Thermocouple Lecture PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

Expt.

4 Temperature sensors and their calibration

Objectives:
• Calibration of thermocouples and thermistors
• Determine the dynamic response (time constants) and of the
given thermocouples

Temperature measuring devices

a) Thermocouples

• When a pair of electrical conductors are joined together, a thermal


emf is generated when the junctions are at different temperatures
• This phenomenon is called Seebeck effect
• Such a device is called a thermocouple
• The resultant emf developed by the thermocouple is generally in
the millivolt range when the temp. difference between junctions is~100oC
A ‘Classic’ Thermocouple
Metal B

At temperature 0.002V Voltmeter →E


T Metal A

Vary this temp. using


a constant temp bath

Ice bath

Measuring the emf of a thermo couple


To calibrate: the temperature of second junction can be varied using a constant
temperature bath and the emf recorded as function of temperature difference
between the nodes
The output voltage, E of such simple thermocouple circuit is usually written
In the form
E=AT+1/2 BT2+1/3CT3
The constants, A, B, and C are dependent on the thermocouple material
Thermocouple Properties

• Commercially available thermocouples usually consist of two leads


terminating in a single junction (no reference temperature, ice bath)

• The leads are connected to a thermocouple signal conditioning box


containing an electrical circuit which provides a reference voltage
equal to that produced by a reference junction placed at 0 oC

• This is called ice point compensation

• Most thermocouple signal conditioners usually display the temperature


directly and or provide a voltage that is proportional to the thermocouple
temperature.

A similar thermocouple signal conditioner with a digital temperature display


and an analog voltage output is used in the present experiment
b) Thermistors

•The thermistor, a thermally sensitive resistor, is made of semi-conducting


material

•Unlike metals, thermistors respond inversely to temperature. i.e. their


resistance decreases as the temperature increases.

•Thermistors are usually composed of oxides of manganese, nickel, cobalt


copper and several other nonmetals

The resistance is generally an exponential function of the temperature as

 R  1 1 

ln  
     
 Ro   T To 
Where Ro is the resistance at a reference temperature, To, while β is a constant,
characteristic of the material
Thermal runaway (generally not a good thing)
• Thermistor resistance varies inversely with temperature.
• The voltage applied directly across a thermistor causes its
temperature to rise, and its resistance to decrease.
• Sufficiently high voltage may cause thermal runaway, in which
case, higher currents and temperature are induced until the thermistor
fails or power is reduced.
Thermal runaway is generally prevented by placing a large resistor in series

 R  1 1 
T
ln       
A
 Ro   T To 

B
Temperature

E
R T

E
Time or power is reduced.
Static Calibration of Temperature Sensors
• Calibration of Thermocouple and Thermistor
• RTD (Resistance Thermal Detector) is used as reference
temperature
RTD display-T oC
Computer-Lab
DAC view-Measure
system Voltage- Vtc and Vtr

Rthermistor
Thermistor
RTD R1
Thermo
Couple
V
Heater

Measure Vtc and Vtr for different ΔV


steady reference temperatures. R3
Draw static calibration curves for R2
the thermo couple and the
thermistor
Why static calibration?
Dynamic Calibration of Temperature Sensors
• Unsteady response of the Thermocouples are measured
Computer-Lab
DAC view-Measure
system Voltage- Vtc(t)

Transfer quickly from


hot bath to cold bath.

Th The time response of


Tc thermo couples are
measured
Heater Ice bath
Different
thermocouples are
used ( size, material)
Time constant is a parameter used to define the
dynamic response of a system.
In this experiment, we measure the ‘time constant’
of various thermocouples
Transient/Dynamic response of a system
Input output
System

• Any system will take a finite amount of time to respond to a


change in input
• This delay in response is called the “rise time” or response
time
• Larger the rise time, slower its system response
Output
Input
(ideal)

time
time
‘Rise’ time
• In many systems (especially Ist order systems, governed by Ist
order ODE’s) the rise or delay of the system is exponential
E Time Constant
Ei

 The transient response function


E(t) 0.632ΔE
E (t )  E ( f )
ΔE is given by  e  at
E (i )  E ( f )
Ef
t

• Time constant is a parameter used to define the dynamic response of


a system.
• By definition, time constant is defined as the time needed for
E (t )  E ( f ) 1
  at  1
E (i )  E ( f ) e
1
t    , Time cons tan t
a
• Time constant is a determined by the physical properties of the
system. Note that 1/e =0.368
• Time constant = Time taken for a system to attain 63.2% of the
difference between initial and final steady state values
Time constant of a Thermocouple
(a model problem)
• Consider this problem, a thermocouple bead is initially exposed to a
temperature T(t=0)=23oC is instantaneously transferred to an ice bath
(T=T∞)
• Find the temperature variation T(t) and time constant of the Thermo
Couple ( assume TC as a lumped system ie Temperature is a
function of time only, Temperature variation in spatial directions are
neglected)

T(t)=?
Ti(t0=23)
Ice bath (T=T∞)

• T(t) can be found by using energy conservation

dT Bi<<1
Ein  Eout  mc Lumped system
dt Bi=hD/Ks
Deriving Time constant of a Thermocouple

dT
Ein  Eout  mc ,
dt
Ein  0, Eout  hA(T  T )
Eout
dE/dt
dT mc dT
ie  hA(T  T )  mc  dt 
dt hA (T  T ) T∞

mc d
define   (T  T )  dt 
hA 

IC t  0,    i
t  t ,   (T  T )
Solving we get
hA hA
T  T  ( )t a
 e mc  e  at mc
Ti  T
t
T  T 

e , where  is The time constant
Ti  T

1 mc
  
a hA
How do you find the time constant from your
dynamic calibration data

t
T(Volts) T  T 

Ti We have e
Ti  T
 T  T  t

ie ln 
 
 Ti  T  
Tf=T∞
 T  T  1
Plot ln  Vs t , then the slope 
 Ti  T  
time
  time cons tan t

You might also like