Plant Integrity Limited Granta Park, Great Abington, Cambridge CB21 6GP, United Kingdom Telephone +44 (0)1223 893994 Telefax +44 (0)1223 893944 e-mail pi@twi.co.uk Web: www.plantintegrity.com

Long Range Guided Wave

Ultrasonic Testing

Pi

CONTENTS Page

Preface 1

Principles of long range guided wave ultrasonics 2

Teletest® system 5 The Teletest Focus™ Unit 5 Teletest Muti-Mode Tools 6 System overview 7 Propagating the guided waves 8 Multimode inspection 10 Collecting and displaying data - Teletest® Software 10

Appendix: 15 Teletest Focus™ 16

Comparison with other NDT methods 18 Methods 18 General 18 Access costs 21 Testing Costs 21 Cost Summary 22

Teletest® Operating Envelope 23

Evidence of Performance 29

Teletest® field applications 32 Power industry 54

Special applications 56

On-going Research and development 59

Summary 66

1

PREFACE

Long range ultrasonic testing (LRUT) is arguably the most significant development in the field of non-destructive testing to have taken place over the last two decades. It is being used increasingly, primarily in the oil and gas industries, for the inspection of pipes and pipelines in order to detect corrosion. LRUT was first introduced as a commercial technique under the trade name Teletest® by Plant Integrity Limited (Pi) - a wholly owned subsidiary of The Welding Institute (TWI) - in 1997. This booklet is designed to: • Provide a simple explanation of the technique and its underlying principles. • Outline work undertaken to validate the technique.

• Give information about the many applications of Teletest in the upstream (exploration and

production), midstream (pipelines) and downstream (process plant) sectors of the oil, gas and petrochemical industries and in the power generation industries.

LRUT has tremendous potential for further development by improving the sensitivity and productivity of current applications and also taking the technology into totally new applications. Research activities being undertaken by TWI and Pi are described in Section 10.

2

PRINCIPLES OF LONG RANGE GUIDED WAVE ULTRASONICS

Ultrasonic testing (UT) is used extensively as a non-destructive testing (NDT) technique for detecting defects in a wide range of structures and components, both during manufacture and in service. Conventional UT uses so-called bulk waves with ultrasound frequencies in the MHz range. Pulses travel along a narrow beam and echoes are detected from defects in the beam’s path. The test range is generally measured in millimetres or centimetres. Plate waves, also known as Lamb waves, can be generated at lower ultrasound frequencies (in the kHz range). They can travel in pulses that penetrate the whole plate thickness over distances of many metres. The plate must be thin enough for waves on opposite surfaces to interact. If the waves on the two surfaces are in phase, Asymmetric (A) plate waves are produced. If the waves are out of phase, Symmetric (S) plate waves are produced (Fig.1).

The velocity of plate waves may vary with frequency. When this occurs, they are described as being ‘dispersive’. Therefore, a pulse of plate waves, containing groups of different frequencies, will broaden as it travels - the lower frequency components travelling more slowly. Guided waves are a special case of plate waves travelling in a wave guide, commonly a pipe. The particle displacements are similar, but because the pipe acts as a wave-guide, the pulses can travel over even longer distances, exceeding 100m under some conditions. Figure 2 illustrates the principal wave modes that can be generated in pipe.

Longitudinal

Torsional

Flexural

Fig.2 Guided waves in pipes

Figure1 Plate waves

Direction of propagation

Asymmetrical wave

Symmetrical wave

Asymmetrical wave

Symmetrical wave

3

The variation of velocity with frequency, wave mode, pipe diameter and wall thickness is illustrated by means of dispersion curves. A simplified illustration showing four wave modes only is shown in Fig.3. These are for a specific pipe diameter and wall thickness. Other diameters and thicknesses will have their own families of dispersion curves.

It can be seen that the Torsional T(0,1) wave is non-dispersive. Its velocity is constant irrespective of frequency. The Longitudinal L(0,1) wave is highly dispersive with wide variations in velocity with frequency. The Longitudinal L(0,2) wave cannot exist at frequencies below about 20 kHz. However, at frequencies above about 40 kHz, the velocity becomes nearly constant with changing frequency. That is the wave becomes non-dispersive. There is a vast range of flexural waves. Only the dispersion curve for the Flexural F(1,3) mode is shown. This wave can only exist at frequencies above ~ 25 kHz. Like the L(0,2) wave, it becomes almost non-dispersive at frequencies above about 50 kHz. As will be seen later, dispersion curves such as those shown are used to select optimum frequencies for testing pipes of given diameters and thickness. The strategy is to use wave modes that are non-dispersive at the test frequency.

Dispersive waves are undesirable because the pulse widens as it travels along the pipe and variation in the velocity of the wave makes it difficult to determine the exact position of any reflector along the pipe.

0.00 0.02 0.04 0.06 0.08 0.100.0

2.0

4.0

6.0

Frequency (MHz)

Vgr (m/s)

Gro

up v

elo

city

0.0 100

Frequency (kHz)

T(0,1)

F(1,3)

L(0,2) L(0,1)

Fig.3 Dispersion Curves

4

Because their velocity is influenced by wall thickness, guided waves exhibit their most important characteristic for NDT; that of being sensitive to changes in wall thickness. They are therefore sensitive to corrosion or erosion, whether it is on the inside or outside surface of the pipe. They are also sensitive to cracks provided that they present a significant planar reflection transverse to the axis of the pipe. This characteristic is based on the physical phenomenon that whenever ultrasound velocity changes at a boundary, a small proportion is reflected. The effect can be caused equally by an increase in wall thickness, at a pipe girth weld for example, or a decrease in wall thickness at an area of corrosion or erosion (Fig.4)

Fig.4 Guided wave reflection at corrosion

5

TELETEST® FOCUS SYSTEM

Teletest® was the first commercially available system to utilise long-range ultrasonic testing for

detecting corrosion in pipelines. TWI developed the technology in response to requests from Industrial Members in the Oil, Gas and Petrochemical industries. These companies required an NDT method capable of detecting corrosion in pipes and pipelines at areas inaccessible for inspection by conventional NDT methods, i.e. insulated and sleeved pipework, or pipelines running under roads or elevated on pipe racks, etc. The original Teletest® equipment has been replaced by a system with greatly enhanced capability and the flaw detector is sold under the Teletest Focus™ trade mark.

The basic system consists of: • A low frequency flaw detector, the Teletest Focus™ unit. • Transducer ring or tool that wraps around the pipe. • A lap-top computer that contains the software for controlling the system • Cable connector between Teletest Focus™ unit and tool. • Umbilical between Teletest Focus™ unit and laptop.

The Teletest Focus™ Unit

The Teletest Focus™ Mark 3 unit (Fig.5) contains the electronics to operate the transducers in sequence in accordance with the inspector's input. The received signals are converted into digital data that can be processed and recorded by computer. The unit is powered by an internal Li-ion battery with sufficient capacity to allow inspections to be completed at more than 25 locations. The distance between the unit and the tool is kept to a minimum to reduce outside signal interference, but the digital data can be transmitted via communication cable up to 100m to the computer. The unit controls and collects 24-channels of ultrasonic data. It is designed for: - • Longitudinal operation • Torsional operation • Multi-Mode operation • Focusing operation

Fig.5 Teletest® Focus unit

6

Teletest Muti-Mode Tools

Teletest® has been designed as a ‘modular’ system. Tools are built up using the appropriate number of modules (for example 36 for testing 12" pipe). Each module has five individual transducer elements as shown in Fig.6. Three are orientated parallel to the pipe’s axis to generate longitudinal waves and two are orientated circumferentially to generate torsional waves. This gives Teletest® ‘Multi Mode’ capability. This is an important feature of Teletest® because for some pipes and flaw types longitudinal testing is most sensitive, whereas, in other situations, torsional testing is more satisfactory. The Multi Mode capability enables the optimum wave mode to be adopted. This is discussed further in a section dealing with Multi Mode inspection.

The modules are forced into contact with the pipe by means of a lightweight inflatable collar Figure 7 shows a collar that has been populated with multi-mode modules. The collar is assembled around the pipe by a rapid clamping mechanism shown in Fig.7. Modules can be quickly removed from the collars and re-fixed in another collar for use on a different pipe diameter. Unlike conventional UT, a liquid couplant between transducer and pipe surface is not applied. There merely needs to be sufficient, evenly distributed pressure of the transducer on the test surface. This is achieved by the air pressure which is supplied and controlled by an ‘on-board’ pump, built into the Teletest Focus™ unit.

A user needs sufficient multi-mode modules for the largest diameter pipe of interest. The appropriate

number can then be selected for any pipe of smaller diameter.

Fig.6 Transducer multi-mode module

Fig.7 Carbon Fibre Collar populated with multi-mode modules (left) and rapid clamp

mechanism (right)

7

System overview

Figure 8 shows the complete Teletest Focus™ system, with the populated tool clamped around the pipe and connected to the unit which is in turn connected to the controlling ruggedised lap-top pc via a PCMCIA communication unit.

Teletest® Focus unit

Tool: with populated modules and inflated collar

Laptop PC

PCMCIA or USB Communication unit

Test pipe

Fig.8 Complete Teletest Focus™ system

8

Propagating the guided waves

The individual transducer elements are removable and can be readily replaced. Teletest® uses piezoelectric transducers to generate and receive the ultrasonic signals. The transducer elements are forced against the pipe as described above. The direction of vibration/oscillation of the transducer is parallel to the pipe surface and ultrasound is generated in the pipe by a shear deformation of the transducer crystal between the transducer block and the external wall of the pipe, as shown in Fig.9.

