Đo nhiệt độ P10 pot

Pyrometer readings, which are measurements of the true target temperature, Tt, are strongly influenced by real, practical, industrial operating conditions as shown in a simplified way in

Practical Applications of

Pyrometers

When describing different types of pyrometers in Chapters 9 and 10 it was assumed that they operate in ideal conditions Pyrometer readings, which are measurements of the true target temperature, Tt, are strongly influenced by real, practical, industrial operating conditions as shown in a simplified way in Figure 11 1 These measurements, which must account for target emissivity, E (Tt) or E,L (T), in the environment of the surrounding walls

of temperature, Tom with emissivity, cw(TW), or, sxw(TW), are taken in the presence of polluted atmosphere of temperature, Ta, having the equivalent emissivity, Ea(Ta),or, ska(Ta),

and an absorption coefficient, aa(Ta)

Referring to Figure 11 1, the pyrometer readings depend on following factors:

" radiation emitted by the target whose temperature is to be measured - (1) influence of target emissivity,

" radiation emitted by the body and reflected from surrounding walls - (2) influence of surrounding walls,

" radiation emitted by the walls - (3) influence ofsurrounding walls,

" radiation absorption by the atmosphere,

" radiation absorption by the sighting window,

" proper emission by the polluted atmosphere,

" presence of solid bodies B in the sighting field of the pyrometer; influence of solid obstacles.

The pyrometer readings are also influenced by:

" temporarily covering ofthe target or its movement,

" dispersed radiation from outside of the viewing cone of the pyrometer.

Temperature Measurement Second Edition

L Michalski, K Eckersdorf, J Kucharski, J McGhee

Copyright © 2001 John Wiley & Sons Ltd ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

WALLS, SURROUNDING T ,E,, ,Ex,

SIGHTING WINDOW

t

Eat) ~B ATMOSPHERE E,(T,) A,(T,I,E,(T,),

kk*TEX(T,)

Figure 11.1 Pyrometric temperature measurement in real operating conditions

In practice the main components ofthe total measurement errors are:

" errors, ATE,depending on the emissivity of the target to be measured,

" errors, ATs, depending on the influence ofsurrounding walls,

" errors, AT,, due to absorption or proper emission by gaseous media,

" errors, ATi, of the indicating instruments, which may be positive or negative

The total error is given by :

The proper errors of the indicating instrument are usually small and do not depend upon the operating conditions

11.2 Influence of Target Emissivity.

A presentation of the methods of calculation, limitation and elimination of the influence of target emissivity which are used, will now be given

11 2.1 Calculation of true temperature

The relevant formulae for the particular types of pyrometers are given in Chapters 9 and 10 : Disappearing filamentpyrometers, as in equation (9.16):

T

1 / T + logEA, / 9613

- precise if the spectral emissivity at Ae = 0.65 ltm is well known

INFLUENCE OF TARGET EMISSIVITY 211

Total radiation pyrometers, as in equation (10 13):

Tt =Tj V1/8

- not very precise

Bandphotoelectric pyrometers, as in equation (10.26):

Tt = Ti n 1 / i,Zt_,12

-not very precise

Spectralphotoelectric pyrometers, based on equation (10.29) :

T

t =

fl (1/Ti)+(/,e/C2 )109Ek

- precise ifthe emissivity at X'e is well known

Two-colour and two-wavelength pyrometers :

- no correction needed for grey bodies

Multi-wavelength pyrometers:

- no correction needed

Calculated examples of the measurement errors of non-black body targets, for different types ofpyrometers, are given in Table 11 1

Table 11.1 Calculated error AT, = Ti - Tt at Tt = 1300 K;

a= 4 = LX1-a2 =0.8

Disappearing filament (9.16) -21

Band photoelectric (n = 6) (10.26) ca -34 Spectral photoelectric

-124 'le = 5 2 prn

11 2.2 Methods of approaching black-body conditions

1 Place a closed end ceramic tube of Ild ? 6, as described in Section 8.2, within the investigated medium or inside the furnace chamber, then measure the internal temperature using a pyrometer as shown in Figure 11 2

2 Apparent increase ofemissivity has been reported by Heimann and Mester (1975) using

an additional reflecting plate of low emissivity placed above the target surface as shown

in Figure 11 3 This method is also used successfully in temperature measurement of a steel sheet in a continuous process of rolling, as shown in Figure 11 4 A more detailed description of this method is given by Honda et al (1992)

3 Any type of pyrometer should be directed at the part of the target covered by black, matt varnish of sz 1

4 The pyrometric sensor, which is directed at the inside of a low emissivity cup, described

in Section 16.4, is periodically brought in contact with the surface, whose temperature is

to be measured Due to multiple inside reflections, the interior of the cup approaches a black body Thus, the measured temperature may also be a reference value for correcting the pyrometer readings The reflected radiation from outside is also eliminated during periods when the cup is near the target surface

5 A further development of the system described above is the Emissivity Enhancer also produced by Land Infrared Ltd (Land Infrared Ltd, undated) In this device the reflecting cup can be placed about 30 mm from the target surface and can be used with Land System 4 or UNO pyrometers

