AU685840B2 - Determination of coating thickness and temperature of thinly coated substrates - Google Patents

Description

P/00/011 Regulation 3.2

AUSTRALIA

Patents Act 1990

ORIGINAL

COMPLETE SPECIFICATION STANDARD PATENT S. Invention Title: "DETERMINATION OF COATING THICKNESS AND TEMPERATURE OF THINLY COATED SUBSTRATES" *000 The following state,. int is a full description of this invention, including the best method of performing it known to the Applicant:-

_I,

1 DETERMINATION OF COATING THICKNESS AND TEMPERATURE OF THINLY COATED SUBSTRATES TECHNICAL FIELD This invention relates to the measurement of the temperature of coated substrates and the thickness of the coating. Such measurements are required, for example, for the efficient operation of continuous paint lines whereby galvanised or similarly coated steel strip is provided with a protective and/or ornamental, organic polymeric coating.

.:oeoi BACKGROUND ART It is well known that conventional single colour radiation pyrometers are subject to large errors when measuring the temperature of metal substrates coated with thin organic films. One of the major problems is that the radiancy of the target depends on both its temperature and its emissivity, and emissivity varies with the thickness of the coating. Without complementary tests it was not possible hitherto to know whether any change in radiancy was due to a real change in temperature or simply a change in coating thickness.

ooooe In continuous paint lines of the kind referred to, so called single colour radiation pyrometers, that is to say pyrometers responsive to radiation of a narrow band of wavelengths, (for example a narrow band centred upon a wavelength of 3.43 pm) are frequently used to measure strip temperature at the exit ends of the primer coat and finish coat curing ovens. The total paint thickness after application of the finish coat is about 20 pm, which is thick enough, for practical purposes, to lu hide the metal substrate. On the other hand, the primer coat thickness is usually only about 6 pm, and the pyrometers can see partly through the coat to the metal underneath. So as to at least partially compensate for this, those pyrometers have been set to an emissivity of about instead of the characteristic emissivity of thick organic materials at commonly used infra-red wavelengths of about 0.90-0.96. The dependency of the emissivity of the surface on primer coating thickness makes pyrometric measurement of the temperature of primer coated steel strip more subject to errors than it is for finish coated strip.

A compounding factor is that most organic films which require heat treatment for curing are raised to a temperature not very far above room temperature. At such low temperatures there is not much short wavelength radiation, so the wavelength to which the pyrometer is sensitive has to be quite long. One of the most widely used long wavelength bands is 8-14 pm, which is well beyond the peak in the Planck radiation curve even for bodies at room temperature. Organic Sfilms absorb and emit quite strongly in the 8-14 pm range, but stray radiation from the surroundings at room temperature also is significant and cannot be neglected. Thus the target is bathed in radiation from the 20 background, and any change in target emissivity results in a change in oooeo target radiancy and a somewhat compensatory change in the background radiation reflected from the target into the pyrometer.

When the temperature of very thin films that are transparent at visible wavelengths (for example, a 1pm thick acrylic anti-soil surface layer) on an aluminium/zinc alloy coated or similar substrate, the above problems are quite severe. Under those circumstances, even using radiation of carefully selected wavelengths does not allow the metal substrate to be hidden by the coating. It might be thought that a I I~ wavelength could be chosen such that the radiation is not strongly absorbed or emitted by the coating, so that the emissivity of the coated material would be that of the metal substrate. However, at long wavelengths the emissivity of the metallic coating is quite low, typically less than 0.1, so that stray radiation reflected from the surface into the pyrometer may actually swamp the intrinsic radiancy of the target.

US Patent specification No.4,881,823 Tanaka et al) describes a method of determining the temperature of a heated target material based on prior knowledge of what is herein termed a "reference function", being in the instance of that Tanaka specification, the relationship between two spectral emissivities for the target material, which emissivities are respectively related to different selected components of the emitted radiation. According to the Tanaka specification, two radiances are measured, one for each selected component of the radiation. A target temperature is assumed and two emissivities are calculated, one in respect of each radiance measurement. If the calculated emissivities satisfy the prior known :relationship (the reference function, as that term is used herein), the assumed temperature is taken to be the true target temperature. If not, 20 a further temperature assumption is made and the calculation repeated until an assumed temperature is arrived at for which the calculated emissivities do meet the relationship. The aforesaid components may be selectFd on the basis of wavelength or the orientation of the line of sight of the radiance measuring means relative to the surface of which the temperature is to be determined or relative to the plane of polarisation of the total radiation after it has travelled through a polarising device.