The frequency of the oscillation must be low enough and therefore the wavelength of the ultrasound long enough for the waves to affect the internal and external surfaces of the pipe simultaneously. For this to happen, the frequency is in the 10-100kHz range.

To limit the number of wave modes generated the transducers are mounted in a ring. Each transducer oscillating on its own would generate a flexural wave, but if oscillating in unison the ring generates an axisymmetric wave. Acoustic coupling with the pipe must be the same for all transducers if this is to occur. Moreover the number of transducers in the ring has to be carefully chosen.

If a single ring of transducers were employed the wave would propagate in both directions (Fig.10). This would be unsatisfactory, since any reflected signals from each direction would be superimposed and the operator would be unsure whether a reflector was in the forward or backward going direction.

Fig.10 Propagation of

axisymmetric guided waves

Fig.9 Method of generating

ultrasound in the pipe

9

To overcome this problem two or more rings of transducers are employed (see Fig.11). The rings are spaced a quarter wavelength apart. There is a time delay between the triggering of the two rings such that the ultrasound travelling in the forward direction from the two rings is additive whereas that going backwards from one ring is cancelled out by ultrasound generated by the second.

In practice a third ring is added when generating longitudinal waves in order to cancel out completely the undesirable dispersive L(0,1) mode. For torsional wave generation two rings are sufficient because the only wave mode is T(0,1) and this is non-dispersive. Thus, the Teletest® Multi Mode system uses five rings.

After firing a pulse of ultrasound, the transducers are stationary for several milliseconds as they wait to receive any reflected pulse. The transmitted waves, whether longitudinal or torsional, are axisymmetric. If the reflector is itself axisymmetric, e.g. a pipe flange or a circumferential weld, the reflected signal will also be axisymmetric. However, an asymmetric reflector, e.g. localised corrosion, causes mode conversion from the axisymmetric wave (longitudinal or torsional) into a flexural wave (see Fig.12).

Teletest® uses this feature to distinguish between symmetrical features (welds and flanges) as opposed to localised corrosion areas.

Fig.11 Transducers oscillating

out of phase

Fig.12 Reflected wave modes

10

The Teletest Focus™ transducer rings are divided into eight octants. As well as recording the total of the received signals, the unit records the difference between the signals received by the top and bottom pair of octants to give V, the vertical flexural component. Similarly the difference between the signals received by the right and left hand pairs of octants is recorded to give H, the horizontal flexural component. Finally, if the flexural response is partly horizontal and partly vertical, as it would be for a flaw at, for

example, the 2 o’clock position, then the equivalent flexural response 2

V+2

H is calculated. If the

flexural response is small compared to the symmetrical response the indication will be interpreted as a weld or flange, whereas reflectors giving strong flexural responses are will be interpreted as corrosion or other types of flaw. For all applications where the position of the welds is unknown, it is absolutely essential to be able to distinguish welds from corrosion. This ability is one of the strengths of the Teletest® system.

Multimode inspection

In the past, most inspections were carried out with a single wave mode, longitudinal or torsional. However, this has proved to be unsatisfactory. Some types of flaw are more readily detected by longitudinal waves, whilst, for some situations, torsional waves give a less ‘noisy’ response. For this reason, Pi emphasise the great advantage of the Teletest® Multi-Mode system. Figure 13 shows a Multi-Mode module.

Collecting and displaying data - Teletest® Software

The inspection is carried out under control of the ruggedised lap-top PC loaded with the Teletest® software. This is written under Microsoft.net and is designed to simplify the task of the inspector and to enable him/her to collect and report the data as rapidly as possible. The software incorporates dispersion curves for all the common pipe schedules. Using these, it selects the nominal optimum test frequencies for both excitation modes (longitudinal and torsional). The tests are then performed at these frequencies and at up to six frequencies on either side of the nominal optimum. The test data are collected by the laptop computer. Since the umbilical between the Teletest® unit and laptop can be up to100m long, the data can be analysed in the comfort of a portable office, a truck for example. The reflections are displayed as signals in amplitude versus distance format, known as the 'A-scan' display. This is similar to conventional ultrasonics, but with a time-base range measured in tens of metres rather than centimetres. The built-in dispersion curves enable the time-base to be calibrated in metres. The Teletest® software displays the A-scans in both the forward and backward directions the A-scan illustrated in Fig.14 shows the data from the backward travelling wave.

Fig.13 ‘'Multi-Mode' transducer

module with 2 transducers

exciting the torsional mode and

3 the longitudinal mode

11

Before the signals on the A-scan can be interpreted, Distance Amplitude Correction (DAC) curves are plotted on the display. It has been found that the signals from girth welds in the pipe that decay away gradually with distance makes ideal reflectors with which to set the DAC. From experience, it is known that the reflection from a girth weld with normal cap and root profile is 14dB (a factor of 5) less intense than the reflection from the pipe end (i.e. total reflection). This is the blue line in .14. Furthermore, experience also shows that an area of thinning which has resulted in a loss of cross-sectional area of 9% in the pipe wall will produce a signal that is a further 12dB less intense than the signal from the girth weld. This –26dB level is used as a threshold for evaluating signals and is the green line in the A-scan.

The Teletest operator uses the A-scan display for interpretation purposes. At the end of an

inspection the software can automatically generate a report in a Microsoft Word template. Separate templates exist for specific situations - road crossings for example. The templates can be customised to suit the inspection company’s own formats. To record signals in the report, the test operator simply selects relevant signals in the A-scan with the screen cursor. The program automatically measures the peak of the signal as a value above or below the –14dB DAC curve, then measures the distance of the leading edge of the signal from the centre line of the transducer ring. An offset may be entered which allows distance to be measured from a known datum point rather than from the ring. The Teletest® report also contains information about the test that was entered into the ‘User Information’ page of the setting up program. For example pipe identification, test location and pipe size. The next three pages are samples from a report. The report template also includes some standard text for the benefit of the end-user describing the technique and the Teletest® system. The report is normally generated from one of several processed data files gathered at each test location, so that it may be the result of observations taken from other processed data files taken with different test parameters, for example at a range of test frequencies.

Fig.14 Teletest® A-

scan display

Teletest® Inspection Report

12

TEST DETAILS

This report contains the findings of a Teletest® inspection on the following:

Pipe Inspected: From Tanks to Plant Location: Test Location 1 Date of Test: 18/01/2007 18:39 Procedure: OPS002 Equipment: Version 1.0.0.3186 Test carried out by: Ashley Jolley

PIPE DETAILS «condition:pipe_Standar dName!=Other»

Material: Ferritic steel Manufacturing Standard: ANSI/ASME B36.10M 'Welded and Seamless

Wrought Steel Pipe' Nominal Size: 8 in Outer Diameter: 8.625 in Nominal Wall Thickness: 7.04 mm Schedule 30 Observed Pipe Condition: Heavily Pitted, Viscous Contents Pipe Orientation: Horizontal Test Direction: Both

DATUM

«condition:datum_Orientation=Horiz ontal »

Datum Position for Measurements: Pipe support behind tool Position: 3.4m Behind Tool Flow Direction: Backwards

Teletest® Inspection Report

13

RESULTS

Features identified in the valid test length are: Distance From Datum Anomaly Description Comments

-24.785m Weld -10.335m Anomaly (minor) -0.092m Pipe Support 7.017m Anomaly (minor) 8.665m Weld 9.908m Anomaly (minor) 11.191m Pipe Support 12.291m Anomaly (minor) 14.013m Anomaly (minor) 18.833m Weld 19.581m Pipe Support 28.572m Pipe Support 29.927m Pipe Support 38.29m Weld 39.427m Pipe Support 40.436m Anomaly (minor) 47.599m Weld 49.152m Anomaly (minor)

Teletest® Inspection Report

14

Summary (Cluster) Plot of Indications Reported

Sample Teletest® Inspection Report

15

APPENDIX:

Test Information

Test Frequency: 30 kHz Wave Mode: Longitudinal Test Direction: Both

Distance from datum Indication Type Comments

7.017m Anomaly (minor) 9.217m Weld 11.004m Pipe Support 14.013m Anomaly (minor) 18.841m Weld 28.791m Pipe Support 38.295m Weld 39.756m Pipe Support 47.555m Weld 49.152m Anomaly (minor)

16

Teletest Focus™

As explained above, the modules in a Teletest Focus™ tool are grouped in eight octants around the tool’s perimeter. The unit triggers the octants separately so that the tool acts as a phased array. Furthermore, the power to each octant can be adjusted to compensate for any variation in coupling. The phasing of the firing of the transducer modules enables ultrasound to be focused at a predetermined position both along and around the pipe. Thus, when a normal screening test has identified the longitudinal position of a flaw which might normally be deemed marginal, ultrasound can be focused at the position and the focal point can then be swung around the pipe in eight steps. This means that it is possible to determine both the circumferential position and the circumferential extent of a flaw. From the latter information, it is possible to estimate the flaws depth and to distinguish between a long shallow flaw and a narrow deep one, both having similar total cross sectional areas. This is obviously a significant advantage since the narrow deep flaw is potentially more detrimental.

