that this is not so.Clearly, the system is unsatisfactory.

=> This example shows the importance of includingdissipation effects in resistive temperaturetransducers. The real answer to this probleminvolves a new design that reduces the thermistorcurrent to value giving perhaps 0.1 ⁰C of selfheating.

8/10/2019 ET ZC341-L5



Thermocouples

• We considered the change in materialresistance as a function of temperature.

• Such a resistance change is considered a

variable parameter property in the sense thatthe measurement of resistance, and therebytemperature, requires external power sources.

• There exists another dependence of electrical

behavior of materials on temperature thatforms the basis of a large percentage of alltemperature measurement.

8/10/2019 ET ZC341-L5


8/10/2019 ET ZC341-L5



Thermoelectric Effects

•

The basis theory of the thermocouple effect isfound from a consideration of the electrical andthermal properties of different metals.

• In particular, when a temperature differential is

maintained across a given metal, the vibration ofatoms and motion of electrons is affected so thata difference in potential exists across thematerial.

• This potential difference is related to the fact thatelectrons in the hotter end of the material havemore thermal energy than those in the coolerend, and thus tend to drift toward the cooler end.

8/10/2019 ET ZC341-L5



• This drift varies for different metals at thesame temperature because of differences in

their thermal conductivities.

•If a circuit is closed by connecting the endsthrough another conductor, a current is found

to flow in the closed loop.

•

The proper description of such an effect is tosay that an emf has been established in the

circuit and is causing the current to flow.

8/10/2019 ET ZC341-L5



• The proper description of such an effect is to say that

an emf has been established in the circuit and is

causing the current to flow.

• Fig. (a)shows a pictorial representation of this effect,

called the seebeck effect ,in which two different

metals, A and B, are used to close the loop with theconnecting junctions at temperatures T1 and T2.

• We could not close the loop with the same metal

because the potential differences across each leg

would be the same, and thus no net emf would bepresent.

• The emf produced is proportional to the difference in

temperature between the two junctions.

8/10/2019 ET ZC341-L5


8/10/2019 ET ZC341-L5



Seebeck Effect: Using solid-state theory, the

aforementioned situation may be analyzed to showthat its emf can be given by an integral overtemperature.

where =emf produced in volts.T1 ,T2 =junction temperatures in K

Q A,Q B = thermal transport constants of the two metals

• This equation, which describes the Seebeck effect,

shows that the emf produced is proportional to thedifference in temperature and, further, to thedifference in the metallic thermal transportconstants.

2

1

( )

T

A B

T

Q Q dT

8/10/2019 ET ZC341-L5



• Thus, if the metals are the same, the emf is zero,and if the temperatures are the same, the emf isalso zero.

• In practice, it is found that two constants, Q A andQ B are nearly independent of temperature and

that an appropriate linear relationship exists asԐ=α(T2-T1)

where α=constant in V/K

T1 , T2=junction temperatures in K

• However, the small but finite temperatureimpedance of Q A and Q B is necessary for anaccurate considerations.

8/10/2019 ET ZC341-L5



• Ex. Find the seebeck emf for a material withα=50μV/⁰C if the junction temperatures are

20⁰C and 100⁰C.

Solution:The emf can be found from

Ԑ=α(T2-T1)

=50(100-20)

=4mV

P l i Eff

8/10/2019 ET ZC341-L5



Peltier Effect

• An interesting and sometimes usefulextension of the same thermoelectric

properties occurs when the reverse of the

Seebeck effect is considered.• In this case, we construct a closed loop of two

different metals, A and B, as before.

•

Now , however, an external voltage is appliedto the system to cause a current to flow in the

circuit, as shown in fig. (b).

8/10/2019 ET ZC341-L5



•

Because of the different electrothermaltransport properties of the metals, it is found

that one of the junction will be heated and the

other cooled ; that is, the device is a

refrigerator.

• This process is referred as the Petlier Effect.

Some practical applications of such a device,

such as cooling small electronic parts, have

been employed.

Th l Ch t i ti

8/10/2019 ET ZC341-L5



Thermocouple Characteristics

• To use the Seebeck effect as the basis of atemperature sensor, we need to establish a

definite relationship between the measured

emf of the thermocouple and the unknowntemperature.

• We see first that one temperature must

already be known because the Seebeck

voltage is proportional to the difference

between junction temperatures.

8/10/2019 ET ZC341-L5



• Every connection of different

metals made in the thermocoupleloop for measuring devices,extension leads, and so on willcontribute an emf, depending onthe difference in metals andvarious junction temperatures.

• To provide an output that isdefinite with respect to thetemperature to be measured, anarrangement such as shown infig. (a) is used.

• This shows that the measured junction, TM is exposed to the

environment whose temperatureis to be measured.

• This junction is formed of metalsA and B as shown.

• Practical measurements with a

thermocouple system often employextension wires to move the reference to

a more secure location.