I II I DISCLOSURE OF THE INVENTION An object of the present invention is to provide a method for accurately determining the temperature of a metallic strip with an organic coating that is not thick enough to absorb all of the radiation reflected from the metal surface, for example hot dip metal coated steel with a primer paint coat or an anti-soil coat thereon (referred to as a thinly coated strip hereinafter).

In summary, in respect of a thinly coated strip it would be iii desirable to operate at a wavelength where the emissivity of the surface is high and stable, because this would minimise the effects of changes in the coating thickness and stray radiation, but unfortunately no such wavelengths exist for such thinly coated strip. Operating at a wavelength where the emissivity is low and stable leads to very large .:i stray radiation errors. In the intermediate region, that is to say at 15 wavelengths where the emissivity is moderately high and dependent on Scoating thickness, emissivity related errors are dominant.

Having regard to the foregoing, rather than attempting to compensate for large amounts of stray radiation, the present invention provides a technique which operates in the intermediate region utilising wavelengths within the range of from 3 pm to 14 and allows the temperature to be determined accurately by substantially eliminating emissivity errors.

Because thin organic coatings are frequently invisible to the eye, it is quite possible for production of thinly coated strip to continue for some time even though, as a result of a coating station failure, no coating is I being applied. Thus a further advantage of the invention is that it may provide an indication of the presence or absence of the coating, in that it also enables the thickness of the thin coating to be estimated to an accuracy at least sufficient for that purpose.

The present invention is based upon a newly derived reference function, being a mathematical model relating the effective emissivity, for a particular wavelength of a coated substrate to the specific spectral emissivity of the coating material, the emissivity of the substrate surface (both of which are inherent material properties that can be accurately determined in advance), the thickness of the film and the ambient temperature.

The "ambient temperature" referred to herein is the temperature of the source that is radiating background or stray radiation onto the target, and, in accordance with the invention, means are provided to define that 15 source and so enable its temperature to be measured.

0*• Having calculated an effective emissivity one may readily convert a black body temperature, as indicated by a radiation pyrometer set for a target emissivity value of unity, to a target temperature, and if the calculated effective emissivity is in fact the true effective emissivity, then the calculated target temperature would be the true target temperature.

According to the invention three data readings are taken substantially simultaneously, namely a thermometer reading of the true ambient temperature, and two radiation pyrometer readings of the apparent black body temperature or radiance of the target utilising radiation of respectively different wavelengths. A film thickness is assumed and the data readings used, by reference to the reference i 6 function, to calculate two effective emissivities and two values for the target temperature. If the calculated target temperatures coincide, then the calculated temperature is the true target temperature and the assumed thickness is the true film thickness. If the calculated target temperatures are not equal, the process is repeated with a different assumed film thickness.

The term "apparent black body temperature" is used to indicate that, where a commercial pyrometer including its conventional control unit is used, the pyrometer reading is taken with the control unit's emissivity setting adjusted to 1.0, so that the pyrometer is acting simply as a radiancy measuring device and indicates a temperature consistent with the target being a black body.

An advantage flowing from using the newly derived reference function is that it enables an accurate value for the effective emissivity of a thinly coated strip to be determined from readily and accurately measurable parameters. Another advantage is its V. C universality, given accurate specific emissivities for the materials concerned; and the way in which an initial calculated temperature discrepancy may be used to give a good indication of

C

the required change in the assumption as to the film thickness.

This enables the required result to be obtained quickly with a minimum of iterations of the calculation.

Thus the invention consists in a method of determining the magnitude of a characteristic of a radiant target comprising a metallic substrate coated with a thin coating of organic material, said characteristic being either the temperature of the target or the thickness of the thin coating, said method comprising the steps of: 'r 7 Ist determining a first radiance value of the target for radiation of a first band of wavelengths within the range of from 3 pm to 14 pm and centred on a first wavelength, determining a second radiance value of the target for radiation of a second band of wavelengths within the range of from 3 pm to 14 pm, being different from, and preferably discrete from, said first band of wavelengths and centred on a second wavelength, determining the value of an ambient temperature, being the temperature of the source of extraneous radiation falling on the target, calculating two magnitudes of said characteristic using the respective radiance values determined by steps 1 and 2, an assumed If value for the coating thickness, the first and second wavelengths and a reference function relating the effective emissivity of the target for radiation of a designated wavelength with the specific spectral emissivity of the coating material, the emissivity of the substrate surface and the coating thickness, comparing the two calculated magnitudes, and in the event of a discrepancy therebetween in excess of a predetermined maximum allowable error, reiterating steps to using a more accurately 20 assumed coating thickness until the said discrepancy is not in excess of said allowable error, o•°oo whereat the calculated magnitudes are correct to an order of a a accuracy determined by the size of said allowable error.

