New and Emerging Inspection Technologies for Flow Accelerated Corrosion in Fossil Power Plants

TP-114349

New and Emerging Technologies for Flow Accelerated Corrosion in Fossil Power Plants
TP-114349 November 1999

EPRI Project Manager Pedro F. Lara

EPRI 3412 Hillview Avenue, Palo Alto, California 94304 PO Box 10412, Palo Alto, California 94303 USA 800.313.3774 650.855.2121 askepri@epri.com www.epri.com EPRI NDE Center 1300 WT Harris Blvd., Charlotte, North Carolina 28262 PO Box 217097, Charlotte, North Carolina 28221 USA 704.547.6100

DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES

THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM: (A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR (B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT. ORGANIZATION(S) THAT PREPARED THIS DOCUMENT EPRI NDE Center

ORDERING INFORMATION
Requests for copies of this report should be directed to the EPRI Distribution Center, 207 Coggins Drive, P.O. Box 23205, Pleasant Hill, CA 94523, (800) 313-3774. Electric Power Research Institute and EPRI are registered service marks of the Electric Power Research Institute, Inc. EPRI. POWERING PROGRESS is a service mark of the Electric Power Research Institute, Inc. Copyright 1999 Electric Power Research Institute, Inc. All rights reserved.

CITATIONS
This document was prepared by EPRI NDE Center 1300 WT Harris Blvd. Charlotte, NC 28262 Principal Investigator I. Lara, Pedro

This document describes research sponsored by EPRI. The publication is a corporate document that should be cited in the literature in the following manner: New and Emerging Inspection Technologies for Flow Accelerated Corrosion in Fossil Power Plants: 1999. TP-114349.

iii

ABSTRACT
This document describes the Nondestructive Evaluation (NDE) technologies that are currently available or under development to help quantify the integrity or the rate of degradation of piping affected by flow accelerated corrosion in fossil power plants. The report classifies the technologies into those that require the pipe to be bare (lightly coated) and those that can measure through insulation or liners. Also, it presents the technologies that are best suited to perform a local metal loss assessment or provide a more complete or global view of the system.

NEW AND EMERGING INSPECTION TECHNOLOGIES FOR FLOW ACCELERATED CORROSION IN FOSSIL POWER PLANTS
Executive Summary Flow accelerated corrosion (FAC) causes leaks and ruptures in carbon steel piping in power plants. Because these ruptures have included worker fatalities and significant property damage in some situations, the issue has been the focus of considerable attention in the past few years. In response to these events, EPRI has published several reports that address various aspects of the corrosion process including mitigation 1, component susceptibility analysis 1,2,3, component degradation prediction methodology 3,4,5, and recommended nondestructive evaluation (NDE) practices 6,7. This report revisits the NDE practices while incorporating the recent technology advances. These new NDE developments provide the utility owner with increased thickness measurement accuracy and better tools to assess wall thickness through insulation. In addition, some of these new technologies are aimed at expanding the inspection area for a very low incremental cost; that is, at providing a more global view of the system for the same dollars invested in inspection. In this report the technologies are divided into contact and non-contact; that is, those that require the piping to be bare (thinly coated), and those that can survey through insulation. Furthermore, the technologies are classified according to their suitability to assess the corrosion rate. EPRIs degradation prediction methodologies include corrosion rate assessments of the components. The corrosion rate estimation is then used as part of the planning process for the next inspection or for replacing the component. This evaluation found that, Monitoring the rate of FAC in bare piping has been made more cost effective with the advent of AutoGrid, a new technology. FAC detection through insulation is feasible, with some limitations, and the availability of new technologies (real-time radiography and pulsed eddy current) have made the inspection operations more cost effective. Measuring the rate of FAC through insulation is feasible and the improvements in the pulsed eddy current and Compton backscattering techniques may make the in-service assessments more plausible. Methods for detection of FAC globally are being developed but are not yet established.

Background Flow-accelerated corrosion (FAC) has been the cause of many carbon steel piping and vessel failures that carry water or steam. Because these ruptures have included worker fatalities and significant property damage in some situations, the issue has been the focus of considerable attention in the past few years. In response to these events, EPRI has published several reports that address various aspects of the corrosion process including mitigation 1, component susceptibility analysis 1,2,3, component degradation prediction methodology 3,4,5, and recommended nondestructive evaluation (NDE) practices 6,7. The mechanism that causes FAC is a combination of metal oxidation from iron to magnetite, and the dissolution or removal of the magnetite layer by the fluid flow. At low fluid velocities, the dissolution of the magnetite is slow enough and a protective layer is formed on the metal surface. Under these conditions the rate of corrosion is relatively slow and controlled by the mass transfer of ions through the magnetite layer. As the fluid velocity increases above the so called breakaway value, the protective oxide film is removed by the surface shear stress and the corrosion accelerates to significantly higher rates, so the term flow-accelerated corrosion 1. The mechanism suggests that impurities contributing to the dissolution of the protective layer enhance the corrosion process. In particular, although the presence of oxygen is a contributor to the corrosion process, the total removal of oxygen from the system is detrimental since no protective layer can be formed. Traces of oxygen in the range of 100 parts per billion are typically recommended 1. FAC occurs in single and 2-phase flow carbon steel systems, and in both small and large bore piping. Components that promote the formation of vortices, secondary flows, or turbulence are more prone to FAC 1. These include: Elbows, Bends, Tees, Reducers, Pipe entries, Downstream of valves and flow control orifices.