Figure 15 shows a normal A-scan. There is an anomaly, designated #88, at a distance of -15.42m from the tool. It is a small horizontal flexural signal, and was chosen to focus on for the purposes of anomaly clarification. In the above A-scan this anomaly represents a cross sectional area change of approximately 3%. Figure 16 shows a focus scan with the focal point at a longitudinal position of -15.42m and at an angle of 45º from top dead centre. The polygon plotted inside the polar plot, on the bottom right hand side of the screen, shows the amplitude of each octant relative to all of the others. The maximum amplitude is displayed as contact between the gray polygon and the black circle. The red dot on the outside of the black circle indicates the circumferential position of the A-scan being displayed in the main window.

In Fig.17 the circumferential position has been rotated to show the maximum amplitude signal. This was located exclusively at the 225º position. Signals in all the remaining seven octants were negligible.

Fig.15 Normal

A-scan

17

Thus using the focusing technique the size of the anomaly can be more accurately determined. In this example the small anomaly, first identified in Fig.15, is concentrated in just one octant. This indicates that the anomaly is more severe than the initial scan suggests.

Fig.16 Focus

scan at 45º

Fig.17 Focus

scan at 225º

18

COMPARISON WITH OTHER NDT METHODS

The Teletest® long-range ultrasonic test system provides a means of inspecting long lengths of pipe (tens of metres) from a single location. From this one location, the, pipe can be inspected in each direction in turn. Thus the length of pipe tested from that point is twice the range over which it is possible to transmit ultrasound in the pipe. This range varies according to the pipe condition, contents, configuration, surrounding insulation or wrapping, etc, but typically is of the order of 30m. Thus 60m can be tested from one access point. Furthermore the system inspects 100% of the pipe wall. Major cost savings can therefore be achieved compared to other NDT methods. This section provides some cost comparisons. The comparisons are made for inspecting 1m of 12inch diameter pipe. The estimates are for the UK and would be approximately correct for any economically developed country. In developing countries the costs of activities such as excavation would be considerably lower. Costs that are common to all methods (mobilisation and demobilisation, etc) are not included. The costs are expressed as a ratio for each technique to the costs for inspection by Teletest®. The cases studied are: - a) Insulated pipe b) Elevated pipe (2m above ground) c) Buried pipe Cases b) or c) can be combined with a). In other words a pipe may be both insulated and buried or elevated. Major differences between the costs of inspection by the various methods hinge on costs of access. For these costs reference was made to published information on the costs of activities such as excavation, roadway reinstatement, insulation removal and reinstatement, scaffolding, etc.

1

METHODS

GENERAL

The methods considered are: -

• Visual inspection • Pulsed eddy current • Manual UT thickness gauging • Magnetic flux leakage (MFL) • Mechanised UT • Teletest® • Profile radiography

Intelligent pigging has not been considered in this note, since Teletest® is not seen as a method for testing long lengths of cross-country transmission lines. However, Teletest® may have a complimentary role to intelligent pigging, in that the MFL devices often used on pigs are understood to become ineffective when the pipe passes through a metal

1 Anon, ‘BMI Building Maintenance Price Book’ Royal Institute of Chartered Surveyors Building Cost Services Ltd., London, 2003.

19

sleeve, as would often occur at road crossings. Thus the pig could be used for the bulk of the line and Teletest® for the sleeved portions. Furthermore, there are limitations to the application of intelligent pigs. Pig launching and receiving facilities must be available. Also for some intelligent pigs there are minimum flow velocity requirements, which cannot always be achieved. In such cases Teletest® may provide a viable alternative. Visual inspection Visual inspection requires direct access to the pipe, and of course internal corrosion cannot be detected unless a camera can be fitted to an internal ‘crawler’. This is often the case in inspected sewer mains, but not with product flow lines. Another problem is that it is often difficult to quantify the amount of corrosion in a given area. The depth of individual pits may be measured with a pin gauge.

UT Thickness gauging The ultrasonic probe must be placed in direct contact with the pipe surface and the surface must be clean and smooth enough to allow coupling of the transducer and pipe wall ultrasonically. A liquid couplant is necessary. This technique is often regarded as unreliable. It is very difficult to replicate the tests exactly so sequential results can be inconsistent with apparent increases in pipe wall thickness due to very small differences in the couplant thickness between probe and pipe surface. When using a digital UT thickness gauge, without an A-scan display, it is possible to take erroneous readings off laminations in the pipe wall. Coverage depends on the spacing of the test points. Typically sampling is carried out at three positions along a 12m length and at these positions thickness is measured at the 12, 3, 6 and 9 o'clock positions. Thus only 12 measurements are made per pipe length and there is a considerable chance that a corroded area will be missed. Even if the testing density is greatly increased, there remains a significant chance that corroded areas will be missed.

Mechanised UT

The same limitations apply as for UT thickness gauging. The mechanical device that scans the UT probe over the pipe surface needs space to operate and without special scanning arms will not operate around the inside of elbows. Figure18 shows a mechanical scanner on a riser pipe. It is scanning around only a one metre wide band of the pipe. However, coverage is 100% of that band and a very detailed map of corrosion on the internal pipe surface is produced.

20

Profile radiography

Profile radiography uses a low energy radiation source and a fluorescent screen to produce a radiograph in ‘real-time’. Its purpose is to test through insulation. It is only sensitive to external metal loss and examines that part of the pipe wall to which the beam is tangential. Thus a single shot provides information on one chord only. Full coverage requires a number of shots. Normal radiographic safety procedures require an area surrounding the equipment to be clear of personnel whilst testing is in progress. Pulsed eddy current This variation of the eddy current NDT method uses pulses of eddy currents. The broad bandwidth of the pulses carries low frequencies that are able to penetrate the pipe wall. It relies on electro-magnetic induction and can therefore be conducted through insulation. It is sensitive to both internal and external metal loss. Coverage will depend on the spacing of the test points. Magnetic flux leakage Direct access to the pipe is required to saturate the pipe wall with magnetism, although the surface can be painted and need not be as clean as for mechanised UT. A corrosion map can be made of 100% of the pipe wall much more rapidly than with mechanised UT. The signal is proportional to volume of wall loss rather than remaining wall thickness. Comment All the above methods, apart from Visual and AUT, only inspect the pipe under the ‘footprint’ of a search device. Direct access to the whole of the outside of the pipe is essential. These methods are contrasted with Teletest® in Figure19.

Fig.18 UT corrosion mapping

of riser

21

Teletest® For this technique, direct access to pipe is required only over a short length to enable the ring of ultrasound transducers of the Teletest® Tool to be placed. From this one position typically 60m of pipe (30m in each direction) can be tested. So road crossings for example, can be covered without excavation. Coverage is 100%. Both internal and external corrosion and erosion can be detected and its position along the pipe length located. Test rates up to 500m per day have already been achieved but, for the purposes of this comparison, a conservative rate of 200m per day will be assumed.

Access costs

The costs of gaining access to pipe to carry out conventional NDT include: Insulated pipe - A number of methods require direct access to the bare pipe wall. Removal and reinstatement of insulation to 1m of 12 inch pipe is estimated to cost £42 (US$75 or Euro 61. This estimate assumes that the existing insulation does not contain asbestos. If it does, the estimate would be considerably higher. Elevated pipe - Fixed scaffolding to access 1m of pipe elevated 2m above ground level is estimated to cost £15 (US $27 or Euro 22). Buried pipe - The cost of excavating and reinstating 1m length of a 2m deep by 900mm wide trench to provide access to a 12-inch pipe is given in the reference (p.678) as £97 (US$175 or Euro 141). A common application is to road crossings. The cost of breaking up and reinstating a 900mm wide trench in a 200mm deep tarmac and hardcore roadway is given (p680) as £64 (US$115 or Euro 93) per metre length. Buried pipe is often contained in a steel sleeve. No estimate has been made of the cost of cutting and reinstating such sleeves. This might be as much as £250 (US$450 or Euro363) per metre.

Testing Costs

In the summary of costs, those for situations where there is no access problem have been provided by benchmarking European/North American NDT service companies. They are average costs and assume 100% coverage in all cases except for manual UT, for which spot checking is assumed.

Weld Metal loss Metal loss

Flange Conve n tional Tran s ducer

Weld Metal loss Metal loss

Flange Teletest® Tool

Guided Wave

100% Inspe c tion

Loca l ised Inspe c tion

(a)

(b)

30m

Fig.19 Comparison between conventional and Teletest® inspection of pipe.

a) Conventional techniques inspect a few cm2 under the device

b) Teletest® inspects 100% of the pipe wall for tens of metres in each direction

22

Cost Summary

Table 1 Estimated costs as a ratio to the costs of a Teletest® inspection for 1m of 12inch diameter pipe

No access problem

Insulated Buried Buried in road crossing

Buried and insulated

Elevated Elevated and insulated

Visual 0.5 4.7 7.8 13.0 10.3 1.8 5.2

Manual UT 0.9 5.1 8.2 13.1 10.6 22 5.5

Mechanised UT

7.2 11.4 13 17.9 15.1 7.9 10.7

Profile radiography

4.7 4.7 11.1 16.0 10.3 5.6 5.2

Pulsed eddy current

5.5 5.5 11.7 16.4 10.9 6.4 5.8

MFL 5.5 10.2 12.1 16.6 13.9 6.4 9.3

Teletest® 1 1 1

1

1 1 1

23

TELETEST® OPERATING ENVELOPE

General

The Teletest® technology was developed to screen pipework for metal loss features such as corrosion and erosion. Originally developed for the inspection of corrosion under insulation in petrochemical plant pipework, the technology is equally suited for application to pipelines including road crossings, bridge piers and poorly accessed pipework generally. Areas highlighted by Teletest® are identified for more detailed assessment using conventional NDT test methods. Teletest® is particularly suited to fingerprinting exercises, allowing the pipe condition to be checked on a periodic basis without the need to remove the entire insulation. The field reporting threshold is area metal loss equivalent to 9% of the pipe wall cross-section (Fig.20). Metal loss features have been detected far smaller than this level. However, a lower reporting level can result in an increase in false calls. Teletest® will provide information on the metal loss feature in terms of range from the transducer (or agreed datum) and severity (minor, moderate or severe). Long range ultrasonic testing, as currently used, cannot distinguish between a wide shallow flaw and a deep axial narrow flaw of similar cross sectional area.