Thus, in a preferred embodiment the invention provides a method of determining the temperature of a hot target comprising a metallic substrate coated with a thin coating of organic material, said method comprising the steps of: II ~Lr~ 8 determining a first radiance value of the target for radiation of a first band of wavelengths within the range of from 3 pm to 14 pm and centred on a first wavelength, determining a second radiance value of the target for radiation of a second band of wavelengths within the range of from 3 pm to 14 pm, being different from, and preferably discrete from, said first band of wavelengths and centred on a second wavelength, determining the value of an ambient temperature, being the temperature of the source of extraneous radiation falling on the target, assuming a coating thickness value, calculating for each radiance value a target temperature utilising said first and second wavelengths and a reference function relating the effective emissivity for a designated wavelength with the specific spectral emissivity of the coating material, the emissivity of the 15 substrate surface and the coating thickness, comparing the calculated target temperatures and in the event of a discrepancy therebetween in excess of a predetermined amount reiterating steps to using a more accurately assumed coating thickness until the said discrepancy is not in excess of said 20 predetermined amount, whereat the calculated target temperatures and the corresponding assumed coating thickness are deemed to be correct.

In preferred embodiments the respective radiance values are the indicated black body temperatures as read by a pyrometer responsive to radiation of the selected wavelength band. If, as is preferred, a conventional commercially available pyrometer is used with a control unit able to be set to an assumed emissivity for the target, then that control is set for an emissivity of 1.0, so that the pyrometer is acting 1. I simply as a radiancy measuring device and indicates a target temperature value consistent with the target being a black body.

In preferred embodiments the pyrometer and an ambient temperature thermometer are housed within a shield intercepting substantially all extraneous radiation that would otherwise fall upon the target. Thus the only substantial source of extraneous radiation is the interior surface of the shield and its temperature, which is readily measurable, constitutes the ambient temperature of the reference function.

10 BRIEF DESCRIPTION OF THE DRAWING i By way of example, an embodiment of the above described boe invention is described in more detail hereinafter with reference to the 0. See accompanying drawing, being a diagrammatic section on a horizontal plane through a temperature measuring station of a steel strip 15 continuous coating line.

BEST MODE OF CARRYING OUT THE INVENTION The development of the reference function establishing the

S

relationship between temperature and radiation from a target body comprising an organic film on a metal substrate, is set forth below.

Radiation is emitted from the coating and the metal. Some of this radiation reaches the interface between the coating and the air, and is either transmitted or reflected back towards the metal. Some of the reflected radiation reaches the metal surface, and is either absorbed or reflected back towards the air. As the radiation moves through the 4 coating a proportion is absorbed by it. These processes are considered :eparately below.

EMISSIVITY OF A THIN COATING At a particular wavelength the organic coating material has a characteristic specific emissivity/absorptivity. The bulk spectral emissivity of a thin coating or film depends on the thickness of the film, so the specific emissivity has the dimensions The emissivity of a very thin layer of film is given by:- 6 8,6x (1) 0 where a. is the specific emissivity of the film material at wavelength X, 10 and 6x is the thickness of the thin layer.

The temperature difference between any part of the organic coating and the surface ofthe metal is negligible.

Consider the film as divided up into an arbitrary number of very thin elemental layers of thickness 5x. Consider the radiation moving towards a sensor spaced from the surface of the film and looking to wards it. The radiancy of each of the layers, and of the first layer in particular, is:- 8Ri 8X.6x.R(, T) (2) where R(X, T) c is the Planck radiation formula for a 1-1 (e k 171 -1) black body at temperature T, ^1 11 c 3.74128E-16 is the first Planck radiation ciostant, and k 0.014385 is the second Planck radiation constant.

