Printer Friendly

Gas-coupled ultrasonic inspection of pipelines.

Magnetic Flux Leakage (MFL) pipe inspection devices ("pigs") measure wall loss in a gas pipeline, but not the remaining wall thickness that determines the remaining strength. In addition, the precision of MFL tools is limited to 10% of the wall thickness, and the technology cannot find cracks, including stress corrosion cracking. For pipeline operators, the 10% limit means extra digs are required to ensure that all severe corrosion has been found. Another drawback is that this inspection technique offers little ability to monitor corrosion growth rates to determine where mitigation is effective and where it needs to be improved.

Measuring pipe wall loss means that the pipeline operator must base repairs on the minimum expected wall thickness rather than the actual measured amount of remaining metal in the pipe wall. Ultrasonic inspection has found cracks and measures the remaining wall with a precision of a few percent. However, ultrasonic inspection currently squires putting a liquid couplant in a gas pipeline. Transducers specialized for inspection in high-pressure gas and specialized inspection methods can eliminate the need for a liquid couplant, bringing the advantages of ultrasonic inspection to gas pipelines.

Ultrasonic inspection is the most accurate method for measuring pipeline flaws. It is also the only method that has successfully detected and measured cracks. Ultrasonic inspection is accurate enough that it can be used directly in calculations of remaining pipe wall strength whether by finite element analysis on by on of the available remaining strength codes, such as ASME/ANSI B31G or RSTRENG.

The most common method used to inspect pipelines is MFL. The technique has been used since the 1960s and is widely accepted as a cost-effective method for maintaining pipeline safety. Yet, MFL has a number of limitations. Chief among these is its accuracy, Current precision is limited to measuring within +/- 10% of the wall thickness 80% of the time.

For example, if MFL estimates that the remaining wall in a corroded area is 50% of the nominal pipe wall thickness, it means that 80% of the time the actual amount of remaining metal is between 40% and 60% of the nominal pipe wall thickness.

Ten percent of the time, less than 10% of the nominal wall thickness is left, and 10% of the time, more than 60% of the wall thickness remains. With that kind of imprecision, confirmatory digs are required--one common criterion being to dig until two successive in-the-ditch measurements find pipe that does not need repair.

MFI, also cannot find cracks and is not good at finding longitudinal defects With MFL technology, signal strength varies with the amount of metal lost MFL, therefore, does not measure the the amount of metal remaining, which is the crucial information. To determine remaining metal, the amount of metal loss is subtracted from the nominal wall thickness. In contrast, ultrasonic inspection measures the remaining wall thickness directly.

In spite of many millions of dollars of R&D for developing a gas pipeline crack detection and measurement pig, none have worked acceptably for routine inspection, The only successful crack detection pig uses standard pitch/catch-mode ultrasonics to detect the crack tip. Gas-coupled ultrasonics would also enable that technology to be used for gas pipeline inspection.

Ultrasonic inspection works well for liquid-product pipelines. Unfortunately, standard transducers cannot be used to inspect gas pipelines because standard transducers need a liquid or gel couplant to get enough of a return signal to measure the pipe wall thickness. However, improvements in transducer technology should provide a high enough signal to noise ratio to make wall thickness measurements in high pressure gas. The Gas Technology Institute plans to develop such transducers and use them to measure remaining wall thickness and crack depths in gas transmission pipelines.

Theory of Operation

Ultrasonic transducers measure the thickness of a material by measuring the elapsed time between the front wall and back-wall reflections. To get the thickness, engineers multiply the elapsed time by the speed of sound in the material. Crack depths are measured by the elapsed time between the front-wall reflection and the reflection from the crack tip.

There are two methods for sending out pulses and detecting the reflections. In the pulse/echo method, illustrated in Figure 1, a single transducer sends out the pulse and measures the time between the reflections. In the pitch/catch technique, one transducer sends out the pulses while a second detects the reflections. Either can be used to measure wall thickness The pitch/catch method is used for crack detection and measurement.

[FIGURE 1 OMITTED]

The major problem in using ultrasound inspection in gases without a liquid couplant is that the back wall reflection is so small that it cannot be detected with standard ultrasonic transducers. The sound properties between natural gas and steel are so mismatched that almost all of the ultrasound is reflected from the front wall of the pipe and very little is transmitted into the pipe wall itself. The situation improves dramatically as the gas pressure is increased to the point where, with custom-designed transducers, it should be possible to make ultrasonic measurements at several hundred psi.

Two transducer modifications are needed. The transducer must be designed to increase its sensitivity to ultrasound waves by matching its properties to those of sound in a gas, and it must have a very high signal to noise ratio (SNR). We calculated that a SNR of 120 dB for the transducer plus the electronics is required.

Objectives

The objectives of the GTI Gas Coupled Ultrasonics project are to develop gas coupled ultrasonics as a very accurate, commercially available, gas pipeline inspection method for measuring the extent and depth of corrosion and remaining wall thickness, and with further effort to adapt the technique for detecting and measuring the depths of cracks and depth and extent of crack colonies.

GTI is conducting research to achieve objectives in three stages:

In the first stage, use gas-coupled ultrasonic sensors in conjunction with MFL pigs to measure the full wall thickness of the pipe. This calibrates the MFL pig and provides the actual (instead of the nominal) pipe wall thickness for calculating remaining strength. This also provides valuable operational experience for gas-coupled ultrasonics at little risk to pipeline inspection results.

In stage two, measure the remaining wall thickness in corroded areas. A gas-coupled ultrasonic inspection pig providing the precision of ultrasonic inspection can then be built.

In the third stage, detect cracks and measure crack depths using gas-coupled ultrasonics. A crack detection and sizing pig can then be built.