Pipe diameters

Teletest® tooling currently held by Pi is suitable for testing all pipe diameters (ANSI/ASME nominal bore) from 1.5 to 48 inches. Other sizes both smaller and larger (based upon standard pipe diameters) are available to order.

Access

Access is required to 0.5m of bare pipe in order to mount the transducer ring. The ring also needs to be at least 1m from the nearest girth weld.

Pipe configurations

Teletest® really scores on straight sections of pipework, where inspection of tens of metres in either direction can be achieved. Testing around swept or pulled bends generally causes no problems. Testing around elbows can result in mode conversion of the guided ultrasound wave and thus reduced testing capabilities. Testing from a main line will not cover branch lines. These should be

tested separately.

Temperatures

Pipe surface temperatures can be up to +125°C.

External coatings

Mineral wool insulation presents no difficulties. Bonded foam polyurethane insulation leads to a loss of ultrasound. However, this merely results in a reduced inspection range.

Some limited success has been achieved in testing pipe passing through concrete walls and pipe encased in lightweight fireproofing cement. However, concrete attenuates ultrasound rapidly and may prevent the effective operation of Teletest®. Bitumastic coatings currently inhibit the effective operation of Teletest®, except where they have dried to a hard finish.

24

Some types of heavy adherent wrapping (Denzo wrap) can result in excessive loss of ultrasound. Newly applied material causes most problems. Testing has been successful on pipe where the tape has dried out and is no longer well adhered to the pipe surface. Testing of this type can be on a trial basis only.

External environment

The Teletest® signal can be transmitted along pipe that is immersed in water, with good results. However, neither the unit itself nor the transducer tool is suitable for sub-sea operation.

Internal environment

As the viscosity of the pipe contents increases, the inspection range decreases due to loss of ultrasound energy. Heavy deposits on the inside of the pipe can also be highly attenuative.

Pipe condition

Teletest® works by detecting echoes from corroded regions of the pipe. Each region acts as a reflector, in turn reducing the intensity of the ultrasound travelling beyond it. On pipework exhibiting general heavy corrosion, ultrasound will be reflected from all the corrosion, effectively reducing the inspection range. It must be remembered that this in itself is a result and the corrosion would be reported accordingly. Heavy corrosion at the place where the Teletest® tool is placed is a particular attenuative because it prevents the formation of a symmetrical wave. Test areas should be examined with a scan from a

conventional 0° ultrasonic probe beforehand.

Test range

The pipe is interrogated first in one direction and then in the other from the one transducer location. Typically ranges of ±30m are achieved. Under ideal conditions, this has gone up to ±180m. However, it can be less, if conditions are unfavourable. Table 2 summarises the factors affecting performance, principally the test range over which adequate signal to noise separation is achieved. As the degree of difficulty of guided wave propagation

increases, so the test range decreases and noise increases. Table 2 Factors affecting performance

Degree of difficulty

Surface condition Geometry Contents

Easy

Difficult

Bare metal Smooth well bonded paint Mineral wool insulation Fusion bonded epoxy Light pitting Heavy pitting Plastic coating Bitumastic coating Concrete coating

Straight lengths Infrequent swept/pulled bends Attachments/brackets

Branches

Many bends

Gas Low viscosity liquid High viscosity liquid Waxy or sludgey deposits

25

Productivity

Test rates of up to 1km per day have been achieved. As with conventional NDT, the rate of inspection depends largely on the condition on the pipework being inspected.

Proven applications

Teletest® has been used commercially over six years. During this time its benefits have been proven on: • Painted pipework • Road crossings • Mineral wool insulated lines • Offshore risers • Polyurethane foam insulated lines • Sleeves sections • Bund wall penetrations • Spirally welded pipe • Buried pipelines • Sleeved sections • High temperature lines (<+125

oC) • Spirally welded pipe

• Mixed phase lines • Stainless steel pipe • Offshore risers • Road crossings

Potential applications

Teletest® is in a continual state of development. Among the applications currently under consideration are: • Wind turbine towers. • Offshore platform jacket structures • Railway lines. • Cables

Requirements on site Teletest® is foremost a rugged site inspection tool, simple to operate and with few parts that can develop faults. For operation it needs: • Vehicle with room in rear to operate PC plus equipment storage

• Vehicle access to within 50m of test location (greater distances can be accommodated provided Pi have warning sufficiently well in advance to obtain an extended umbilical)

• 110/240v, 50/60Hz power supply or cigarette lighter socket in above vehicle • Compressed air supply or cylinder with ⅝″ BSP, RH thread, cone recessed female connector • Teletest® operation requires a Hot Work Permit.

Philosophy of screening

Teletest® does not provide a direct measurement of wall thickness, but is sensitive to a combination of the depth and circumferential extent of any metal loss, plus the axial length to some degree. This is due to the transmission of a circular wave along the pipe wall, which interacts with the annular cross-section at each point. It is the reduction in this cross-section to which the guided wave is sensitive. Figure 20 illustrates that the technique is sensitive to flaw area as a proportion of the pipe-wall cross-section. It is equally sensitive to internal and external flaws. The effect of multiple flaws is additive.

26

Corrosion is the most common cause of failure in pipes. It is imperative that any pipe carrying hazardous or high pressure fluid be inspected at appropriate intervals to detect any corrosion sufficiently early for the system to be shut down and for remedial measures to be taken before failure or leakage occurs. Because Teletest® is so much cheaper and requires less preparatory work (excavation, insulation removal, etc) than other inspection techniques it can be used more frequently in this monitoring role. Only when a suspect position on the pipe is identified is that part of the pipe exposed for more detailed examination by a conventional NDT method. This would usually be visual inspection for external flaws or ultrasonic thickness gauging for internal corrosion. In comparison with other methods of monitoring Inspections can be carried out more frequently with Teletest® because: • Scaffolding is not necessary. • Pipes do not have to be dug up. • Only a small area of insulation has to be removed. • Plant does not have to be shut down. Greater coverage is achieved with Teletest® because: • On average, 60m of pipe can be inspected from one location. • Internal and external corrosion are detected simultaneously. • The complete pipe circumference is inspected. More consistent results between repeat inspections are achieved with Teletest® because: • Placement of the Teletest® Tool is replicated exactly. • Calibration of the A-scan is exactly the same; the same pipe welds are used to set the Distance

Amplitude Correction curves that determine the test sensitivity.

t

D

A2

A1

Percentage loss of cross sectional area is given by

100% 321×

++=∆

Dt

AAAA

π

A3

Fig.20 Sensitivity to loss of cross section

27

EVIDENCE OF PERFORMANCE

Teletest® pipe inspection procedures have been validated in extensive trials and performance demonstrations.

Initial sensitivity trials

Figure 21 shows initial trials used to set sensitivity levels. The plot shows the Teletest® responses relative to the response from a cut end in the pipe (100% reflector) on the y axis against the percentage loss of pipe cross sectional area for a series of 30 artificial flaws of varying depth, width and aspect ratio on the x axis. The target reporting level set by the oil companies that supported the development of Teletest® was 9% loss of cross section. Flaws of this size produce reflected signals that are 26dB down from the 100% reflector. There is some scatter on either side of the best-fit line. It is interesting to note that the stronger reflections come from deep narrow flaws compared to shallow wide ones having the same percentage loss of cross section. This is good because it is clearly more important to detect the deep narrow flaws.

Blind validation trials

The first blind trials of the Teletest® system were undertaken within the European Reliability Assessment for Containment of Hazardous materials (RACH) project.

-60

-40

-20

0

0% 10% 20% 30% 40% 50%

% loss of pipe wall area

Signal Amplitude, dB

-26dB flaw response best fit

Reporting level

Fig.21 Signal versus flaw area as a percentage of total cross sectional area

28

This was managed by University College, London and included the gathering of NDT data from controlled corroded 6" diameter pipe samples. Eight different NDT methods were compared. The data were collected in ‘blind’ trials and evaluated by an independent team from Bureau Veritas. Figure 22 shows Teletest® results from 36 defects. The plot is of defect depth against defect circumferential length. Lines representing 3% and 9% loss of cross sectional area are also shown. The plot shows that the limit of detection is at the 3% level, only one flaw below this level being detected. All flaws greater than the 9% of cross section target were detected. The data show the classic probability of detection characteristics, with an increasing likelihood of detection with area above the 3% level. These results are important as they demonstrate that the performance of the technique, determined from 'open' tests on specimens with known flaws, could be reproduced when testing real corroded pipes with unknown (internal) flaws.