The radiation produced by elemental layer 1 is partially absorbed by each of the succeeding layers, so the amount remaining after it has passed through n-1 layers is:s1-, 6R n-1 (3) Similar absorption takes place for radiation from each of the other layers, so that the direct radiation reaching the detector from the ith layer Is:- 6 bRi 1 (4) Summing the elementary intensities, and allowing n to become very large and 6x to become very small, it can be readily shown that the total radiancy seen by the detector is given by:- R R(X, T) (l-e 6 X) where x is the thickness of the coating.

9 REFLECTION AT AIR/FILM INTERFACE Equation does not take account of internal reflection at the surface of the film. A proportion of the radiation is transmitted, and the remaining proportion is reflected back into the film.

r= 1-t (6) In practice, for organic coatings:t 0.96, and r 0.04 (8) The reflected portion of the radiation (rRsub) heads back towards the metal surface, being absorbed gradually along the way. The amount of radiation that actually reaches the metal surface is rRsu b -@X n which as n-oo becomes rRsube- x n a REFLECTION AT METAL/COATING INTERFACE At the metal surface a proportion e etal of the absorbed, and the remainder is reflected:- Amount reflected (rRsube-x) (l-EMetal) radiation is a This reflected radiation makes its way back to the film/air interface, suffering attrition along the way. Only a proportion e-ax actually gets back to the surface. Thus the amount of radiation that reaches the surface after two passes is given by:- RSUb2 Rsub I +r 912X (1-EMtail) (11) Extending (11) to many passes gives a geometric series which converges to:- R T) (1-e-x) Rsubn -re-x (1 Eme 1-re (1-emetal) (12) The amount of radiation that escapes and is seen by the pyrometer is:-

L

R T) film out ex2 I 1-rea (1-ecea

I

(13) THERMAL RADIATION FROM THE METAL SURFACE The spectral radiancy of the metal surface is given by:- RMeta EMeta R T) (14) *c e« Tracking this radiation backwards and forwards as before gives the amount of radiation from the metal surface which is transmitted to 5 the pyrometer as RMetal out t ~talR (I T) e x 1-re -8,2 (1 _EMetal) *a a EMISSIVITY OF THINLY COATED SURFACE Summing equations (13) and (15) gives the total radiation emitted by the resin coated metal surface as Rcomposi te tR T) 1 etale-xx) 1-re -2x (l-eMetal) (16) Therefore the emissivity of the composite surface is: Ecozposjte t EMeale -ex) l-re -2x (1 -EMeta (17) Although equation (17) does not take account of interference effects from the multiple reflections within the coating, it is quite accurate for thinly coated strip for the following reasons I L_ 14 -interference is not noticed in practice at visible wavelengths, and similarly would not be detectable at the wavelengths under consideration, the metal substrate and the organic film are optically rough, the transmission coefficient for radiation passing from the film to the air is large.

STRAY RADIATION AND EFFECTIVE EMISSIVITY One more factor must be considered to completely account for the 10 radiation received by the pyrometer, namely extraneous or stray radiation from the surroundings. Stray radiation is particularly important at longer wavelengths (8-14 pm), where tests have shown that it may equal the intrinsic radiation from the target.

*0 The total radiation received is Rpyzo Rcmposite (1-EComposite)R(X, Tanb) (18)

S

where Tab is the ambient temperature.

Effective emissivity may be defined as Pyr (19) Eff- R(IT (19) which is the ratio of the radiancy seen by the pyrometer to the radiancy of a black body at the temperature of the target.

~II

CHOICE OF WAVELENGTHS There are several factors which control the wavelength bands which are available for use in methods according to the invention. The short wavelength end is limited by the need to have enough radiant energy to operate the detector. For temperatures as low as 40 or 500C the shortest wavelength band in commercially available pyrometers is a narrow band centred on 3.43 im.

At the long end of the wavelength range it is inevitable that the I organic film will contribute less to the emissivity of the composite surface 10 as the wavelength increases, because the wavelength starts to become ,.very much greater than the film thickness. Thus the emissivity of the composite surface becomes less dependent on the film thickness, and the inventive method would break down. Another difficulty arises because the emissivity of the metal substrate becomes lower at longer 15 wavelengths, so that stray radiation dominates the intrinsic radiancy of the target. The practical long wavelength limit for use in effecting the invention in commercially available pyrometers currently available is the band 8-14 pm.

a Between 3 and 14 pm there are stro~ig absorption bands from water and C02 in the atmosphere. It is preferable to avoid these bands, as they will tend to hide the target from the pyrometer. The danger wavelengths are around 2.9 pum, and longer than 12 pm. C02 has a strong absorption band at about 4.2 pm.

