Temperature Measurement Best Practices: Steel Continuous Annealing Line

Precise control of process temperature is critical for the continuous annealing of steel.  The mechanical properties of the product are directly affected by the temperatures attained during this process and by the rate of temperature change.  And, because the process is energy-intensive, precise temperature control is essential for cost efficiencies.  For the continuous annealing process, no parameter is more important than the precise control of temperature.

During this process there are two complicating issues that affect the ability of an infrared pyrometer to produce an accurate measure of temperature:  hot background reflections and a significant emissivity variation of the strip.  Temperature measurement best practices for the steel continuous annealing line effectively address both issues including two separate measurement techniques to produce a highly accurate measure of temperature for effective control of product properties and process costs

Multi-reflective Roller Wedge Measurement Technique

For the roller wedge technique, an infrared pyrometer is installed to view into the roller nip where the strip and roll meet.  This alignment technique serves two purposes.  First, this technique prevents the pyrometer from viewing a reflected image of the hot furnace wall, as the pyrometer views a reflected image of the roll instead.  Next, when measurement conditions are ideal, this technique creates a near-blackbody emissivity condition, as off the steel strip the pyrometer views a reflected image of the roll, and off the roll the pyrometer views a reflected image of the strip, and with each successive reflected image the effective emissivity is increased.  When the measurement conditions are ideal, this technique results in an emissivity of 1.000 at the point of measurement.

The roller wedge technique produces an average temperature reading between the temperature of the strip and the temperature of the roll.  If the two surfaces are not the same temperature, then the reported temperature reading is not the true temperature of the strip.  Similarly, the effective emissivity at the point of measurement is very close to 1.000 only when the pyrometer is precisely aligned to the so-called sweet spot, and the sweet spot is only sufficiently large when the pyrometer is viewing from an appropriate angle.  An angle too close to parallel to the strip and even the slightest change in the angular alignment of the pyrometer will result in a large displacement at the point of measurement.  An angle too far from parallel to the strip and the sweet spot is too small to be viewed clearly.  And finally, an angle too far from perpendicular to the centerline of the strip, and the strip is not wide enough to produce enough multiple reflections to create the necessary near-blackbody condition.

An argument may be made that the temperature reading need not be precisely accurate and that a measurement that is repeatable is good enough for process control; but, when examined carefully, this argument loses validity because an inaccurate measure of temperature is also a non-repeatable measure of temperature.  When the strip and the roll are not the same temperature, the weighting given to the temperature of each surface is proportional to the difference in emissivity between the strip and the roll and when the pyrometer is not aimed precisely at the sweet spot, then the effective emissivity is less than 1.000, and any variations in temperature, emissivity or alignment can introduce significant and variable measurement errors.

Cold rolled steel, for example, is more reflective than hot rolled steel, and so when measuring cold rolled steel the reading is more weighted towards the temperature of the roll and when measuring hot rolled steel the reading is more weighted towards the temperature of the strip.  A changing emissivity of the strip likewise changes the weighting factor between the temperature of the roll and the temperature of the strip, and non-repeatable measurement variations caused by this difference have been measured as great as 30°C.

Roller wedge measurement conditions are valid when the effective emissivity is high and stable.  If the measured emissivity is lower than 0.940, then the roller wedge measurement is not accurate and if the measured emissivity value is not stable, then the measurement is not repeatable.   When using a single-wavelength pyrometer it is not possible to know if the reading is valid or invalid because only one parameter, temperature, is calculated.  To address the important accuracy concerns, the recommended best practice for the roller wedge technique is to use a two-wavelength pyrometer able to produce a measured emissivity reading in addition to the measured temperature reading.  This second measured parameter is used to confirm valid measurement conditions and to identify invalid measurement conditions when they exist.  The measured emissivity value will be low when there is a significant temperature difference between the strip and the roll, when the geometry is unable to produce a high emissivity condition, when the pyrometer is misaligned to the sweet spot, and when the viewing port window is dirty.  Not only is the two-wavelength ratio measurement better able to correct for these process variables, but also monitoring the measured emissivity value provides important warning as to the validity of the measurement and any issues that may influence the accuracy of the measured temperature reading.

Direct View Measurement Technique

For those measurement positions where the multi-reflective measurement technique is not available or does not produce valid readings, the pyrometer must be installed to directly view the surface of the strip.  Here, without the benefit of a roller wedge to enhance the emissivity and to shield against hot background reflections, both interference sources must be accounted for.

If the pyrometer is mounted in a section of the furnace where the background is not hot, then no special provision must be made to account for hot background reflections.  If the pyrometer is mounted in a section of the furnace where the background is hot, then the area of measurement must be shielded from hot background reflections by installing a cooled viewing tube.  In this way, the pyrometer views a reflected image of the cooled surface rather than a reflected image of the hot furnace wall.  If the cooled surface is significantly cooler than the temperature of the strip, then it will have no influence on the infrared pyrometer reading.

Once background reflections are accounted for, then emissivity variation becomes the only remaining complicating issue.  Emissivity varies with alloy, surface texture, degree of oxidation, and elemental migration.  For any given point on the steel strip, the emissivity of the steel changes as it passes through the annealing furnace due to oxide formation, oxide removal and elemental migration to the surface, and for any given coil, the emissivity changes from the head to the middle to the tail.  This emissivity variation prevents an accurate temperature measurement using single-wavelength technology, and because the emissivity variation is different at different wavelengths, ratio (two-color or dual-wavelength) pyrometers are similarly inaccurate.

Ratio pyrometers (two-color or dual-wavelength pyrometers) are designed to compensate for changing emissivity; however, this pyrometer technology assumes that the emissivity and any emissivity variation is the same at both measured wavelengths.  For cold rolled and high strength steel alloys, neither of these conditions is true.   For an accurate measure of temperature at the continuous annealing furnace a multi-wavelength pyrometer is required.

A multi-wavelength pyrometer is used to measure the temperature of non-greybody materials. A non-greybody material is one where the emissivity varies differently at different wavelengths. Multi-wavelength pyrometers are commonly used for the measure of aluminum, copper, zinc, steel annealing lines, and high-alloy steel. This type of pyrometer uses a mathematical algorithm to model the emissive nature of the steel across the wavelength set of measurement. As the emissivity increases and decreases, as the surface texture changes, as the level of oxidation changes, and as alloying elements migrate to the surface, the multi-wavelength pyrometer automatically adjusts to these changing surface conditions to report an accurate measure of temperature.

Precise temperature control throughout the continuous annealing and hot dip coating process provides significant benefits to both product quality and to cost savings. Plants that have adopted these best practices report increasing throughput and lowering costs with benefits totaling millions of dollars per line per year. These significant benefits are achieved making no process changes other than the change to a more effective, accurate and precise pyrometer technology.