Another series of validation trials was carried out under the auspices of the Pipeline Research Council International (PRCI). In this case blind tests were carried out on 24" pipes with 86 machined artificial flaws. Teletest® detected all those with areas greater than 2% of the pipe cross section.

The results of the RACH and PRCI trials can be used to estimate an approximate probability of detection (POD). To determine a true POD would require the testing of a far greater population of flawed pipes. The results of the two sets of trials are shown in Fig.23.

0%

20%

40%

60%

80%

100%

0% 5% 10% 15% 20% 25%

% Circumference

% Wall thickness

Detected Not Detected 9% Area 3% Area

Fig.22 Results of RACH trials

29

It can be seen that the probability of detection of flaws in the 24" PRCI pipes is higher than that for the 6" RACH pipes and approaches 100% when the flaws exceed 6% of pipe cross section.

Demonstrations

Teletest® is demonstrated regularly to potential clients in order to prove capability. Two of the more the critical demonstrations were:

Jay oil field, Alabama/Florida

Teletest® was being used to inspect a high-pressure water injection line buried in and around the main Jay facility. The pipe was buried in a light sandy soil and wrapped in plastic. The test results were acted upon immediately. The position of any anomaly on the Teletest® A-scan was measured and paced out form the Teletest® tool placed on exposed pipe in a bell-hole. One such indication is shown in Fig.23.

Fig.23 A-scan of corrosion near weld WeldCorrosion

Probability of Detection

0%

20%

40%

60%

80%

100%

0 2 4 6 8 10 12

Flaw Area %

POD %

RACH data

PRCI

data

Fig.22 Plot of probability of detection (POD) against flaw area for the RACH and PRCI blind trialsFig.23 Plot of probability of detection (POD) against flaw area for the RACH and PRCI blind

trails

30

The anomaly, indicated by the (+) in the figure just beyond the weld, approaches the moderate threshold level - the blue DAC curve. A bell hole was dug down to the pipe at the indicated position, where corrosion was revealed. It was decided to cut out the corroded section of pipe (Fig.24). This confirmed the presence of severe corrosion at the position indicated, just beyond the weld.

North Slope oil field, Alaska

Before this major survey of road crossings in the oil fields at Kuparuk and Prudhoe Bay, the Teletest® system had to pass field trials to detect examples of ‘weld-pack’ corrosion. Figure 25 shows an

example of this corrosion.

The corresponding Teletest® A-scan is shown in Fig.26. It has been magnified, a useful feature of the software, to view the signal just in front of the weld. The corrosion gives rise to a moderate anomaly, about 750mm in front of the signal from the weld. The signal in the anomaly indicated by a peak in the red line indicating a horizontal flexural response due to the corrosion being concentrated near the three or nine o’clock positions.

Fig.25 Weld pack corrosion

Fig.24 Cut-out of pipe showing severe corrosion

31

Fig.26 Anomaly at weld pack

position

Weld

Corrosion

32

TELETEST® FIELD APPLICATIONS

Overview

Teletest® was developed for inspecting pipe that is inaccessible to conventional inspection techniques, because it is buried, elevated or insulated. This sort of pipe inspection is a major problem in the oil, gas and petrochemical industries and it is in these industries that Teletest® has found the most widespread use. However there is also a growing demand for such services in the power generation industry. Here pipes are generally more accessible for condition monitoring by conventional NDT methods, but the benefits of Teletest® in providing more complete coverage and greater reproducibility between repeat inspections is being recognised. There are also many applications for Teletest®, which are special ‘one-offs’ and are industry specific. Some of the Teletest® field applications, as performed by Pi, are described in this section. They have been divided into five categories (Fig.27).

Oil and gas industry

Work for the oil and gas industry is divided into upstream (exploration and production), midstream (transmission pipelines) and downstream (petrochemical plant).

Power generation

Offshore oil and gas • flow lines • risers

Teletest® Case

Studies

Petrochemical • flow lines • jetty lines • tank farms

Onshore oil and gas • Flow lines • Crossings

Special applications

Fig.27 Overview of Teletest® applications

33

Upstream onshore Field production flow lines

Most field production lines cannot be ‘pigged’. This may be for one of a number of reasons - a pig trap is not present, the pipe has tight bends or the pipe diameter is too small. Teletest®, can: • Detect damage on the internal and external surfaces with equal sensitivity • Test from only a 500mm wide section of exposed pipe • Test up to 300m of pipe from one test location. • Test several adjacent pipes in sequence • Perform under arduous site conditions • Operate whether the pipe is in or out of service • Provide highly repeatable test results • Detect very small changes in pipe condition between inspections.

Table 4 Selected work in field production flow lines

Client Application Location

IRS for Caltex 4" line pipe in oil field Indonesia Chinook for Shell 10" plastic wrapped buried pipe around

Waterton gas field. Alberta

ExxonMobil High pressure water injection line around the St Regis processing facility.

Florida

With Fairshores for Shell Assa manifold piping and 8" delivery lines in Assa and Ahia Oilfields.

Nigeria

Caltex Miri oil field, Indonesia In common with many on-shore oil fields, pipelines from the wells to the collection stations are beyond their design life. In the Miri field some of the lines have been in use since the late 1950s. Most are above ground, running through jungle. Several 4" lines were inspected in a variety of locations, including some underground. It was found that earth and corrosion products attenuated the signals. However, ultrasound could be transmitted 10m or more below ground. In some cases, the pipe temperatures exceeded 100ºC but this was within the upper limit for Teletest® tools of 125ºC.

Shell Waterton gas field, Alberta. The Waterton gas field is on the edge of the Rocky Mountains so the pipelines run through difficult country for several kilometres until they reach the gas processing stations. They are all buried and wrapped in polyethylene. The wrap has a bitumastic substrate. Where this had aged and become brittle, test ranges in excess of 30m were possible. Bell holes down to the pipe could be sunk at 25m intervals to allow some overlap between shots. One of the pipelines was being re-wrapped with new polyethylene. Test ranges under the new wrap, because the bitumen was still viscous, were reduced to only a few metres. Exxon Jay Field, USA Part of a 10" diameter, 17.68mm wall thickness high-pressure water injection line was tested within the boundary of the St Regis gas processing plant.

34

The urgency of the work led to immediate investigation of any indications detected by Teletest®. This was a true test of the system performance and the confidence of the operators. Four anomalies were recorded and confirmed. Pipe elbows could be identified in the A-scans, which was helpful when tracking the line after it had taken some unexpected turns in the plant. Shell Assa and Ahia oilfields Niger Delta. The network of pipes connecting the oil wells to the pumping and collection stations run through dense plantations and, until recently, pipe leaks were tolerated. However, pressure from environmentalists and local villagers had led Shell to rethink its pipe maintenance strategies and a major inspection programme had been initiated. As a feasibility study, a 0.5km section of an 8" delivery line was inspected with Teletest® in the Ahia field. A further network of pipes of various diameters at a manifold in the Assa field was also inspected.

The line was tested successfully by sinking deep bell-holes at about 25m intervals (Fig.28). Even deep holes were excavated in minutes by labourers from a local village. A manifold in the Ahia field was tested in a variety of locations to demonstrate the versatility of the Teletest® system (Fig.29). The 8" lines that ran to adjacent well heads from the manifold were known to be corroded. The high level of signals in the A-scans confirmed this, but the corrosion at the Teletest® ‘tool’ placement points reduced the test ranges to less than 10m. However this did not prevent the correct interpretation of unusual geometric features such as ‘pup’ pieces welded in to join pipes that did not reach each other. These were confirmed by excavating the pipe.

Fig.29 Ahia manifold

Fig.28 Teletest® tool on pipe in bell-hole

35

Upstream offshore Teletest® inspections are now a regular occurrence on offshore oil production platforms (Fig.30). On risers, they offer the only reliable method of inspecting through the ‘splash zone’ that extends between the high and low water marks.

Production platform flow lines Space is very restricted on offshore production platforms, so the pipework is stacked closely together, short in length, and densely packed with valves, elbows and other fittings. This makes access for inspection by conventional methods difficult. Yet inspection of offshore pipe-work is vitally important. Since the Piper Alpha disaster, pipework on platforms is regularly subject to full volumetric inspection with radiography or ultrasonics. A severe corrosion risk arises because of the nature of the products flowing through the pipes. Unlike other petrochemical plant, there is limited control on the fluids from the oil or gas field. Bacteria in the product can initiate corrosion. Even on straight, featureless lengths of pipe, they can create a corrosion ‘cell’ leading to rapid penetration of the pipe wall. Such isolated corrosion pits are difficult to detect. The most sensitive ultrasonic methods use mechanical scanners to orbit the sensors around the pipe. Computer imaging software produces ‘corrosion maps’ of the pipe’s internal surface. In these, the isolated corrosion pits can be identified. However, this process is extremely slow and cannot be applied to elbows. There also has to be sufficient clearance between racks of pipe for the scanner and this is often not available. Some sections of pipe may be over open water between platforms and require scaffolding to reach them with the probe scanner. Teletest® is often the only viable option. It can accurately locate areas to be evaluated more closely with ultrasonics or radiography. Among its advantages in offshore operations are: • 100% of pipe is inspected. • Pipes can be closely spaced • Inspection can be done around bends and elbows • Test range may be sufficient to cross a bridge from test points at either end. • Modular tools can be made up to fit pipe diameters from 1" upwards. • The equipment is small enough for easy transport to and from the platform. • Two-person operation saves on transportation costs and accommodation resources. Offshore in the Danish sector of the North Sea Plant Integrity joined the FORCE Institute’s inspection services for Maersk in Esjberg for a series of offshore inspections in the Danish sector of the North Sea. In the Tyra and Gorm fields a major inspection programme of all the process pipework on the platforms was instigated after corrosion had been found on one specific pipe. Extensive ultrasonic corrosion mapping using a computer aided UT system was supported with Teletest® inspections. A range of pipe diameters from 3 to 24" was covered. The compactness of the

Fig.30 Offshore platform complex

36

equipment was a major benefit in transporting it from and to shore and between platforms. Although over 12 different tool diameters were needed, the modules were interchangeable between collars, an important saving when a 24" ‘tool’ contains 56 modules and 168 individual transducer elements. On the production header lines, which included 18"diameter x 39mm WT pipe and 12"diameter x 33mm WT pipe, some problems were encountered in using the normal procedure of calibrating the DAC curves using signals from the welds. This was because the weld caps were not as pronounced in comparison with the thick wall of the pipes as would be usual for the more common smaller wall thicknesses. However corrosion associated with the pipe support clamps was successfully identified.