For fossil power plants the following systems have been determined to most likely experience FAC 5: Piping around boiler feed pump, Tubesheets and tubes in the high and low pressure heaters, Heater drain lines, Economizer inlet tubing, Piping to economizer header, Deaereator shell, Heat recovery steam generator tubing.

This report revisits the NDE practices applicable to carbon steel piping while incorporating the recent technology advances. In this report the technologies are classified according to the level of technical maturity into 3 categories: established, new, or emerging. Established techniques exhibit wellaccepted deployment procedures and accuracy levels. New techniques have proven performance under some scenarios but have not yet gained wide acceptance. Emerging technologies have not yet demonstrated field ability to size damage with any accuracy. Established technologies provide the user with an accurate assessment of the local pipe conditions. The user then estimates the overall piping integrity by inferring wall condition based on these localized measurements (statistical inference). In order to help reduce the risk of not identifying a deteriorated area while performing these limited surveys, EPRI has developed CHECWORKS, 8, a comprehensive predictive model that helps the utility operator target for inspection those pipe locations that are likely to suffer the highest corrosion rates. Some of the new and emerging NDE technologies discussed in this report are aimed at further minimizing the integrity uncertainty by expanding the area of inspection for a very low incremental cost. The objective is to provide the utility operator with a more global view of the system for the same dollars invested in inspection. In this report the technologies are divided into contact and non-contact, that is, those that require the piping to be bare (thinly coated), and those that can survey through insulation. Furthermore, the technologies are classified according to their suitability to assess the corrosion rate. EPRIs degradation prediction methodologies include corrosion rate assessments of the components. The corrosion rate estimation is then used as part of the planning process for the next inspection or for replacing the component 5. The inspection technologies are listed in Table 1 (page 4) for bare piping, and in Table 2 (page 5) for insulated piping. These technologies are discussed next.

Table 1 Inspection Technologies for Bare Piping

Established Local Metal Loss Detection Radiography New Emerging Accuracy +/- 5% for Ug=0.02 inch +/- 0.005 in. Wide-scan Real Time Radiography Narrow-scan Real Time Radiography Low Frequency Electromagnetics Corrosion Rate Assessment Global Metal Loss Detection Autogrid + Ultrasonics Magnetostrictive Guided Waves Piezoelectric Guided Waves Qualitative assessment Comments Provides accurate wall thickness measurement and good pit resolution. Expensive when per unit area of coverage. Use Iridium sources for equivalent steel thickness < 2", Cobalt for equivalent steel thickness < 6". Provides accurate wall thickness measurement of bare or coated piping. Accuracy affected by pipe temperature and roughness. Suitable for an rapid wide-area scan to identify areas that need further evaluation in straght water filled piping 8" to 36" in diameter and up to 24" diameter elbows. Can resolve a flaw with a radiographically projected image of 1/4" in diameter and survey 1" away from obstructions. Requires 4" and 12" clearance at the detector and source side respectively. Suitable for an rapid narrow band scan to identify areas that need further evaluation for pipes up to 18" in diameter including insulation. Can detect a 35% 3/4" diameter FBH in a 6 inch diameter pipe. Sensitivity will be less for larger diameter piping. Suitable for an rapid wide band scan to identify areas that need further evaluation in water filled piping. Can detect isolated pits through coatings up to 0.2" thick. Performance may be affected by material property changes. Suitable for corrosion monitoring. Combines the accuracy of ultrasonic measurement with precise sensor placement. Suitable for qualitative evaluation of large inaccessible areas. Can detect a 1% pipewall cross section reduction with a 7" spatial resolution. The inspection range best for piping up to 16" in diameter and depends on carrier fluids and piping geometry. Device still under development for global detection of piping damage. Can detect a 0.46% reduction in cross-section with a spatial resolution of 1.5". Range of detection dependent on wave-mode. Rules for wave-mode selection still under development. Suitable for qualitative evaluation of inaccessible areas. Can detect a 10% volume loss over the 1.5" wide sensor-to-sensor band. Performance may be affected by the presence of internal deposits. May detect 25% FBH with diameters equal to nominal wall thickness. Lesser defect depths may not be detectable. Performance affected by the coating thickness. Vendors various

Ultrasonics

various IHI Southwest

Not established Not established +/- 0.005 in.

