Pipe Inspection by Infrared Thermography

This month's  NDT Solution shows how infrared thermography can be used to detect wall thinning in pipes.  The technique provides a qualitative alternative to commonly used techniques such as ultrasonics, X-ray, and inspection via borescope. This technique has the advantages of being fast, safe, and nonintrusive. The feature shows how pulsed active  infrared thermography works to detect the wall thinning on an elbow when the part is loaded with transient thermal gradients. Quantitative results are presented correlating thickness to time at which abnormal thermal patterns appear. G.P. Singh Associate Technical Editor

 

Figures 1-2
Figures 3-4
Figures 5-7

Introduction
Pipes submitted to flow of fluid can be severely internally damaged after a certain period of time because cavitation erosion locally reduces wall thickness. This type of corrosion is most severe in bent portions of pipes (elbows) where turbulences and vortices take place. To prevent damages due to explosion of severely corroded pipes, current practice is to replace pipes on a statistical basis, for example every 10 years (blind replacement). An alternative is to rely on nondestructive evaluation (NDE) to assess wall thickness and then replace only critically damaged pipes. Borescopes, X-rays, ultrasonics, and infrared thermography are useful for such purposes. In this text pulsed active infrared thermography (PAIRT) is examined.

In PAIRT, a transient thermal perturbation of the inspected part allows detection of subsurface defects because of their different thermal properties (such as thermal diffusivity) with respect to surrounding sound material. Such different thermal properties are revealed through the evolution of the surface temperature history observed with an infrared camera (Kaplan, 1993; Maldague, 1994; Maldague, 1993). Interestingly, the thermal perturbation can be either external or internal. PAIRT has the advantages of being noncontact, fast (area scan), harmless, useful for either the reflective or transmissive method, and easy to deploy (no dismantling).

---

Pulsed active infrared thermography is a convenient and cost effective approach to detect corrosion damages in pipes.

---

IRT Inspection Station
Figure 1 presents a schematic diagram of a PAIRT inspection station. Because the infrared (IR) camera is the central piece, if surface emissivity is sufficiently high, relevant temperature differences on the surface being tested can be measured. Emissivity is a surface property describing the ability of the surface to emit energy, it varies from zero for a perfect reflector, such as a mirror, to one for a perfect emitter, such as a blackbody. Low emissivity can be corrected by such methods as applying an high emissivity (around 0.9) paint (black painting). If water soluble paint is used, it can be wiped off after the inspection procedure (black tempera paint is particularly convenient for such purpose). IR images (thermograms) are preferably recorded by direct digital transfer or on video tape, but the latter choice is far less advantageous due to additional required manipulation and image degradation. In either case, a time sequence of thermograms acquired following the transient heat perturbation becomes available. The third element of the inspection station is the heat perturbation source which is often application dependent.

 

Thermographic Inspection In Transmission
Under stationary conditions, temperature differences between areas of different wall thicknesses within a pipe are nonexistent. Significant thermal contrasts are obtained under transient conditions in which case a practical relationship relates the depth, z, of a structure of interest (wall thickness in the case of pipes) with the time of observation, t, when the corresponding thermal contrasts appear on the front surface (t is counted from the start of the thermal perturbation) (Maldague, 1993):

(1)      t = z² /

where is the thermal diffusivity of the material (m²/s).

Following Equation 1 , if a portion of a pipe has its wall thickness reduced by a factor of two due to corrosion, a sudden thermal disturbance will reach the outer surface of this portion after a time period four times smaller than the surrounding portions. In other words, areas with thinner walls will have their temperature affected first. By measuring the time of observation, t, it becomes possible to evaluate quantitatively the wall thickness providing the thermal diffusivity, , of the pipe.

These concepts can be applied experimentally by a simple method:

1. Generate a thermal transient inside the pipe by changing the temperature of the circulating fluid.

2. Observe the temperature distribution of the outside surface with the IR camera and note abnormal temperature patterns.

3. Evaluate wall thickness with Equation 1.

A typical test station is shown in Figure 2. In this case a severely corroded, 90 degree bent portion of a steel pipe is connected to two taps, one cold 6 °C (43 °F) and one hot 40 °C (104 °F), which are supplied by outlet water with a flow of 120 L (32 gal) per minute, to generate thermal transitions. Figures 3 and 4 present results for hot to cold and cold to hot transitions. Change in type of temperature transition does not change the abnormal temperature patterns which clearly reveal damaged zones. However, change in flow direction locally affects vortexes with a small incidence on observation time as illustrated by Figure 5. Also shown in Figures 3 and 4 is the fact the stationary regimes (hot or cold) do not reveal any damage.

As stated before, based on Equation 1, an approximate quantitative wall thickness evaluation is possible. On Figure 6 the wall thickness (depth) versus time of observation is plotted with steel = 10 x 10^-6 m ²/s (10.8 x 10^-5 ft²/s). Also indicated on that figure are the times of observation of a few relevant structures highlighted in Figures 3, 4, and 5 with corresponding measured exact wall thicknesses obtained from visual inspection after the corroded pipe was cut into pieces.

Equation 1 provides a good approximation in cases of isotropic materials such as steel. In case of anisotropic components, such as graphite fiber reinforced plastic (GFRP), more accurate methods such as thermal modeling are needed to perform such investigation (Maldague, 1993; Maldague, 1994).

 

Thermographic Inspection in Reflection
Alternatively, if a change in the water flow is not possible, a reflective method can be implemented. In this case a uniform heat source is applied to the exterior of the pipe using, for instance, a heat gun, and, because of differences in thermal diffusivities between water ( water = 0.14 x 10^-6 m²/s [1.51 x 10-6 ft²/s]) and steel, the thermal front reflected at the interface wall-water reaches the exterior surface at different times depending on wall thicknesses (Equation 1). This allows for the pinpointing of local wall corrosion.

These concepts can be applied experimentally by a simple method:

1. Generate a thermal transient outside the pipe by heating the surface with an heat gun. The circulating fluid in the pipe must be stopped, and the pipe surface temperature must be uniform prior to start.

2. Observe the temperature distribution of the outside surface with the IR camera, and note abnormal temperature patterns.

3. Evaluate wall thickness with Equation 1.

Figure 7 shows result of the reflection method, a 5.4 x 10⁶ J (1500 W) heat gun model induces thermal contrasts. Although the damaged area is visible, the contrast is not as good as with the method in transmission due to much smaller temperature differences involved.

 

Conclusion
Pulsed active infrared thermography is a convenient and cost effective approach to detect corrosion damages in pipes. Both the transmission and reflection methods are easy and fast to implement without dismantling the piping. Measurement of the thermal propagation time following the transient perturbation reveals pipe wall thickness. Among the advantages of the transmission method are high thermal contrast on the pipe surface and absence of reflective noise since temperature perturbation is provoked inside the pipe.

 

References
Kaplan H., "Practical Applications of Infrared Thermal Sensing and Imaging Equipment," SPIE, vol. 13, 1993.

Maldague X., ed., Infrared Methodology and Technology, NY, Gordon and Breach, 1994.

Maldague X., Nondestructive Evaluation of Materials by Infrared Thermography, London, Springer-Verlag, 1993.

*  Electrical and Computing Engineering Department, Université Laval, Quebec City (Quebec), G1K 7P4, Canada; (418) 656-2962; fax (418) 656-3594; e-mail maldagx@gel.ulaval.ca

Copyright © 1999 by the American Society for Nondestructive Testing, Inc. All rights reserved.

[ Materials Evaluation ]