6 Application of a parallel polarising filter Aiming the pyrometer at an angle of about 45° to the measured surface results in an increase in the apparent emissivity An optimum angle has to be found experimentally Walther (1981) has pointed out the really inconvenient necessity of a rigid filter and pyrometer mounting The filter

CERAMIC TUBE

PYROMETER 9

Ild>6

Figure 11 2 Pyrometric measurement ofthe temperature ofa furnace chamber using a sighting tube

CYLINDER Figure 11.3 Low emissivity screen for Figure 11 4 Pyrometric temperature

pyrometric measurement of the temperature measurement of steel sheet in a rolling process

of non-black bodies

INFLUENCE OF TARGET EMISSIVITY 213 attenuation decreases the pyrometer output signal, limiting the application range of the method, which can only be applied to metallic surfaces

11.2.3 Other methods

1 In the majority of photoelectric pyrometers, as well as in some total radiation pyrometers

it is possible to set the emissivity value of the target by changing the measuring channel gain (Ircon Inc., 1997) The relevant element, equipped with emissivity scale, is placed before the lincarising circuit, as shown in Figure 11 5 If the true spectral emissivity, ei, differs from the set value,Es, some additional measurement errors, Ad , are observed These can be determined from Table 11.2 using the formula (Ircon Inc., 1993):

E E

At9=-100 s c At91io

(11 2)

Et

where A61% is the measuring error occurring at 1 % difference ofES - Et, as given in Table 11 2 In the case of emissivity change by x %, the corresponding total error is given by the formula:

(11 3)

As shown in Table 11 2, AYE errors increase with increasing values of effective wavelength, ~, If the emissivity of the body as a function of temperature is known a priori, such as for induction heated steel or graphite, then the dependence of the emissivity, s (T), can be stored in the memory of the measuring arrangement and taken account of as a correcting signal during the measurements.

2 If the condition of constant emissivity is not fulfilled, the pyrometer readings could be less influenced by the emissivity changes by convenient choice of operating infrared wavelength range Using Planck's law, the spectral heat flux densityqk as a function of wavelength, and of target temperature, T, is:

qz =E,~

eC2/AT-1

HEAD

°C M

Figure 11.5 Emissivity setting in a photoelectric pyrometer by gain change

Table 11.2 Emissivity errors, A,91%, in °C of spectral and band pyrometers due to 1 % difference between the set £s and true £t emissivity values (Ircon Inc.,1997)

Target Effective wavelength, l,e (gym)

temperature

where cl and c2 are Planck constants Also, using theStefan-Boltzmann law:

where 6o is the Stefan-Boltzmann constant

Calculating the partial derivatives of the above formulae relative to emissivity and temperature, yields respectively:

The heat flux density for wavelength range ~ 1 to /12 and x>>4 is given by:

q;,1 -~z = f~ q;d,,=£6 O TIA2 x (11 6)

INFLUENCE OF TARGET EMISSIVITY 215

a(6j4)=6j4 ; d- (6, T4)=4e6.T3 ; (11 7)

Finally the ratios of rate of signal change relative to temperature and to emissivity changes are respectively:

U Thus, using pyrometers, which operate in a narrower wavelength range, the readings are less influenced by emissivity change, even for bodies of low emissivity (e ; 0.1 - 0.3) The ratio ofthese readings is aboutx/4;sometimes even as high asx/4 ;zz 4 - 6

3 For a certain assumed target emissivity, the data of Table 11.2 also indicate that the influence of changing emissivity on spectral and band pyrometer readings can be reduced by using pyrometers operating at theshortest possible effective wavelength, as shown before in Section 10.4.3

4 In repeated production processes, measuring the true temperature, Tt, is avoided Instead,another, more precise methodis used to measure the true temperature, Tt, at the same time determining the relevant indicated temperature, Ti This apparent temperature,

Ti, may serve as a reference for the repeatability of the process, as long as the target emissivity is constant.

5 Kelsall (1963)describes a special method ofautomatic compensation ofthe influence of

irradiated with two heat fluxes One heat flux is the combined effect of the true heat flux due to the temperature of the target with emissivity, el, and another component due to the reflection of the heat flux from the target:, with reflectivity, pl, initially coming from

a heating element of emissivity, c2 = 1 This combined heat flux,(DI,is:

The direct heat flux from the heating element is:

DETECTOR LENS

12.1, T2

HEATING ELEMENT

31 E1 T1 11TARGET

Figure 11.6 Automatic compensation ofthe influence of emissivity

In the formulae (11 11) and (11 12), Ej and £2 , are the respective emissivities oftarget and heating element, pj is the reflectivity of the target, kj is a coefficient, depending on the diameter of the aperture in the rotating disk associated with the flux 4)j , k2 is a similar constant for 02 Changing and measuring the temperature, T2, of the heating element until the condition (D j = 02 is reached at kj = k2 when TI =T2 and thus regarding that E2 = 1, the pyrometer readings are correct