Lixi

Testex

Southwest Research Institute Southwest Research Institute Penn State University; Plant Integrity Ltd. (UK) Magna-Tec All-Tech Thermal Wave Imaging

Not established Not established

Tangential Guided Waves Active Infrared Thermography

Not established Not established

Table 2 Inspection Technologies for Insulated Piping

Established Local Metal Loss Detection Radiography New Emerging Accuracy +/- 5% for Ug=0.02 in. Wide-scan Real Time Radiography Narrow-scan Real Time Radiography Pulsed Eddy Current Compton Backscattering Corrosion Rate Assessment Pulsed Eddy Current Compton Backscattering Global Metal Loss Detection Magnetostrictive Guided Waves Piezoelectric Guided Waves Qualitative assessment Application Provides accurate wall thickness measurement and good pit resolution. Expensive per unit area of coverage. Use Iridium sources for equivalent steel thickness < 2", Cobalt for equivalent steel thickness < 6". External corrosion products may interfere with the measurement. Suitable for an rapid wide-area scan to identify areas that need further evaluation in straight water filled piping 8" to 36" in diameter and up to 24" diameter elbows including insulation. Can resolve a flaw with a radiographically projected image of 1/4" in diameter and survey 1" away from obstructions. Requires 4" and 12" clearance at the detector and source side respectively. Suitable for an rapid narrow band scan to identify areas that need further evaluation for pipes up to 18" in diameter including insulation. Can detect a 35% 3/4" diameter FBH in a 6 inch diameter pipe. Sensitivity less for larger diameter piping. Suitable of general wall loss detection of piping through insulation. Will not detect isolated damage. Vendors various

IHI Southwest

Not established +/- 0.04"

Lixi

RTD APTECH Nuclear Measurement Corp. RTD APTECH Nuclear Measurement Corp. Southwest Research Inst. Penn State University; Plant Integrity Ltd. (UK)

Not established +/- 0.04 in.

Device still under development. May be suitable for measurement of localized damage through insulation. Exhibits repeatibility of +/- 0.005 inch which is suitable for corrosion monitoring. Grid points can be painted on the insulation, or tool integrated with Autogrid. May be suitable for corrosion monitoring once the tool's repeatability is quantified. Grid points can be painted on the insulation, or tool integrated with Autogrid. Suitable for qualitative evaluation of large inaccessible areas. Can detect a 1% pipewall cross section reduction with a 7" spatial resolution. The inspection range best for piping up to 16" in diameter and depends on carrier fluids and piping geometry. Device still under development for evaluation of inaccessible areas. Can detect a 0.46% reduction in cross-section with a spatial resolution of 1.5". Range of detection dependent on wave-mode. Rules for wave-mode selection still under development.

Not established Not established Not established

FLOW ACCELERATED CORROSION INSPECTION OF BARE PIPING NDE technologies for bare piping include those that are used for localized assessment, those that are suitable for corrosion rate monitoring, and those that are targeted to provide a global view of the system. Local Detection of Metal Loss in Bare Piping Established technologies for localized metal loss assessment include ultrasonics and radiography. These techniques have been extensively documented elsewhere including two comprehensive EPRI reports 6,9. These reports list the essential variables, accuracy, and operating procedures of both ultrasonics and radiography. Because of the availability of these documents only a brief description of the techniques will be given here. Ultrasonics Ultrasonic thickness measurement can be performed on bare ambient temperature piping with an accuracy of +/- 0.005 inch. As the piping temperature increases, the accuracy of the ultrasonic measurement decreases and calibration corrections are warranted. Temperature affects the couplants viscosity, the delay line velocity, and to a small degree the speed of sound in the steel (the delay line is the nonmetallic material that is placed in front of the transducer for protection purposes). When surveying hot piping, ASTM E797 recommends a -1% per 100F correction to the thickness measurement when the calibration procedure calls for ambient temperature calibration blocks 10. Ultrasonic thickness measurement accuracy also decreases as the backwall surface roughness increases because the uneven surface scatters and attenuates the backwall echo. When surveying corroded surfaces, the use of dual-crystal transducers with a focal distance close to the nominal thickness is recommended. Radiography Radiographic techniques can be used to measure the pipe wall thickness as well as detecting the aerial extent of localized damage. FAC measurements with radiography use the tangential technique (Figure 1) 6,9. In this technique a radiographic image of the pipe wall is obtained by configuring the equipment so that the source-to-film centerline is tangent to the pipe section to be surveyed. The pipe wall thickness is obtained from the image by multiplying it times the calculated magnification factor. Alternatively, a specimen of known dimensions can be placed on the pipe surface at the tangent point, and the magnification factor estimated from the specimens image. Wall thickness measurements from tangential radiography have an estimated accuracy of +/- 5%, when the geometrical unsharpness is less than 0.02 inch.

Figure 1. Radiographic Techniques 9

Radiographic measurements can also be performed using the double wall technique. This technique is useful to identify localized damage on a component and to accurately assess the aerial extent of the damage. It will not, however, accurately size the depth of wall damage. Therefore, the double wall technique must be supplemented with other NDE methods when performing component integrity or life assessments. New technologies for metal loss detection in bare piping include wide- and narrow-scan realtime radiography. Wide-Scan Real-Time Radiography A new radiographic method, called ThruVu, has been developed under EPRI funding to scan piping over a wide area and display the acquired image in real time 11. This real-time radiographic method permits identification of wall thinning areas, as well as localized damage at a much lower cost per unit of inspected area when compared to conventional radiography. Once deteriorated areas are qualitatively identified, subsequent examination with more accurate metal loss measurement techniques, such as ultrasonics, is warranted (Figure 2).