8/10/2019 ET ZC341-L5



• Two other junctions are then formed to a common

metal , C, which then connects to the measurement

apparatus.

• The “reference” junction are held at a common,

known temperature TR, the reference junctiontemperature. When an emf is measured, such

problem as voltage drops across resistive elements in

the loop must be considered.

• In this arrangement, an open circuit voltage is

measured(at high impedance) that is then a function

of only the temperature difference(TM-TR) and the

type of metals A and B.

8/10/2019 ET ZC341-L5



• The voltage produced has a magnitudedependant on which temperature is larger,

reference or measurement junction.

• Thus, it is not necessary that the measurement junction have a higher temperature than the

reference junctions, but both magnitude and sinof the measured voltage must be noted.

• To use the thermocouple to measure atemperature, the reference temperature must be

known, and the reference junctions must be heldat the same temperature.

• The temperature should be constant, or at leastnot vary much.

8/10/2019 ET ZC341-L5



• In most industrial environments, this would bedifficult to achieve if the measurement junctionand reference junction were close.

• It is possible to move the reference junctions to a

remote location without upsetting themeasurement process by the use of extensionwires , as shown in fig b.

• A junction is formed with the measurement

system, but to wires of the same type as thethermocouple. The extension wires now can berun a significant distance to the actual reference

junctions.

Thermocouple Types

8/10/2019 ET ZC341-L5



Thermocouple Types

•a First material is more positive when the measurementtemperature is more than the reference temperature

•b Constantan, chromel, and alumel are registered tradenames of alloys.

Type Materialsa Normal Range

J Iron-Constantan -190⁰ to 760⁰C

T Copper-constantan -200⁰C to 371⁰C

K Chromel-alumel -190⁰C to 1260⁰C

E Chromel-

constantan

-100⁰C to 1260⁰C

S 90%

platinum+10%

rhodium-platinum

0⁰C to 1482 ⁰C

R 87%

platinum+13%

rhodium-platinum

0⁰C to 1482⁰C

• Certain standard configurationsof thermocouples using specificmetals( or alloys of metals)have been adopted and givenletter designations; examplesare shown in Table 4.2

•

Each type has its particularfeatures, such as range,linearity, inertness to hostileenvironments, sensitivity, andso on, and is chosen for specificcases, such as oven

measurements, highly localizedmeasurements, and so on.




• These curves of thermocouple voltage versus

8/10/2019 ET ZC341-L5



temperature for a reference show the different

sensitivities and nonlinearities of three types.

•

The curves of voltage versustemperature in fig. areshown for a referencetemperature of 0⁰C and forseveral types ofthermocouples.

• The type J and Kthermocouples are noted fortheir rather large slope. Thatis high sensitivity-makingmeasurements easier for agiven change in temperature.

• The type S thermocouple hasmuch less slope and isappropriately less sensitive.




8/10/2019 ET ZC341-L5



• It has the significant advantages of a much

larger possible range of measurement,

including very high temperatures, and is highly

inert material.

• Another important feature is that these curves

are not exactly linear.

• To take advantage of the inherent accuracy

possible with these devices, comprehensive

table of the voltage versus temperature have

been determined for many types of

thermocouples.

Thermocouple polarity

8/10/2019 ET ZC341-L5



Thermocouple polarity

• The voltage produced by TC is differential in the sense that it is

measured between the two metal wires.

• As noted in the footnote to Table 4.2 , by convention the

description of a TC identifies how the polarity is interpreted.

• A type J thermocouple is called Iron-Constantan. This means that

if the reference temperature is less than the measurement junction temperature, the iron will be more positive than the

constantan.

• Thus, a type J with 0⁰C reference will produce +5.27mV for a

measurement junction of 100⁰C,meaning that the iron is morepositive than the constantan.

• For a measurement junction of -100⁰C, the polarity changes, and

the voltage will be -4.63 mV meaning that the iron is less

positive than the constantan.

Thermocouple Table

8/10/2019 ET ZC341-L5



Thermocouple Table

•

It simply give the voltage that results for a particulartype of thermocouple when the reference junctionsare at a particular reference temperature, and themeasurement junction at a temperature of interest.

• Referring to the tables, for example, we see that for a

type J thermocouple at 210⁰C with a 0⁰C reference, thevoltage is

V(210⁰C)=11.34mV (type J, 0⁰C ref)

• Conversely, if we measure a voltage of 4.768 mV with a

type S and a 0⁰C reference, we find from the table T(4.768 mV)=555⁰C (type S, 0⁰C ref)

• In most cases, the measured voltage does not fallexactly on a table value.

8/10/2019 ET ZC341-L5



• When this happens, it is necessary to interpolatebetween table values that bracket the desired value.