The reflection spectrum of an acrylic coating on commercial aluminium/zinc alloy coated steel strip shows strong absorption peaks at I 3.43 and 6.85 pm (C-H bond resonance), 5.77 pm (C=O double bond), and 7.92 pm (C-O bond), as well as several other smaller peaks. One of the water absorption bands is very close to 5.77 pm, so this wavelength is not very suitable. Of the remaining candidate wavelengths it is believed that commercially available pyrometers only cover very narrow bands centred on 3.43 and 7.92 pm and 8-14 pm.

Thus those wavelength bands are currently preferred for use in methods according to the invention.

As indicated earlier, putting the present invention into effect 10 requires a knowledge of some inherent properties of the materials involved, that is to say properties that are independent of the variable coating thickness and strip temperature, but which are constants in the reference function defining how radiance is related to those variables -moil• and ambient temperature. These constants may be determined by :i 15 experimental data derived from measurements taken in the laboratory S-using test samples of known dimensions of the coated strip to be monitored at known temperatures.

Specifically the material dependent constants concerned are the specific emissivity of the coating material the emissivity of the substrate surface (EMetal) and the reflection coefficient of the coating material/air interface Briefly stated, the determination of these material dependent constants involves measuring the effective emissivities of the known test samples, which may be compared directly with the results of equation Fitting the theoretical effective emissivity equation to the experimental data by adjusting E, metal, r for different wavelengths and ambient temperatures gives the experimental values for these parameters.

Values so determined in relation to an acrylic over coated aluminium/zinc alloy coated steel sheet are given in the following Table.

TABLE

Wavelength (um) 3.43 11 Specific Emissivity 220,000 120,000 Metal Emissivity 0.14 0.06 Reflection Coefficient 0.04 0.04 As shown in the drawing, the embodiment of the invention using the above derived reference function now being described may be oe applied to the determination of the coating thickness and temperature of an aluminium/zinc alloy coated steel strip 1 issuing vertically from an induction oven for curing an acrylic film newly applied over the alloy coating in a steel finishing mill.

The relevant values for EMetal and r are determined S 15 experimentally in advance.

Immediately adjacent to the oven exit, two commercially available single colour radiation sensors, so called pyrometer heads, 2 and 3 respectively are mounted within a cylindrical or other hollow shield 4 with an open end directed towards the strip 1. The shield 4 may be, for example, about 300 mm in diameter and may be mounted so that it terminates about 40 mm from the strip suriace. Stray radiation from outside the shield 4 is then practically eliminated from the field of view of the pyrometer heads 2 and 3. The inside of the hollow shield 4 is preferably painted matt black to absorb radiation from the strip 1, and a thermocouple 6, RTD (a device with an accurately known thermal -oefficient of resistance) or other thermometric device may be installed on the inner surface of the shield 4 tube to measure the ambient temperature seen by the pyrometers 2 and 3. Ideally the shield 4 is made from "non-magnetic" material to avoid hot spots or excessive heating from stray fields from the induction heater of the furnace. A shield material with a good thermal conductivity, such as copper or aluminium, is preferred to help ensure that any hot spots are evened out, and thermal insulation is preferably applied around the outside of the shield 4.

10 The pyrometer heads 2 and 3 may be sensitive to narrow bands of radiations centred on 3.43 pm and 1 1pm wavelength respectively and V may be mounted on the shield 4 itself. The two pyrometer heads 2 and 3 are mounted at angles so that the field of view 5 on the strip is common to both of them. Both pyrometer heads are interfaced to a conventional controller unit, with emissivity settings adjusted to unity.

'ndicated temperature readings may be transmitted via 4-20 mA loop to a controlling PLC or computer, which intc.prets the data and exhibits or o records the temperature and thickness measurements.

The PLC or computer, as the case may be, may be programmed to function as follows. The input variables are the ambient temperature and the black body temperatures indicated by the two pyrometers.

These temperatures are used to calculate the spectral radiancies seen by the two pyrometers, by inverting equation 'This is preferred to the alternative of utilising radiance value signals direct from the pyrometer heads, because it is cheaper to use commercially available pyrometers including control units, and because of the testing and control facilities built into such control units.

I

19 In one variation a value is assumed for the thickness of the film, and this is used to calculate strip temperatures for the first and second wavelengths, utilising equations (18) and The difference in calculated temperatures gives an estimate of the error in the assumed film thickness, and this error is used to estimate a new thickness.