Similar problems were found with two 12"diameter x 33.23mm WT lines on bridges between E and B platforms and between B and C platforms in Tyra East (Fig.31). Without Teletest®, extensive scaffolding would have been needed for the UT corrosion mapping equipment to gain access to the pipes, which ran under the bridges. With Teletest®, scaffolding was only needed to confirm indications of corrosion. On the Dan field, a leak from the inside of a bend in a 24", 9.53mm WT gas pipe led to extensive inspection. The pipe crossed a 100m wide bridge between C and F platforms and Teletest® was used to inspect this from just three locations. Only one location, the one in the centre of the bridge, required scaffolding. The test range extended around the bends at each end of the bridge, where the corrosion was readily detected. Risers Risers present one of the most challenging environments for design, since the damage caused by seawater and marine growth is accentuated by the high temperatures caused by the product flowing through the pipe. Risers must also withstand wave action and storm damage. The consequences of riser rupture can be catastrophic both in terms of the environment and health and safety of those working on the platform. Some risers are connected to a wellhead directly below the platform but others are connected to a flowline on the seabed. There is thus a J-bend in the pipe. The riser ends with a flange connection to the pipework on the platform, usually combined with a valve for emergency cut-off. The riser is protected from corrosion by a cathodic protection system, but this is less effective in the splash zone. The pipe in the splash zone must have additional protection. This might be a coating of a

Fig.31 Flow lines across bridge

37

special substance such as ‘Splashtron’ or the riser may be sealed inside a sheath of stainless steel or Inconel. Alternatively, several risers may be enclosed in a caisson. Despite precautions to prevent corrosion, the protection system may break down. The constituency of the product may change so that the cathodic protection is not optimised or there may be damage caused by fatigue from platform vibration or even collisions. Another factor to be taken into account, particularly in the North Sea where regulations to prevent fire are exceptionally stringent, is the presence of thick fire prevention coatings around the riser. From time to time therefore, risers have to be inspected. Above water the access is good, the riser is visible and conventional NDT methods can be applied to determine the condition of the riser’s internal surface. Below water the inspection can be done, but with great difficulty, by divers or by ROV. Underwater inspection is very expensive and there is the cost of surface preparation and removal of marine growth. In the splash zone inspection from the outside is impossible. Inspection must therefore be conducted from inside the riser pipe. For this purpose special, ‘intelligent pigs’ have been developed from pipeline applications. These carry NDT sensors vertically down inside the riser to detect damage on the outer and inner surfaces. The inspection is very time consuming and requires bulky equipment both of which are premium costs in offshore work. Teletest® offers significant advantages specifically for riser inspection; • Internal and external corrosion is monitored • Only one access point is required above the sea and below the flange that attaches the riser to

the platform pipework. • The inspection is conducted in only one test. Data gathering can take less than five minutes. • The riser flow line does not have to be detached from the riser in order to introduce a ‘pig’. • An unlimited range of pipe diameters from 1" upwards can be inspected. • The tests are highly repeatable, so new risers can be fingerprinted before installation and the

progress of corrosion monitored over long periods in operation. Plant Integrity has performed many riser inspections using Teletest® (Table 6). Table 6 Selected Plant Integrity work on risers

Client Application Location

LRM for BG Risers on Neptune and Cleeton platforms

North Sea, southern sector

Force for Maersk O&G Pump caisson around riser TWE-A on Tyra West platform.

Danish sector of North Sea

PTTEP Risers on Barykat offshore platform Thailand TechCorr for PDVSA Risers on un-manned platform in Lake

Maracaibo Venezuela

Monitoring the condition of risers on Neptune and Cleeton platforms

Modern life-cycle management of structures and components requires an accurate assessment of condition before entering service and condition monitoring during service. BG were having risers manufactured at the Ardersier fabrication yard of Barmac under supervision from BP. The design of these particular risers called for them to be completely sealed inside caissons during fabrication. Plant Integrity was called upon to test the risers to give an initial fingerprint of their condition, and then compare the results with periodic future Teletest® inspections once installed offshore. Teletest® is ideal for monitoring applications, because the tests are repeatable. A-scans can be digitally recorded and then superimposed one upon the other to reveal even minor changes. The Neptune Platform risers are assembled in bundles before insertion into caissons. The caisson tops are sealed with lids, each riser having an anchor flange which sits in a recess in the lid. Once sealed in the caissons, direct access to the risers is no longer possible.

38

It was found that transmission of ultrasound past the anchor flanges was possible and that the sprayed aluminium coating had no adverse effect on propagation. The pipe ends, 63m away, were readily detected and the possibility of detecting 9% metal loss flaws in the splash zone, 18-20m from the transducers was demonstrated. After successful completion of the on-shore trials, BG International then asked for a baseline survey of all accessible offshore risers after final installation. A second Teletest® survey was therefore conducted on 12 and 16" risers on the Cleeton platform. The Teletest® survey was this time only partially successful. Data were obtained over a range of 50 to 70 metres from the tops of the caissons. This covered the objectives to examine the splash zone area and a substantial proportion of the enclosed riser. However, a large amount of reverberation was observed on all the recorded images, due to the proximity of the anchor flange exacerbated by the coated surface. Much ‘cleaner’ signals were obtained from the previous Teletest® survey of risers carried out at the Ardersier Fabrication Yard

. Here they were carried out on the flame sprayed aluminium alone,

whereas offshore the flame sprayed aluminium was over-coated by epoxy paint. Attempts were made to improve the coupling contact by dressing with emery cloth, and this significantly improved the data quality. There was however a practical limit to what could be achieved by these means in the initial offshore trials and plans were set for a second offshore trail. On the second offshore visit, the area of transducer application was shot-blasted back to the sprayed aluminium surface and a single coat of primer applied for corrosion protection. This produced an ideal situation, in that good sound coupling was achieved with adequate temporary pipe protection. This approach will be utilised in any further riser inspections. Operational riser R7 and 'Future use' risers R1, R2 & R5 were surveyed in the Cleeton Riser Tower (Fig.32). All four risers had adequate access and surface preparation so that good inspection conditions were achieved. In each case, data were gathered from the anchor flange, beyond the

splash zone and down to the first bend at the seabed.

Operational risers R1 & R2 were surveyed in the Neptune Riser tower. Both had adequate surface preparation but the risers had poor access due to a total of five pipes sharing one caisson. It was impossible to wrap the transducer tools completely around the risers. Due to the Teletest® technique of filling the whole pipe volume with ultrasound, these restrictions were overcome in the following way. The R1 riser (10") was inspected by removing two transducer modules from each end of the mounting collar allowing it to tuck into the narrow gap with the adjacent riser. (60mm between pipes is required to successfully mount the transducer tool).

Fig.32 Teletest® tool strapped to a riser

39

The R2 riser (16") had three restrictions in one area which precluded using the above technique but was instead inspected using a 12" transducer tool which fitted snugly to the riser with the gap left adjacent to the obstructions. In both cases, sufficient ultrasound was generated to produce an acceptable result, which could be repeated at a future date if required. As with the Cleeton risers, both the above were successfully inspected from the anchor flange, through the splash zone to the first bend at the sea bed. Inspection of riser caisson

Where risers are protected inside a caisson, the condition of the caisson can itself be a cause for concern. The riser on A-Platform of the Tyra-west complex in the Danish sector of the North Sea had received a spot corrosion check with an automated ultrasonic thickness mapping system. This had detected some thinning, but the scanning was limited to a small region just above the splash zone. To determine whether this thinning extended into the splash zone, Plant Integrity was called in to conduct a Teletest® inspection of the caisson. The inspection level was 20m below the platform deck, just out of reach of the waves so a period of calm weather had to be chosen. At 30", the diameter of the caisson required one of the larger Teletest® tools. This was lowered to the inspection level with the where two men manhandled the inflatable collar around the pipe. An umbilical cable ran to the laptop control on the platform deck. The results showed the thinning to be localised and not a significant problem. Inspection of risers in South China Sea The Thai State Oil Company PTTEP called in Plant Integrity to inspect ten 355mm diameter risers on the Barykat oil production platform. The two problems successfully overcome were the non-standard pipe diameter and the presence of clamps below tool placement positions (Fig.33). Test ranges of 20m were achieved down to the seabed.