• In general, the value of temperature can be foundusing the following interpolation equation:

TM=TL+[(TH-TL)/(VH-VL)](VM-VL)

•

The measured voltage, VM , lies between a highervoltage, VH, and a lower voltage, VL, which are intables.

• The reverse situation occurs when the voltage for aparticular temperature, T

M

, which is not in the table, isdesired.

• Again, an interpolation equation can be used, such as

VM=VL+[(VH-VL)/(TH-TL)](TM-TL)

Ex. A voltage of 23.72 mV is measured with a type K

h l 0⁰C f Fi d h f h

8/10/2019 ET ZC341-L5



• Solution:

From the table, we find that VM=23.72 liesbetween VL=23.63mV and VH=23.84 mV withcorresponding temperatures of TL=570⁰C andTH=575⁰C, respectively.

The junction temperature is found from TM equation.

TM=570⁰C+(575⁰C-570⁰C)/(23.84mV-23.63mV)*(23.72mV-23.63mV)

=570⁰C+5⁰C/0.21*(0.009mV)=572.1⁰C

thermocouple at a 0⁰C reference. Find the temperature of the

measurement junction.

Ex. Find the voltage of a type J thermocouple

8/10/2019 ET ZC341-L5



g yp p

with a 0⁰C reference if the junction temperature

is -172⁰C.

• Solution: Apply the interpolation directly, the

junction temperature lies between a high TH=-

170⁰C and a low TL=-175⁰C.

The corresponding voltages are VH=-7.12mV

and VL=-7.27mV. The TC voltage will be

VM=-7.27mV+(-7.12+7.27)/(-170+175)*(-172⁰C+175⁰C)

=-7.27mV+(0.15mV/5⁰C)*(3⁰C)=-7.18mV.

Ch f T bl R f

8/10/2019 ET ZC341-L5



Change of Table Reference

• It has already been pointed out that thermocoupletables are prepared for a particular junctiontemperature.

• It is possible to use these tables with a TC that has a

different reference temperature by an appropriate shiftin the table scale.

• The key point to remember is that the voltage isproportional to the difference between the referenceand measurement junction temperature.

• Thus, if a new reference is greater than the tablereference, all voltages of the table will be less for thisTC.

Th l ill b j h l f h

8/10/2019 ET ZC341-L5



• The amount less will be just the voltage of the newreference as found on the table.

• Consider a type J thermocouple with a 30⁰C reference.The tables show that a type J thermocouple with a 0⁰Creference produces 1.54 mV at 30⁰C.

• This is the correction factor that will be applied to anyvoltage expected when the reference is 30⁰C. Consider

a temperature of 400⁰C. V(400⁰C)=21.85mV (Type J,0⁰C) andV(30⁰C)=1.54mV(Type J, 0⁰C ref)

• The correction factor is subtracted because thedifference between 400⁰ to 30⁰C is less than thedifference between 400⁰ and 0⁰C, and the voltagedepends upon this difference.

Therefore, V(400⁰C)=20.31mV (type J, 30⁰C ref)

8/10/2019 ET ZC341-L5



• To avoid confusion on the reference, all TCvoltages will henceforth be represented with a

subscript of the type and the reference.• Thus, VJ0 means a type J with a 0⁰C reference ,

and VJ30 means a type with a 30⁰C reference.

• To consider a couple more temperatures, verify

the following:VJ30 (150⁰C)=8.00-1.54=6.46mV

VJ30 (-90⁰C)=-4.21-1.54=-5.75mV

•

In the last case, the magnitude of the voltage islarger because the difference between -90⁰C and30⁰C is greater than the difference between -90⁰Cand 0⁰C, and the voltage depends on thisdifference.

8/10/2019 ET ZC341-L5



• In summary, tables of T voltage versustemperature are given for a specific referencetemperature.

• These tables can be used to relate voltage andtemperature for a different reference

temperature by using a voltage correction factor.• This correction factor is simply the voltage that

the new reference would produce from thetables.

• The correction factor is algebraically subtractedfrom table values, if the new reference is lessthan the table reference, and added if the newreference is less than the table reference.

• Fig Shows the correction processA change of reference from 0°C to 20°C

i i l t t lidi th TC

8/10/2019 ET ZC341-L5



• Fig. Shows the correction process

graphically.

• The curve for a 0⁰C reference is

assumed to be the table values.

• For a reference of 20⁰C, the curve

is simply reduced everywhere by

the correction voltage of 20⁰C, the

curve is simply reduced

everywhere by the correction

voltage of 1.02mV.

• In effect, the original curve slides

down by 1.02mV.

• Note: It is not simply add orsubtract the new reference

temperature as a correction factor.

Correction is always applied to

voltages.

is equivalent to sliding the TC curve

down in voltage.




N t Cl

8/10/2019 ET ZC341-L5


Next Class

• Types of temperature sensors and their

characteristic features

• Design Considerations

Documents

ET ZC341-L5