Iteration of the film thickness is continued until the two calculated temperatures agree satisfactorily.

In experimental runs it was found that in some instances wherein the film thickness was deliberately increased to values much in excess 10 of 1 pm, the two pyrometers may never give the same temperature, even though both calculated temperatures become stationary. Thus the computer program preferably provides for the iterations to be stopped when the time rate of change of calculated temperature falls below an acceptable limit.

It was also found that for very thin coatings stability was hard to achieve, especially when the calculated thickness went negative, which may happen if the coating thickness is close to zero and the noise on the measurements is large. Thus the computer program preferably curtails negative thicknesses.

In the event that significant noise errors are present in the signals fed to the computer, it is possible for the iterated calculated values not to converge unless the "gain" of the loop is sufficiently small to give the required stability.

Taking these factors into account it is possible to produce a program that is stable under all reasonable conditions, and which gives tolerably small errors for likely random variations in input data.

aim By way of example, a satisfactory program is set forth hereinafter written in the BASIC language. That program is written as for a single determination; in practice, in situations where continual monitoring of a nominally unchanging strip is concerned, various unchanging inputs, such as wavelength and material dependent constants would simply be stated instead of appearing as input data.

An alternative program uses the estimated thickness and the radiancy seen by one of the pyrometers to calculate the strip temperature. The calculated strip temperature and the radiancy seen by 10 the second pyrometer then give a new estimate of the film thickness.

Iterations continue until successive estimates of the thickness converge.

Benchmarking of those alternatives over a wide range of conditions showed that the second version of the invention is up to three times faster, but has 50% greater temperature errors when the film 15s 15 thickness is very small.

21

PROGRAM

100 REM INITIALISE DATA 110 120 130

INPUT

INPUT

INPUT

L1 L2 Ecl a.

*SSS**

.SSSS

150 INPUT Ec2 170 INPUT Eml 190 INPUT Em2 210 INPUT refll 15 230 INPUT refl2 250 INPUT x :REM Wavelength 1 (for example 3.43E-6) :REM Wavelength 2 (for example 11E-6) :REM Specific emissivity of coating at L1 (for example 22E4) :REM Specific emissivity of coating at L2 (for example 12E4) :REM Emissivity of metal substrate at L1 (for example 0.14) :REM Emissivity of metal substrate at L2 (for example 0.06) :REM Coating/air coeff. of refln. at L1 (for example 0.04) :REM Coating/air coeff. of refln. at L2 (for example 0.04) :REM Nominal thickness of coating (for example 1.0E-6) :REM ambient temperature C :REM indicated temperature from pyrometer 1 C :REM indicated temperature from pyrometer 2 C 270 280 290 300 310 320 330 340 350 360 370 INPUT Ta INPUT T1 INPUT T2 c 3.74128E-16 k 0.014385 REM START OF Ta Ta+273 T1 T1+273 T2 T2+273 I= 1/18E6 T1g Ti 100 :REM First Planck constant :REM Second Planck constant OUTER LOOP :REM Convert to absolute scale :REM Initial convergence parameter :REM Initial old guess temperature I a U a a e a.

a a a 380 390 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540 550 560 570 580 590 600 610 620 630 640 650 REM SPECTRAL RADIANCY R1 R2 c/(L2-5)/(EXP(k/(L2*T2))-1) REM START OF INNER LOOP REM CALCULATED TEMPERATURE AT Li ex-I EXP(-x*Ec1) El -refll -exi +Eml *exl )I(1-refli *exl -Em 1)) R R1 (1-E1)*(cILf5)/(EXP(kI(Li*Ta))-1) Ti gold T1 Tg :REM Save old temperature Tig li,,(Li*LN(1+E1*cI(R*L^5))) 273 REM CALCULATED TEMPERATURE AT L2 ex2 =EXP(-x*Ec2) E2 -refl2)*(1 -ex2+Em2*ex2)I(l -refl2*ex2^2*(1 -Em2)) R R2 (1 -E2)*(c/iL2-5)I(EXP(k/(L2*Ta))-1) T2g kI(L2*LN (1 +E2*cI(R*L2^5))) 273 dT =Ti g- T2g :REM Apparent temperature error dTt T1ig T1igold :REM Time change of temp.

dx dT*I :REM Correction for x I I*.9 :REM Shrink convergence param.

x=x- dx :REM New guess at x IF x 0 THEN x x/10 :REM Cushion against large neg.