11.3 Influence of Surrounding Walls

The surrounding conditions can exert a marked influence on pyrometer readings For instance, when measuring the temperature of a charge placed inside a furnace chamber whose wall temperature is different from that of the charge, 6, the pyrometer is aimed at the charge through a sighting window as shown in Figure 11 7 Assuming a non-transparent charge, the pyrometer readings depend upon the signal, Ecfj (t9c ) , emitted by the charge surface of temperature, 0, and emissivity, Ec , and also upon the signal,

EW (1- E,)f2(t9,,) , emitted by the walls of temperature 6v, , and emissivity, EW , reflected from the charge

The overall signal, s, determining the pyrometer readings, is then:

Any non-linear dependence of the pyrometer readings upon the measured temperature, which is represented by the functions, fj (Oc ) and f2 (6w) , depends on the pyrometer

INFLUENCE OF SURROUNDING WALLS 217

"9v E V

WINDOW

E, -TEMPERATURE AND EMISSIVITY OF CHARGE 4.J -TEMPERATURE AND EMISSIVITY OF WALLS

used Similarly, the emissivity of walls, eW , and ofcharge, Ec,which can be either total or band or spectral emissivity, also depends upon the pyrometer type If the charge emissivity

is high, Ec-~1, the error due to radiation reflected from the walls can be neglected, because (1- ec ) ) 0 Hence, the signal, s, given by equation (11 13), only depends upon the charge radiation.

When e,, < r9c measurement errors, due to radiation reflected from the walls, are usually negligibly small, especially when the charge emissivity is correctly set on the emissivity corrector scale of the pyrometer It is then advisable to use a pyrometer with a short effective wavelength Ae, because the reflected radiation has a mainly long wavelength.

When OW = 6, no errors are observed.

When 6w > Oc , the pyrometer readings are too high, giving errors which increase with increasing wall temperature, Ow, and with decreasing charge emissivity, ec As the correct setting of the emissivity corrector does not prevent the errors, it is advisable to use pyrometers with a long effective wavelength, Ae

The following methods can be used to reduce the influence ofradiation from walls :

1 Application of a pyrometer of such an Effective wavelength at which the charge emissivity is as high as possible.

2 A water, air or nitrogen cooled sighting tube protecting the sighted area from wall radiation, as shown in Figure 11 8 Water cooled tubes can be even 1 5 m long (Land Infrared, 1997a).

3 Directing ofpyrometer sensing tube such that reflected environmental radiation does not influence the readings, as shown in Figure 11 9 (Ircon Inc., 1997) Only in the case when the charge emits dispersed reflected radiation may a small amount arrive at the pyrometer.

HEATING ELEMENTS FALSE CORRECT

POSITIONING POSITIONING

PYROM ETER

AGENT

- CHARGE RADIATION SIGHTING TUBE CHARGE - WALLS RADIATION

A numerical example illustrates the calculation ofthe true charge temperature

Numerical example

charge temperature 9c

Solution:

Using the pyrometer characteristic shown in Figure 11 10, the pyrometer output signal

1 75 = 0 5xfl(9a) + (1-0 5)x2 96

or

fi(dc) = (1/0 5)x(1 75 - 0 5x2 96) = 0 54mV

9c = 565 °C

6

E

_5

W

4

z 2,96 3' r- 2-1,75 a

1 ,0,54

0 3~=565°C 4 7114 ,A=B40°C 0

500 600 700 B00 900 1000

TEMPERATURE ,A , "C

Figure 11 10 Determining true charge temperature, 9c, based upon the characteristic of a total

INFLUENCE OF SURROUNDING WALLS 219

4 A twopyrometer method (Beynon, 1981 ; Roney, 1992) based on the use of the difference signal of two pyrometers The first one, aimed at the charge, has an output signal :

The first term depends on the temperature, 0c , of the charge with emissivity, E, and the second one on radiation from the walls, at a temperature, 6, and emissivity, Ew, reflected from the charge The functionf, describes the scale defining equation of the pyrometer

The second pyrometer, of identical functionf, is aimed at that part of the wall, whose temperature, Ow , equals the average temperature of all the walls The signal of this second pyrometer is :

Correcting both signals in the function of charge emissivity, the resulting difference signal is free from any influence of radiation from the walls Formation and conditioning

of the difference signal is made by microprocessor A simplified diagram of the system, which is shown in Figure 11 11 (Ircon Inc., 1996a), uses a MaxIine microprocessor To ensure correct operation of this system, it is possible to choose one out of eight spectral operating bands for the pyrometer, at which the spectral emissivity of the charge is the highest The system operates correctly when the spectral emissivity of the charge is known and when no absorption by the furnace atmosphere takes place

5 In thetwopyrometer method with an additional, cooled reference radiation source, the first pyrometer is aimed at the charge The second one is aimed at the water cooled reference radiation source, placed in the furnace chamber and emitting only the radiation reflected from the walls It is advisable, that this reference source has the same emissivity,Er,as that of the charge ,e. The difference signal ofboth pyrometers is :

CHARGE TEMPERATURE SENSOR

FURNACE

1 2 9 rrr

rr r rr

v CHARGE

1

o WALL TEMPERATURE SENSOR

MICROPROCESSOR SYSTEM

Figure 11.11 Two-pyrometer method for eliminating the influence of reflected radiation

(Ircon Inc., 1996a)