Inspection on unmanned platform in Lake Maracaibo

Plant Integrity has been collaborating with a local company, TechCorr, to bring Teletest® technology to Venezuela. A number of demonstrations have been carried out for Venezuelan oil and petrochemical companies.

Fig.33 Teletest® tool on Baryat riser

40

One of particular interest concerned the inspection of offshore risers. PDVSA own a number of small, unmanned gas platforms in Lake Maracaibo. They are concerned about the possibility of corrosion affecting the risers in the splash zone. The purpose of this exercise was to demonstrate in principle that Teletest® was capable of inspecting this zone. Figure 34 shows the transducer ring clamped around a 6" riser. The range achieved was 13m, but was only limited by a change in pipe section. The ‘Splashtron’ coating, whilst causing some slight signal reduction did not significantly affect the ability to inspect the critical region.

Midstream transmission lines Midstream facilities are taken here to include long distance, cross country transmission lines, in-field lines and the pipework in compressor plants and similar. For the most part, transmission lines are buried and therefore have to be inspected from inside using ‘intelligent pigs’. These carry arrays of electro-magnetic or ultrasonic sensors for detecting corrosion, cracks and third party damage caused by excavators, etc. The lines are tested regularly by introducing a ‘pig’ into the pipe at a ‘pig-launcher’ and allowing it to flow with the pipe contents to a similar exit point (the ‘pig receiver’). Several hundreds of miles may be covered in one inspection. This generates enormous amounts of data, which, along with difficulties in determining the position of the pig at the time that it picked up an indication, makes precise interpretation difficult. Any pipe with a significant indication has to be dug up. If the damage is then not visible, the area must then be tested using radiography or ultrasonics. This may involve exposing several tens of metres of pipe until the precise location of the indication is found. Using Teletest®, an indication detected by the ‘pig’ can be confirmed from a single bell-hole sunk to the pipe to attach the tool. Moreover, Teletest® will locate the indication with an accuracy of ±100mm There are many lines where pig-launching facilities have not been provided or where the fluid in the line flows too slowly to transport a pig. In these circumstances, Teletest® may be used on a selective basis to inspect ‘high consequence’ areas of the pipe. Teletest® has been used in local inspections of transmission lines (Table 7)

Fig.34 Teletest® tool on riser

in Lake Maracaibo

41

Table 7 Selected Pi work on transmission lines

Client Application Location

BP Amoco Wall penetrations at beach valve station of a 36" line at the CATS terminal.

Middlesbrough, UK

Aitec West/Chinook Headers at gas field compressor station USA With TechCorr for PDVSA 10" gas line through Las Monachas Venezuela

Advantica 42" gas transmission line penetrating inspection pit.

Gloucestershire, UK

Wall penetrations at beach valve station of a 36" line at the CATS terminal. A 36" gas line passes through this station at either side of control and ESD valves, housed in a concrete pit. The aim of the inspection was to determine the condition of the line inside the 800mm thick concrete walls. On one side, the Teletest® tool had to be positioned inside a pup-piece and could not be placed more than 800mm from the wall due to a forged T-piece in the line. The wall therefore occurred within the ‘dead-zone’ of the A-scan. However, in the event good quality data were obtained by performing frequency sweeps.

Gas pipeline in Venezuela One of the most rapid pipeline inspections achieved to date was along a 9.5km stretch of 10"diameter x 9.8mm WT gas pipeline at Las Monachas, Venezuela. Three hundred metres of pipe was tested from each location, resulting in an inspection rate of over 1.5km of pipe per day, a productivity record. Working in conjunction with TechCorr, the local inspection company, Plant Integrity tested just over 3km of 10" 9.8mm wall thickness above ground gas pipe in 12 working hours. One reason for this high productivity was the large range achieved. Figure 35 shows a typical A-scan. The line was free from significant corrosion. The peaks shown are reflections from the welds. Weld M is 150m from the transducer assembly. This means that 300m was inspected from each transducer location.

Fig.35 Long range A-scan

M

42

At one point along the pipe there was a valve station as shown in Fig.36. Note that there are three

branches connecting to the line, two horizontal and one vertical.

The A-scan obtained by ‘shooting’ from a position just outside the fence is shown in Fig.37. Peaks A and B show strong horizontal flexural responses (red line) and are the signals from the two horizontal branches. Peak C, with a strong vertical flexural response (blue line) is from the vertical branch. Note that it was possible to inspect the line beyond the branches.

Inspection of headers in gas compressor stations Plant Integrity were asked by their Canadian customer, AITEC West, to assist in a novel Teletest

®

inspection of headers in gas compressor stations in Montana and North Dakota in the USA (see Fig.38). The final client, the station’s owner, was Northern Borders Pipeline (NBPL). AITEC were sub-contractors to Mears Engineering LLC, NBPL's principal inspection company. The challenges presented by these inspections were: - • The presence of some twenty 12" branches (Fig.38) • The large diameters - 36, 37 and 42" • The significant thickness - 44mm

Fig.37 A-scan from valve station

Fig.36 Valve station on 10" line

43

As Fig.38 shows, the headers were supported on concrete blocks. The aim of the inspections was to detect possible atmospheric corrosion at the six o'clock position at the interface between the headers and the concrete supports.

Because of the thicknesses involved, it was decided to inspect using torsional wave excitation. The Teletest® collar was mounted at the quarter length positions of the headers, which were up to 60m (180 ft) long. Despite the intervening branches, it was possible to 'see' to the dome ends. The presence of the branches meant that the top of the pipe was not fully inspected, particularly 'downstream' of each branch. However, this was not a problem, because, as stated above, the zones of potential corrosion were at the supports at six o'clock. NBPL were completely satisfied by these inspections. The plan now is to use Teletest® to inspect the headers on a regular three-yearly basis. Crossings Where pipes cross roadways, railways and rivers they have to be buried. Here they become inaccessible to inspection, even by intelligent pigs, as the presence of an external metal ‘sleeve’ can reduce the sensitivity of the pig’s electro-magnetic sensors. Crossings are therefore a major problem when assessing or monitoring pipe condition. Even for buried pipe, there is a greater risk of failure at crossings. The cathodic protection system may be adversely affected by contact between the sleeve and the pipe. Furthermore, corrosion may be accelerated due to drainage of corrosive fluids from the roadway. Crossings are often at low points in a pipeline and therefore may be affected by intermittent flooding. The cost of digging up a crossing to uncover a pipe for inspection will be several tens of thousands of dollars. The costs are incurred from: • Disrupting traffic • Digging trench • Implementing statutory Health and Safety regulations for exposing trenches • Cutting open a sleeve from around the pipe if present • Removing protective pipe coating • Inspecting visually the external surface of the pipe and using radiography or ultrasonics to detect

any flaws on the internal surface of the pipe • Reinstating protective coating • Replacing sleeve • Filling in trench • Repairing road Teletest® offers many advantages for inspecting pipes in road, rail or river crossings:

Fig.38 Gas compressor station

header

44

• The testing is done from each side of the crossing with only a small 0.5m band of pipe surface exposed for attaching the transducers.

• Internal and external surfaces of the pipe are tested simultaneously. • Normally two tests are required, the Teletest® tool ‘shooting’ from each side of the crossing.

Overlap between A-scans can confirm anomalies. Short crossings, less than 30m wide may be tested from one side only.

• If several pipes run through the same crossing, they can be tested sequentially, only changing the tool diameter as necessary.

• The test can be controlled from a vehicle standing on the roadway. • Inspection can be carried out beyond elbows. • Crossing widths of 60m are normally possible, although under certain test conditions widths of

200m have been achieved. • It is not necessary to take the pipeline out of service. • From the same test position, the pipe can be tested in the direction away for the crossing as well

as into it, doubling the coverage. • Tests are highly repeatable since the test conditions, equipment calibrations, etc. can be

replicated at each test. Very small increases in pipe corrosion can therefore be detected. Some of the Teletest® inspections of crossings in which Pi has been involved are given in (Table 8) Table 8 Selected Pi work on oil and gas pipe field crossings

Client Application Location

ConocoPhillips Road crossings in Kuparuk and Prudhoe Bay Arctic oil fields.

Alaska

Cooperheat-MQS for Loop LLC

Pipe under Bayou crossing. Louisiana

ConocoPhillips Alaska The oil fields on the North Slope of Alaska border the Arctic Ocean in a nature conservation area. The permafrost is extremely sensitive to damage. There is therefore a zero-tolerance of leaks from any of the pipelines and process plant. The pipelines are laid on supports 1-2m above the ground and are regularly inspected with automated ultrasonics and/or radiography. However where the pipes cross the roadways, which are built up on 1-2m high gravel banks (Fig.39) to protect the permafrost, these methods cannot be used without digging up the pipe. Because the problems of disruption are magnified in Arctic operations, the cost of digging up one crossing may exceed $500K. An alternative solution was therefore sought by ConocoPhillips and was found in Teletest®.

The pipes around the oil fields of the North Slope are insulated with polyurethane foam. A particular corrosion problem exists at the so-called 'weld packs'. The pipes were pre-insulated. However, about half a metre of pipe is left un-insulated until after welding on site. The remaining metre of pipe is insulated with a site-installed band, the weld pack.