Figure 2. Wide Scan Real Time Radiography The ThruVu System 11

ThruVu can be used with various radiographic sources. However, it typically deploys an Amersham 660 camera with a low intensity Iridium-192 source (30 or more curies). The tungsten collimator has a narrow V-shaped slit that produces a fan shaped beam of penetrating radiation. This low radiation intensity level limits the personnel exclusion zone to about 50 feet. ThruVu deploys an array of solid state detectors that are sensitive to variations in radiation intensity. The size and number of solid state elements determines the systems spatial resolution. The linear detector model currently used is 16 inches long and includes 128 1/8-by-1/8 elements. With this configuration, the system is capable of resolving wall loss areas that exhibit a radiographically projected image of in diameter. ThruVu is most efficiently utilized when performing double-wall assessments. The current spatial resolution suggests that its applicability to tangential surveys is limited. The sensor array, when laid perpendicular to the scanning axis, allows the system to perform double-wall radiographic images of straight piping ranging from 8 to 36 in diameter, and elbows up to 24 in diameter. To perform a survey, a temporary track consisting of two guides is secured on the pipe using a pair of brackets. Cords or straps are then used to secure the tracks in place. A motorized cart assembly that includes the source and the detector is placed on the pipe and locked to the track, the latter guiding the cart as it performs the survey. This cart-track design permits the survey of horizontal, vertical, and elbow configurations as long as 4 and 12 inches of clearance is available at the detector and source side respectively. The side clearance requirement is only 1 inch, and the unit can survey within 1 inch of obstructions.

Narrow-Scan Real-Time Radiography Alternatively, a real-time radiographic survey can be performed along a narrow band using a different system design 12. This new technology, the Lixi Profiler, is packaged as a portable hand held unit, which permits the operator to scan at will any accessible piping area for wall loss after performing a simple calibration. The device is suited for identification of FAC followed by further verification with more accurate metal loss evaluation techniques (Figure 3).

Figure 3. Narrow Scan Real Time Radiography The Lixi Profiler 12

The Profiler can be operated as a hand held unit because it uses a very low intensity Gadolinium 153 gamma ray source and its weight is low. The exposure to the operator with this source is typically less than 2 mR, hence, it does not require an exclusion zone, and the device weighs only 9 pounds. Piping up to 18 inches in diameter can be surveyed with this technology. The Profiler detector uses a scintillator material and a photomultiplier tube that measures the light intensity of the scintillator material. Various detector shapes are available including a commonly used slit 1 long by 1/8 wide. A palm-sized computer records and displays the data in graph form with equivalent steel thickness in the Y-axis and time in the X-axis. In order to convert the detected radiation intensity into an equivalent steel thickness, a calibration procedure must be followed. This procedure must be performed with a pipe of similar dimensions and filled with fluid at the same level as the actual field pipe. Changes in the water level will affect the baseline values. Because the data is stored as thickness versus time, it requires that the operator record the position of the probe independently in order to relate the archived values to piping locations. One method to establish the probe position is to survey the segment at a constant pace and record the time lapsed with a stopwatch. Low-frequency Electromagnetic Method Emerging technologies for the assessment of FAC of bare piping include the low frequency electromagnetic method 13. This recently developed device allows the operator to scan a wide band of accessible and hard to reach piping by running a lightweight, hand held sensor unit and acquiring the data on a laptop computer (Figure 4).

Figure 4. Low Frequency Electromagnetic Method Testex PS 2000 13

Because the device transmits and receives electromagnetic signals, it does not require liquid couplant or the removal of the coating, and needs minimal surface preparation. The device inspects the pipe along a wide band by deploying an array of sensors (up to 64). The amplitude and phase information of the electromagnetic signal is displayed graphically in real time by the laptop computer. These anomalies are then correlated to flaw depth via calibration. When operating the unit, it is recommended that the scanner be moved at constant speed over a piping segment. This allows the identification of flaw locations from the recorded data, as the unit does not currently have a wheel encoder to establish the flaw positions on the computer screen. Once the flaws are identified and ranked, verification with a more accurate wall loss measurement technique is recommended. Corrosion Rate Assessment in Bare Piping Corrosion rate estimations are an important part of the FAC management strategy since the corrosion rates affect the operational acceptance of the component, and influence the system repair plans by determining the component life estimates. A component is judged to be suitable for continued service if the predicted wall thickness at the time of the next planned inspection is greater than the minimum accepted value 4, in accordance with the formula, Wall predicted > = wall (Corrosion Rate x Time-next-inspection x Safety Factor) Likewise, a component is considered to have a given remaining useful life in accordance with the formula, Life = (Wall Minimum acceptable thickness) / (Corrosion Rate x Safety Factor)

10

EPRI recommends that the corrosion rate of a component be determined by establishing an inspection grid on the component, comparing the thickness measurement at each grid point with that obtained in the previous inspection, and selecting as the corrosion rate the maximum value obtained from all the points in accordance with the formula,
Corrosion Rate = Max [(Wall Wall previous inspection) / (time between inspections)]

The suggested grid patterns for components of various diameters are given in Table 3.
Table 3 Maximum Grid Sizes for Standard Pipe Sizes 5
Pipe Size inch 2 3 4 6 8 10 12 14 16 18 20 24 >24 Outside Diameter inch 2.375 3.5 4.5 6.625 8.625 10.75 12.75 14 16 18 20 24 Maximum Grid Size inch 1.00 1.00 1.17 1.73 2.25 2.81 3.33 3.67 4.19 4.71 5.23 6.00 6.00

Errors in the corrosion rate estimation can be introduced either by Changes in the equipment calibration between inspections, or Variations in the probe placement between surveys.