Ultrasonic Transducer Development

In previous research, standard transducers with low-noise electronics showed that defect and crack detection and measurement are possible, but that much better transducers were needed to greatly reduce the noise and improve reliability We therefore stalled a program of looking for appropriate transducers (ones that could ring down by 80dB in 2 [micro]seconds) and developing electronics with a signal to noise ratio for that transducer of 120 dB.

The first transducers manufactured under this program still showed clutter signals from spurious radial resonances and side lobe responses. It was clear that better modeling capability was required. GTI worked with Weidlinger Associates (a company that developed the PZ-Flex software package that is in common use in transducer development for the medical industry) to model the transducer and develop design specifications for the Phase II prototype.

Our calculations showed that eliminating the spurious response required a reduction in kerf width from .002-inch to .001-inch. Prototypes based on that design were better than the original transducers used in the project, but they have not yet met specifications We are making improvement. Meanwhile, GTI continued its search for other transducer sources

One promising lead developed through a paragraph on a medical application of gas-coupled ultrasonics in Physics Today. Ultran Laboratories had developed transducers used to monitor burn patients. The transducers evaluated the severity of burns and how well they were healing without the need to touch the patient, which is very painful, These transducers are also in use for inspecting composite materials, coupling ultrasound directly through air into the composites rather than needing a liquid couplant. These transducers have proven to be the best transducers so far; but they are expensive and also still fall short of specifications.

Figure 2 shows the back wall reflections detected by our best Ultras transducer from a 9 mm thick pipeline steel plate The initial huge pulse is the reflection from the front of the 9 mm plate. The smaller pulses are a series of "back wall" reflections from ultrasonic waves trapped in the plate. Just as it is difficult to get ultrasound into the plate using gas as a couplant once in the plate the sound cannot leave and keeps bouncing between the front wall and the back wall.

[FIGURE 2 OMITTED]

Current Status

The Ultran transducer was the quietest and most sensitive of the transducers tested. The signal to noise ratio is better at 500 psig than that of most sensors at 1,000 psig. The pulses are a little wider than desired for good timing measurements and the first back-wall reflection is partly buried in the front wall reflection. Still, this sensor and a second Ultran transducer were, overall, by far the best two sensors that have been tested to date, and they are more than adequate for proof of feasibility in a pipeline, provided the sensors can withstand the pipeline environment. Back-wall reflections from both sensors were still well above noise levels at pressures less than 300 psig.

Encouraged by these results, GTI tried unsuccessfully to measure the remaining wall thickness in a coupon of corroded curved pipe with rust on both sides. We detected nothing but noise where the back wall reflections should have been.

Fortunately, researchers do not live in a vacuum. On the expert advice of George Alers (who worked on this project while at the National Institute of Standards and Technology) that the pitch/catch technique was more robust, GTI tried that method on a larger coupon of 24-inch pipe that also had corrosion on the outside, but was smooth on the inside. Remaining wall thickness was measurable in most of the corroded area of the pipe using unfocused transducers, although interpretation of the signals in many areas was not straightforward. Focused transducers should improve the results since one of the problems was that the beams were of the same dimensions as the corrosion pits and therefore the signals combine reflections from too wide a range of depths.

We realized shortly after getting these good results that the first of the objectives--using gas coupled ultrasonic inspection for calibrating MFL pigs--was now well within reach. We need to measure the tolerance of the transducers to small mechanical alignment errors to design a pig-mounting device, and transducers need to he more rugged for use in a pipeline. We are developing more robust versions of our transducers and have had favorable discussions concerning mounting these on an MFL pig. Barring unforeseen obstacles, GTI expects to test gas-coupled ultrasonics in an operating pipeline later this year or early next year.

ACKNOWLEDGEMENTS

The author thanks the Gas Technology Institute (GTI) and the Pipeline Research Council International (PRCI) for funding the project, and the GTI/PRCI Corrosion and Inspection Technical Committee for its support and advice The author would like to acknowledge the following individuals for their important contributions to the development of gas-coupled ultrasonics for gas pipeline inspection: the late Christopher Fortunko, NIST; Jerry Jackson, Southwest Research Institute (retired); and Alfred E. Crouch. Southwest Research Institute

Author: Albert Teitsma is Program Manager in Pipeline Operations Technology at GTI. He received his B.Sc. in Physics and Ph.D. in Nuclear Physics from McMaster University. His contributions include developing new technologies for nondestructive inspection of transmission and distribution pipe. Before joining GTI. Dr. Teitsma worked for Pipetronix Ltd. and TransCanada PipeLines Ltd., developing in-line inspection techniques.
COPYRIGHT 2003 Oildom Publishing Company of Texas, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:2003 Special PPSA Section
Author:Teitsma, Albert
Publication:Pipeline & Gas Journal
Geographic Code:1USA
Date:Aug 1, 2003
Words:1976
Previous Article:Engineers master challenges to commission large Vietnam pipeline.
Next Article:Buyer's guide.
Topics:


Related Articles
Crack detection tool set for initial U.S. trials.
British Gas uses ultrasonic vehicle for assessing pipeline integrity.
Life extension program on cross country gas trunklines.
Pipeline inspection failure blamed for oil spill. (Newsreel).
Pigs and the pipeline: Alyeska uses the Geopig to inspect the trans-Alaska oil pipeline.
An additional tool for integrity monitoring. (Optical Remote Sensing).
Pigging today is a mature, sophisticated science at last.
Major pipeline outlines its compliance plan for integrity management.
ABB continues to score Caspian Sea contracts.
GE, AGR in pipeline partnership.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters