7.03.2017

A method of measuring minimum wall thickness for pipes and plates in inaccessible places

by Dr Konstantin E Nikitin, R&D Manager of Dacon Inspection Services Company Limited.

Method

Dacon Inspection Services has developed a unique system for detecting corrosion in pipelines at hidden support areas and other inaccessible places. The Corrosion Under Support (CUS) system is able to quantify corrosion and measure the remaining wall thickness at inaccessible areas, such as where the pipe is resting on supports or held by clamps.

The system has an independently verified accuracy of +/- 0.3 mm and can be used to gain significant cost advantages when assessing pipeline and process piping integrity, and planning plant shutdowns, as well as scheduled maintenance. Tried and tested in the field by a number of the world’s largest oil and gas companies, this system should form an important part of any inspection program.

Previously available tools have been able to quantify corrosion under support areas however; Dacon’s CUS system is capable of measuring the actual remaining wall thickness. This is carried out by utilizing the direct measurement time of flight technique, as used for traditional Ultrasonic thickness measurements. After testing and verification and a number of years in commercial use, Dacon CUS has proven to be a valid tool to precisely assess wall loss at pipe support areas, thus minimizing the time and expense involved in lifting up a pipe.

The inspection method is based on L-S-L conversion where a primary longitudinal wave (L1) is converted on the surface into a shear wave (S), and then back to a secondary longitudinal wave (L2). As a result, the signal is comprised of a combination of shear waves and longitudinal waves which freely propagate along the pipe wall. This is a continuous process that takes place in each section of the pipe as shown by the below simulation.

Software

The CUS software is based on measuring time of flight between primary and secondary longitudinal (L) waves and recalculating it in to the remaining wall thickness.

The wall thickness (h) is related with the time of flight of the sound wave (T) through the following simple equation:

Where k is the coefficient of proportionality defined speeds of sound; cl and ct for longitudinal and shear waves, the following relationship applies:

Therefore, when corrosion is present resulting in a smaller wall thickness; the combined L-S-L wave has a shorter time of flight than of that in an area of good material.

By measuring the time of flight for a signal generated at an area presenting corrosion, which is then observed on an A-scan as a small pulse, before the secondary longitudinal wave (L2), we can measure the remaining wall thickness of an inaccessible area of the pipe between the transducers.

Since both surfaces take part in L-S-L wave conversion, the remaining wall thickness is the same for both external and internal defects with the same wall thickness, and so it is not possible to separate external with internal defects using this technique.

This method of measuring remaining wall thickness is applied for general corrosion types with more or less even and parallel surfaces. It will not present accurate results for pitting or other type of corrosion with an average dimension of less than the diameter of the chosen transducers.

Case study

Below are shown the experimental results for sample of 14 inch pipe with wall thickness 11 mm with the following artificial external cylindrical defects simulating corrosion defects:

• Defect 1 with diameter 20 mm and depth 4 mm simulating average level corrosion with 36% metal loss
• Defect 2 with diameter 20 mm and depth 6 mm, simulating high level corrosion with 55% metal loss

Signals on good material

Signals at defect 1

Signals at defect 2

The time of flight on the above oscillograms for the good material and defects 1 and 2 is given as: 2.8 us, 1.9 us, and 1.3 us. For these values the above formula gives the wall thicknesses at: 10.7 mm, 7.2 mm, and 4.9 mm. This demonstrates a high level of accuracy when compared to the real values of wall thickness: 11 mm, 7 mm and 5 mm.

ปิดโหมดสีเทา