In order to reduce the errors in the equipment calibration, plants typically prefer to inspect the components when the equipment is shut down and the components are at room temperature. This procedure avoids the errors associated with surveying hot surfaces with ultrasonics, as discussed above. To minimize the errors introduced by variations in probe placement, EPRI funded the development of the AutoGrid system. This technology will be discussed next. AutoGrid The AutoGrid system 14 combines the accuracy of ultrasonics with a novel low-cost method of recording the survey locations. This system, developed by Southwest Research Institute under funding from EPRI, accurately locates the ultrasonic transducer in three-dimensional space by triangulation using an array of acoustic microphones. Using this system the operator can accurately return to a surveyed location and compare the new measurement with one previously acquired (Figure 5).

11

Figure 5. The AutoGrid System 14

AutoGrid locates the position by attaching two acoustic emitters to the transducer. These airborne sound signals are received by 3 microphones, which compute the transducer location by triangulation. The system requires that 3 permanent reference locations be established on the pipe. Then, a grid pattern is constructed for the area of inspection by the computer relative to these points. The grid points are virtual; that is, their coordinates are stored in a laptop computer rather than having them painted on the pipe surface. This avoids the difficulty of laying out the grid pattern and having the marks subsequently wear off. In addition, the system helps the operator find the marked location by continuously displaying the position of the probe in relation to the target point. Thickness measurements obtained with the help of AutoGrid can be organized as part of a comprehensive data management system such as CHECWORKS, 8. The data can be archived, annotated, displayed, and exported to CHECWORKS for further analysis. Global Detection of Metal Loss in Bare Piping Global external inspection techniques include longitudinal guided waves, tangential guided waves, and active thermography. These technologies are still under development for application to FAC and, therefore, have been classified as emerging. Two types of longitudinal guided wave technologies are currently under development using magnetostrictive and piezoelectric sensors. Magnetostrictive Guided Wave The magnetostrictive guided wave system is currently suited to scan large sections of piping from one location. The system can detect defects with an equivalent cross-sectional reduction of at least 1% and place the defect location with an accuracy of 2 inches 15.

12

The magnetostrictive guided wave system is reported to achieve its largest inspection range in straight, drained pipe, where it can inspect sections in excess of 100 feet long. The inspection range is reduced when the piping is filled with water, when the inspected section includes sharp bends, tees, and elbows, or when the coating is of the bituminous type. The current tool performs best when surveying piping of 16-inch diameter or less. The inspection range deteriorates for larger diameter piping. The technology measures the net reduction in cross-sectional area. Accordingly, defects with different morphologies, but same net cross sectional reduction, have equivalent signals. That is, localized deep pits and widespread shallow defects with the same loss in cross-sectional area are ranked equally in severity. Furthermore, the signals acquired have a spatial resolution of 7 inches. Defects that are within 7 inches are lumped together and reported as one large defect with a cross-sectional area reduction equal to the sum of the individual contributions. In addition, areas that are severely corroded will provide a large number of return echoes that will be superimposed, making it difficult to perform accurate sizing analysis. In these situations the technology provides a 3 tier grading system namely, good condition, lightly corroded, or heavily corroded. Because of these performance characteristics, once defects are detected and located, follow up measurements should be performed with more accurate local metal loss evaluation tools. The magnetostrictive system deploys lamb-wave transmitter and receiver units that consist of encircling coils and biasing magnets. The lamb wave is generated in the pipewall by disturbing the bias magnetic field established by the magnets with a current pulse applied to the encircling transmitter coil (Joule Effect). Conversely, the defect signals are detected by monitoring the current that is generated in the receiver coil as the lamb wave crosses its bias magnetic field (Villary Effect) (Figure 6).

Figure 6. Magnetostrictive Guided Wave Method Southwest Research System 15

Piezoelectric Guided Wave Guided wave inspection technologies that deploy piezoelectric transducers are also under development 16. One of these technologies uses either fully or partially encircling comb

13

transducers (Figure 7). This technology has been reported to detect defects with an equivalent cross-section reduction of 0.5%, and has been capable of inspecting 24-inch diameter pipelines.

Figure 7. Piezoelectric Guided Wave Inspection Pennsylvania State University Comb Transducer System 16

The comb transducer is reported to allow the control of the Lamb wave modes generated. This mode control permits the optimization of defect detection as well as minimizes the loss of energy to the transported fluid or the external coating, hence, maximizing the range of inspection. The inspection procedures, however, are still under development, and tool operation requires supervision from an expert in the process. The comb transducer consists of an array of transducers placed normal to the pipe surface. Each transducer in the array produces a periodical vibration in phase and at the same frequency as to generate a guided wave with a wavelength equal to the transducer spacing. This relationship between the transducer spacing and the wave mode permits the optimization of the inspection range. Tangential Guided Waves Global ultrasonic inspection can also be performed with tangential guided waves. This emerging technology is suited for evaluation of hard-to-reach locations such as piping laying on saddle supports, or as a rapid scanning device to qualify piping areas for further evaluation with more accurate metal loss techniques. The system allows 360 examination of piping ranging from 2 to 36 in diameter. Surface preparation is minimal since the system deploys electromagnetic acoustic transducers (EMAT) that require no couplant, and can perform surveys through coatings up to 0.02-inch thick. The device uses a saddle-like, hand-held sensor unit that is placed on top of the pipe with transmitter and receiver electromagnetic acoustic transducers (EMAT) located 90 apart (Figure 8). The transmitted wave reaches the receiver along the short cord (90), by taking the long path (270), or by performing a full trip around the circumference before reaching the transducer

14

(450). The time-of-flight of these signals are recorded and correlated with the volumetric metal loss along each of the travel paths. In this way, the technology surveys a band 1.5-inch wide that completely encircles the pipe. The operator then moves the sensor unit to attain 100% coverage 17 .