IF ABS(dT) 0.2 THEN GOTO 620 :REM Normal break IF ABS(dTt) 0.05 THEN GOTO 620 :REM Backup break GOTO 410 :REM Inner loop IF x 0 THEN x 0 :REM Do not show negatives PRINT x*1 E6, Ti g :REM Output data (pm, 'C) GOTO 320 :REM Outer loop

END

*.aaae a a

Claims (7)

1. A method of determining the magnitude of a characteristic of a radiant target comprising a metallic substrate coated with a thin coating of organic material, said characteristic being either the temperature of the target or the thickness of the thin coating, said method comprising the steps of: determining a first radiance value of the target for radiation of a first band of wavelengths within the range of from 3 rpm to 14 pm and centred on a first wavelength, determining a second radiance value of the target for radiation of a second band of wavelengths within the range of from 3 um to 14 pm, being different from said first band of wavelengths and centred on a second wavelength, determining the value of an ambient temperature, being the temperature of the source of extraneous radiation falling on the target, calculating two magnitudes of said characteristic using the respective radiance values determined by steps 1 and 2, an assumed rA&lue for the coating thickness, the first and second wavelengths and a reference function relating the effective emissivity of the target for radiation of a designated wavelength with the specific spectral emissivity of the coating material, the emissivity of the substrate surface and the coating thickness, comparing the two calculated magnitudes, and in the event of a discrepancy therebetween in excess of a predetermined maximum allowable error, reiterating steps to using a more accurately assumed coating thickness until the said discrepancy is not in excess of said allowable error, whereat the calculated magnitudes are correct to an order of accuracy determined by the size of said allowable error. 7C 0' I"

2. A method according to claim 1 wherein said iterations are stopped and said calculated magnitudes deemed to be correct if the change of the calculated magnitudes from one iteration to the next falls below a predetermined minimum amount.

3. A method according to claim 1 wherein said first wavelength is 3.43 pm and said second wavelength is 11 pm.

4. A method according to any one of the preceding claims wherein said ambient temperature is determined by interposing a thermally conductive shield between the target and surrounding sources of extraneous radiation and measuring the temperature of the surface of the shield exposed to the target.

A method according to claim 4 wherein said surface of the shield is matt black. .:oo.i

6. A method according to claim 1 wherein the respective radiance values are the black body temperatures as read by a pyrometer responsive to radiation of the selected band of wavelengths and having a control unit able to be set to an assumed emissivity for the target, wherein that control unit is set for an S emissivity of to"

7. Apparatus for effecting the method of claim 1 comprising oeoooo a thermally conductive shield substantially preventing direct exposure of the target to extraneous sources of radiation, two radiation pyrometers, which comprising a sensor head and a control unit, of which at least the sensor heads are mounted on or within said shield for direct observation of the target and of which the control units are able to be set for an assumed target emissivity of 1.0, ambient temperature measuring means responsive to the temperature of a surface of said shield that is exposed to the target, and a programmable computing means connected to receive signals from said pyrometers and ambient temperature measuring means, and programmed to effect said iterations of the calculation of the magnitudes of said characteristic. Applicant: BHP STEEL (JLA) PTY. LTD. Date 17 September 1997 Attorney: ROBERT G SHELSTON F.I.P.A.A. of CARTER SMITH BEADLE o** S o a *I, 2 ABSTRACT A method of determining the temperature of a thinly coated steel strip comprising the steps of: determining a first radiance value of the strip for radiation of a first band of wavelengths centred on 3.43 pm, determining a second radiance value of the strip for radiation of a second band of wavelengths centred on 11 pm, determining the value of the strip's ambient temperature, being the temperature of the source extraneous radiation falling on the strip, calculating two magnitudes of the temperature of the strip using the respective radiance values determined by steps 1 and 2, an Poe.*: assumed value for the coating thickness, and a reference function relating the effective emissivity of the strip for radiation of each of said 15 wavelengths with the specific spectral emissivity of the coating material, the emissivity of the substrate surface and the coating thickness, comparing the two calculated temperatures, and in the *O6 event of a discrepancy therebetween in excess of a predetermined maximum allowable error, reiterating steps to using a more 20 accurately assumed coating thickness until the said discrepancy is not in 0 So.: Sa excess of said allowable error, whereat the calculated temperatures are correct to an order of accuracy determined by the size of said allowable error. Figure 1 I