Figure 8. Tangential Guided Wave Method Magna Scan System 17

The system can detect a 10% volume loss over the surveyed band; however, it provides no information on the type of corrosion damage. Therefore, the corrosion indications are typically marked for further evaluation with more accurate metal loss techniques. Active Infrared Thermography Active Infrared Thermography is another emerging technology under development for global inspection of FAC 18. This technology may be suited for detection of defects with diameter equal to the wall thickness and 25% deep. The performance deteriorates in the presence of thick coating. Active Infrared Thermography detects thin areas by quickly heating the pipes external surface with pulsed flash lamps then recording the change in the surface temperature with an infrared camera. With this procedure, internal pipe defects are identified by contrasting their hotter appearance from the cooler image of the pipe at large. The cameras are designed with a small form factor that can be manipulated by a single operator in tight environments. Furthermore, the system highlights defect indications automatically, for further comparison with calibration signatures.

15

FLOW ACCELERATED CORROSION INSPECTION OF INSULATED PIPING NDE technologies for insulated piping are classified in the same way as in the bare piping section; that is, localized assessment, corrosion rate monitoring, and global surveying. Local Detection of Flow Accelerated Corrosion in Insulated Piping Local detection techniques include radiography, real time radiography, pulsed eddy current, and Compton backscattering. Radiography Internal wall loss assessment through insulation using radiography is well established. Internal wall loss assessments with an accuracy of +/- 5% can be obtained using the tangential configuration as long as the geometrical unsharpness remains within 0.02 inch 6,9. Since the film and the source (by necessity) are placed outside the insulation, the source-toobject distance must be increased relative to bare piping procedures to attain the geometrical unsharpness value of 0.02 inch. Therefore, measurements through insulation require longer exposure times to obtain the recommended exposure densities of 1.5 to 2. EPRI recommends the use of fine grain high-resolution film for wall thickness assessments 9. However, if this technique leads to exposure times that are impractical, then faster speed films can be used. As was discussed in the bare piping case, double wall exposures can be used to locate the damage and assess the aerial extent of the affected area, but actual verification of the remaining wall thickness should be done with other techniques. Real-time Radiography FAC can also be detected through insulation with real-time radiography for the commonly deployed low-density insulation materials such as urethane or polystyrene foams. The detection capabilities of the wide-band ThruVu system or the narrow-band Lixi Profiler are little affected by the presence of these types of insulation systems. The advantage of these newly developed techniques is that the cost per unit area of inspection is much lower when compared with conventional radiography. However, because these tools use the double-wall assessment technique, the thickness values provided must be verified with more accurate metal loss procedures. Higher density insulation systems such as calcium silicate or asbestos have been reported to interfere with the ThruVu system. When pieces of these types of insulation are missing, an apparent wall loss is recorded in the image 19. Pulsed Eddy Current FAC can also be detected with pulsed eddy current technology. The wall thickness of pipes with diameters greater than 2 inches can be measured with an accuracy of +/- 0.04 inch 20,21.

16

Pulsed eddy current systems are easy to deploy. Since no surface preparation or coupling media is required, the operator simply places the probe at the insulated location and, within a few seconds, obtains a thickness reading (Figure 9). Furthermore, since the technique is not sensitive to probe alignment, it is very tolerant of the operators skill, which helps attain good measurement repeatability in the field.

Figure 9. Pulsed Eddy Current Method The Incotest System 21

Pulsed eddy current technology is not suited for the detection of isolated damage. The eddy current field, when used through 2 inches of insulation, illuminates an area 4 to 8 inches in diameter. Isolated pits may go undetected when applying the technique. Pulsed eddy current uses a transmitter and a receiver coil. During operation, a train of current pulses is sent to the transmitter coil, and the voltage decay of the induced eddy current field is monitored by the receiver. The remaining wall thickness is determined by measuring the decay time (or decay rate) of the eddy current field and comparing it to calibration standards. Use of the technology does require some survey planning, particularly in pipe racks or in the vicinity of carbon steel vessels. The proximity of large carbon steel masses will affect the measurements and may require the realignment of the coil sensor away from the interfering mass. Compton Backscattering Compton backscattering is another technology currently under development for measurement of pipe wall thickness through insulation.

17

This process has been shown to have an accuracy of 5% for wall thickness less than 0.25 and 10% for thickness greater than 0.25 but less than 0.5. For thickness greater than 0.5 the performance of the system degrades. These accuracy targets are achieved when the material, the internal fluids, and the survey geometry are well-characterized 22. This technology has the advantage of exhibiting a spatial resolution of 0.75 and 1 inch for liftoff distances of 2 and 4 inches respectively. This resolution enables the technique to detect pits with 0.5-inch diameter and quantify their depth when their diameter exceeds 1 inch in 4-inch thick insulated piping. The Compton backscattering device includes a high-energy radioactive source with photon energies greater that 0.1 MeV. The detectors, which can be of the scintillation type, are placed near the entry point of the source radiation to capture the very weak backscattered radiation emanating from the tested material. When the location of the source and the detector relative to the specimen are accurately established, the intensity of the scattered radiation is proportional to the density and thickness of the material. Since the density of steel is known precisely, the pipe wall thickness can be accurately determined 23. The intensity of this backscattered radiation also depends on insulation thickness, and service fluid. Since insulation height is typically variable and unknown, practical implementation of this emerging technology will require the development of liftoff corrections to the detected signal. Corrosion Rate Assessments through Insulation The NDE procedures to establish corrosion rate through insulation in piping susceptible to FAC are not yet established. However, the pulsed eddy current and Compton backscattering techniques can potentially be used for this service. Pulsed Eddy Current The pulsed eddy current technology can potentially be used to monitor the rate of FAC online, in insulated piping. Because the monitoring procedure has not been validated, the application has been classified as emerging. The pulsed eddy current technique is suited for corrosion rate monitoring because it exhibits a repeatability of +/- 0.005-inch, which is equivalent to that exhibited by ultrasonics. Furthermore, monitoring can be performed online, while the component is hot, without deteriorating the tools repeatability. As was mentioned above, the device is very tolerant of variations in the field conditions. No surface preparation is required, and changes in the probes alignment will not affect the performance. In addition, because the tools hardware consists of a voltage measurement unit, an accurate clock, and coil probes, no field adjustment of the device is required. Therefore, the operators skill level is not critical to achieve the repeatability target. As in the case of bare piping, a monitoring grid pattern should to be established on the component. This grid pattern can be painted on the insulation, rather than on the surface of the component. Also, the pulsed eddy current device, in principle, can be integrated with the AutoGrid system, although this service is not yet available.

18

Because the pulsed eddy current system illuminates a relatively large area, about 4 inches in diameter for a 2-inch liftoff, less monitoring stations are required when compared to ultrasonics to attain a similar aerial coverage of the component. Accordingly, the grid pattern can be made sparser than that suggested in Table 3. Compton Backscattering Corrosion rates can also potentially be monitored with the Compton backscattering technique for piping with wall thickness less than 0.5. This emerging technique is still under development, as its repeatability has not yet been established. However, since the technology requires that a calibration curve of radiation-countsversus-wall thickness be established before performing the survey 22, its repeatability is likely to be less robust than that exhibited by the pulsed eddy current technique. However, because the Compton backscattering technique has a good spatial resolution, and the errors brought about by the liftoff variations are eliminated once the baseline values at grid points are established, the technique represents an alternative for online monitoring of FAC. As with pulsed eddy current devices, the grid pattern can be established on the insulation rather than the component itself, either by painting the survey stations or by integrating the device with the AutoGrid system. Because the spatial resolution is 0.75 inch, the grid spacing suggested in Table 3 is applicable. Global External Inspection of Insulated Piping Global external inspection techniques for insulated piping include longitudinal guided waves with either magnetostrictive or piezoelectric sensors. As mentioned when referring to bare piping inspection, these are emerging technologies. Magnetostrictive Guided Wave The magnetostrictive guided wave systems can have an inspection range greater than 100 feet as long as the sensors are placed directly on the pipe surface by removing a short piece of insulation. When placing the coils and magnets in close contact with the pipe surface, the system can detect defects with a cross-sectional reduction of 1%, with spatial resolution of 7 inches, and locate defects with an accuracy of +/- 2 inches. As mentioned before, the tool performs best when surveying piping of 16-inch diameter or less. The inspection range deteriorates for largerdiameter piping. The magnetostrictive guided wave system can be operated with the transmitter and receiver coils placed outside the insulation. However, this configuration reduces the inspection range and deteriorates the systems performance. The preferred manner is to remove a narrow band of insulation all the way around the pipe and place the sensors directly on the pipes external surface. Guided waves are then produced which travel along the length of the pipe (beneath the insulation) and detect damage.

19

Piezoelectric Guided Wave Guided wave inspection technologies that deploy piezoelectric transducers can also be used to inspect insulated piping, but they require removal of the insulation at the location where the transducers are to be coupled to the pipe. These systems have been reported to inspect insulated piping with minor range deterioration once the wave-mode is properly selected 13. This technology requires that the partially encircling comb be coupled to the pipe surface at the location where lamb waves are to be launched. This guided wave technology has been reported to detect defects with an equivalent cross-sectional reduction of 0.5%, in piping up to 24 inches in diameter 16. Conclusions This evaluation on new and emerging technologies for FAC concluded that: Monitoring the rate of FAC in bare piping has been made more cost effective with the advent of AutoGrid, a new technology. FAC detection through insulation is feasible, with some limitations, and the availability of new technologies (real-time radiography and pulsed eddy current) has made the inspection operations more cost effective. Measuring the rate of FAC through insulation is feasible and the improvements in the pulsed eddy current and Compton backscattering techniques may make the in-service assessments more plausible. Methods for detection of FAC globally are being developed but are not yet established.

References 1. Chexal, B., et.al., Flow Accelerated Corrosion in Power Plants, EPRI Report No. TR106611-R1, 1998. 2. Dooley, B., Condition Assessment Guidelines for Fossil Fuel Power Plant Components, EPRI Report No. GS-6724, 1990. 3. Gosselin, S.R., Ammirato F.V., Walker, S.M., Risk-Informed Inservice Inspection Evaluation Procedure, EPRI Report No. TR-106706, 1996. 4. Chexal, V.K., Recommendations for an Effective Flow Accelerated Corrosion Program, EPRI Report No. NSAC-202L-R1, 1996. 5. Dooley, R.B., et.al., Guidelines for Controlling Flow-Accelerated Corrosion in Fossil Plants, EPRI Report No. TR-108859, 1997. 6. Walker, S.M., Ammirato, F.V., Nondestructive Evaluation of Ferritic Piping for ErosionCorrosion, EPRI Project No. NP-5410, 1987. 7. Nottingham, L.D., Sherlock, T.P., NDE Guidelines for Fossil Power Plants, EPRI Report TR-108450, 1997. 8. Chexal, B., CHECWORKS Flow Accelerated Corrosion, EPRI Report No. TR-103198P1, 1998. 9. Becker, F.L., Lube, B.M., Walker, S.M., Guide for the Examination of Service Water Piping, EPRI Report No. TR-102063, 1994.

20

10. ASTM Standard Recommended Practice for Measuring Thickness by Manual Ultrasonic Pulse-echo Contact Method, Designation E-797-90. 11. Gothard, M., Field Trials and Testing of Prototype ThruVu Real Time Radiographic Device, EPRI Report GC-110152-SI, 1998. 12. Walker, S.M., Demonstration of the Prototype Lixi Portable Density Profiler, EPRI Report GC-109055, September 19, 1997. 13. Ramchandran, S, Ramchandran, S., McDougal, L., Non-Destructive Testing of Boiler Tube from the Fireside (Outer Diameter), Proceedings: Third International Conference on Boiler Tube Failures in Fossil Plants, EPRI Report TR-109938, Pg. 5-33, April 1998. 14. Walker, S.M., Development and Commercialization of Next Generation NDE Devices for Flow-Accelerated Corrosion, EPRI Report MI-107959, 1998. 15. Brophy, J.W., Assessment of Magnetostrictive Sensor Technique: Detecting FlowAccelerated Corrosion in Feedwater Piping, Revision 1, EPRI Report No. TR-108449-R1, 1997. 16. Spanner, J.C., Ultrasonic Guided Wave Inspection of Piping, EPRI Report TR-107419, November 1997. 17. Kenefick, S., Evaluation of the Quality NDT Magna Scan Ultrasonic Examination System, EPRI Piping and Bolting Memorandum, May 1997. 18. Zayicek, P., Shepard, S.M., Summary Report of Advanced IR NDE of Service Water Piping Systems, EPRI Report TR-107463, October 1997. 19. Angell, B., Personal Communication, May 1999. 20. Walker, S.M., Martinez, E.Q., MacDonald, D.E., Evaluation of the Transient Electromagnetic Probing (TEMP) System Through for Detection of Wall Thinning Through Insulation, EPRI Report TR-101680, December 1992. 21. Brett, C.R., Assessment of the Pulsed Eddy Current Technique: Detecting Flow-Accelerated Corrosion in Feedwater Piping, EPRI Report No. TR-109146, December 1997. 22. Klevans, E.H., et.al., An Erosion-Corrosion Monitor Using Gamma Ray Backscatter, EPRI 5th Piping and Bolting Inspection Conference, San Antonio, June 1999. 23. Bryant, L.E., Nondestructive Testing Handbook Radiography and Radiation Testing, American Society of Nondestructive Testing, Volume 3, 1985.

21

About EPRI EPRI creates science and technology solutions for the global energy and energy services industry. U.S. electric utilities established the Electric Power Research Institute in 1973 as a nonprofit research consortium for the benefit of utility members, their customers, and society. Now known simply as EPRI, the company provides a wide range of innovative products and services to more than 1000 energy-related organizations in 40 countries. EPRIs multidisciplinary team of scientists and engineers draws on a worldwide network of technical and business expertise to help solve todays toughest energy and environmental problems. EPRI. Powering Progress

1999 Electric Power Research Institute (EPRI), Inc. All rights reserved. Electric Power Research Institute and EPRI are registered service marks of the Electric Power Research Institute, Inc. EPRI. POWERING PROGRESS is a service mark of the Electric Power Research Institute, Inc. TP-114349 Printed on recycled paper in the United States of America EPRI 3412 Hillview Avenue, Palo Alto, California 94304 PO Box 10412, Palo Alto, California 94303 USA 800.313.3774 650.855.2121 askepri@epri.com www.epri.com EPRI NDE Center 1300 WT Harris Blvd., Charlotte, North Carolina 28262 PO Box 217097, Charlotte, North Carolina 28221 USA 704.547.6100