Fig.39 Road crossing on the North

Slope

45

Moisture can penetrate at the joints between the weld pack and the pre-installed insulation, causing localised corrosion. This is therefore characterised by being 500mm to 1m away from the weld. In lines carrying multiphase products, internal corrosion may also be present often taking the form of long narrow pits, orientated in the axial direction. These features are generally found between the five and seven o’clock positions. After showing the adequacy of Teletest® in detecting corrosion associated with the weld packs, Teletest® surveys were started in the summer of 1999 and are repeated every summer for a six to eight week period. High productivity is essential because of the shortness of the weather ‘window’.

For each inspection, a band of insulation 500mm wide is removed from the pipe close to the crossing. The surface is lightly power-brushed to remove residues of insulation before mounting the Teletest® tool (Fig.40). The test is controlled and the results interpreted from a truck parked up on the roadway. During the 1999 season 64 crossings were inspected in the Kuparuk oil field and 92 in the adjacent Prudhoe Bay field. Each crossing may have six pipes covering a diameter range of 6, 8, 10, 12, 14, 16, 18 and 24". In these fields there are also caribou crossings that allow migrating herds to cross the pipelines. These can be over 100m long, are of similar design to the road crossings, and also require inspection. Inspections are also carried out on pipes under the gravel ‘pads’ that support the plant at the well-heads, water injection plants and pumping stations. These usually require bell holes to be sunk to the buried pipe between pipe ends. One hundred and ten crossings were inspected in 2000, 79 in 2001, 63 in 2002, 70 in 2003 and 63 in 2004. ConocoPhillips have now established a rolling programme in which 20% of the road crossings are inspected each summer, so that each crossing is inspected every five years. In this way corrosion can be accurately monitored. Loop LLC

This 48" pipe became at the time, the largest diameter pipe tested using a specially constructed Teletest® tool (Fig.41). Because of its size, two modular tools were strapped together. The line could not be pigged because of the two elbows. With Teletest® it was possible to inspect from the horizontal section past the first elbow into the vertical section and then past the second elbow into the horizontal section under the bayou. In particular it was also possible to inspect the pipe under the strap clamping it to the support.

Fig.40 24-inch tool near

culvert at a road crossing

46

Upstream - petrochemical industry

PROCESS PLANT

A typical refinery contains many thousands of metres of pipe, much of it tightly packed, and under insulation or at heights that cannot be reached without extensive scaffolding (Fig.42). Until recently it was acceptable to reduce the risk of catastrophic failure in pipe by adopting a leak before break philosophy. The pipes were merely viewed to look for leaks. The vessels and other containers on the other hand received extensive inspection. This is no longer acceptable as environmental pollution is not tolerated. Most plant operators need a rapid pipe inspection technique to screen their pipework. Teletest® meets this need.

Flow lines

Process flow lines often carry toxic fluids at high temperatures and pressures and are therefore generally at high risk in plant operations. Pipe monitoring is important for health and safety as well as for environmental and economic reasons. Pipes may be subject to damage from both internal and external corrosion, and possibly also from stress corrosion or creep. If the pipe contents include particles travelling at high velocities or in turbulence, there is also the possibility of erosion. Volumetric inspection, of both internal and external surfaces, is necessary using radiographic or ultrasonic methods. However, these are slow, coverage at a test location will only be a band around the pipe a few centimetres wide and the surfaces have to be exposed, prepared and made accessible, which may mean digging up the pipe, removing protective coatings, and erecting scaffolding. A particular inspection problem in process plant is caused by the presence of insulation. Perhaps 60% of a plant’s pipes are insulated, often with asbestos. The cost of removing insulation for inspection invariably exceeds the cost of the inspection itself.

Support and clamp

Transducer ring

Fig.41 48" bayou crossing – Loop LLC

Fig.42 Process plant pipework

47

The insulation itself can cause corrosion if it becomes saturated with water. Corrosion under insulation (CUI) has been identified as a major inspection issue. The first field use for the Teletest® long-range ultrasonic test technique was for detecting CUI. As well as insulation, other problems encountered in inspecting flow lines include: • Wide range of pipe diameters • Pipes are in short lengths with frequent elbows • Pipes are very close together, making access difficult for scanning devices • Pipelines are interrupted with valves, branches and support collars • Contents are varied, ranging from light fuel oils to viscous foodstuffs • Pipes are usually elevated in some sections • Pipes are often at high temperature, but can also be at low temperature.

Teletest® offers many advantages in testing pipes in process plant: • Lengths of pipe up to 200m can be tested from one test site • If insulation is present, only a 500mm wide band has to be removed for attaching the tool • Both internal and external corrosion can be detected • The exact distance from the tool to any indication can be determined • If several pipes are running in a rack, they can be tested sequentially only changing the tool

diameter as required • The tests can be controlled over a large area from one point using an umbilical • Inspection can be carried out beyond bends and elbows • The line can remain in use during testing • The damage in the pipe can be graded between severe, moderate and minor A summary of recent Teletest® inspections carried out by Plant Integrity on flow lines in process plant is given in Table 9. Table 9 Selected work on flow lines

Client Application Location

Equistar 14" line from ammonia reactor at Channel View

Texas

MB Inspection for BP Fuel oil line at Coryton Refinery UK Canspec Slurry line damaged by erosion. Canada Inspection of 14" ammonia line This insulated line was the feed to a reactor vessel in a chemical plant. It emerged from the reactor 2m above ground level, ran vertically for 7m, then horizontally for a further 10m. Visual inspection was difficult owing to the insulation and access was not feasible to the elevated section without scaffolding. External CUI was suspected. The Teletest® transducer was attached near the base of the vertical section, to examine the vertical and elevated horizontal legs. An example of the Teletest® A-Scan output is shown in Fig.43. The large peak 5m from the transducer is a weld at the elbow where the pipe turned to the horizontal. A number of flaws were reported from 13 to 19m from the transducer (marked '+' on the plot).

48

On removal of the insulation for cleaning and visual inspection, these were confirmed as areas of CUI attack, some of which are shown in Fig.44.

12" heavy oil fuel line This inspection of a 3.5km pipeline (Fig.45) was carried out over a four-day period at a UK refinery by Plant Integrity acting as sub-contractor to Northern NDT (now MB Inspection). Specific lengths of line were selected visually as having potential corrosion problems - road crossings, damaged insulation or corroded canning, leaking from the steam trace lines, evidence of flooding surrounding the pipe, etc.

Following this preliminary survey, 800m of line was inspected from 30 locations. The line was insulated with mineral wool.

Fig.43 A-scan of anomalies

in ammonia line

Fig.44 Ammonia line showing

corrosion

49

The heavy fuel oil was found to attenuate the signal slightly but did not hinder inspection significantly. Furthermore, the transducers were readily deployed, despite the presence of the steam trace lines. The longitudinally welded pipe supports at 5m intervals had no adverse effect on transmission. 'Moderate' and 'minor' anomalies, representing suspected metal loss features, were located at 12 locations and were recommended for detailed local examination. Good correlation was found between the Teletest® predictions and the flaws present. Inspection of 24" slurry line along lake shore fort McMurry This line carried water-based slurry. It was not insulated and was at ground level, so that access was not difficult. However, the main concern was local high levels of internal erosion where turbulence in the flow caused eddies and consequent high impact of particles on the inside of the pipe.

Since service history had shown that these occurrences were difficult to predict, spot thickness measurements were ineffective in detecting thinned areas prior to leakage.

Fig.45 Fuel oil line at Coryton refinery

Fig.46 Slurry line

50

Teletest® overcame this problem as 100% of the pipe wall is examined. During initial trials, a test was carried out on a section where a small leak had already occurred (Fig.46). The result is shown in Fig.47. The signal approximately 12m from the datum is from a butt weld. The very large signal which follows at 14m (marked '+') coincided with the location of the leak. This suggested extensive metal loss.

It was found by subsequent examination that there was a band of erosion almost through the wall for the majority of the pipe circumference. The pipe was therefore at the end of its life.

Terminals and storage Large storage tanks are a common feature of the petrochemical landscape. A network of pipes interconnects the tanks with on-loading and off-loading bays and allows transfer between tanks. Often placed near ports for easy connection to ocean going tankers, they include jetty lines. Tank farms Leakages from tank farms can be cause catastrophic damage to the environment, particularly as these farms are frequently on river or estuary banks or on the coast, to allow tanker access. Maintenance of the structural integrity of storage tanks and the pipelines that interconnect them is therefore essential. Important codes of practice such as API 570 set out the inspection requirements. The pipes were generally designed for ‘leak before break’, that is to say, long before any catastrophic rupture of a pipe occurs, it will leak its contents. These will be detected visually, as long as the pipe is visible so that corrective action can be taken. Until recently therefore, inspection of pipes was principally visual. However much tighter environmental legislation has made this unacceptable and pipe damage must be detected before leaks occur. The problems that arise when inspecting the pipelines in a tank farm include: • Long pipe lengths - one tank farm may have over 100km of pipe. • Some pipe may be buried, insulated or elevated so that there is no ready access. • Wide range of pipe diameters, typically from 3 to 18". • The pipes can be coated. • Pipelines all penetrate ‘bund-walls’. Such penetrations are common corrosion sites. • Pipelines are interrupted with valves, elbows and branches.

Fig.47 Anomaly in slurry line Weld

51

• Pipe contents are varied, ranging from light fuel oils to viscous foodstuffs. • Insulated pipes often have steam heating lines or tracers running along the surface. • There are numerous sites for potential corrosion in pipes including under wet insulation, pipe

support crevices, and concrete accelerated corrosion in bund walls. • To meet environmental legislation and prevent leaks, tank farm operators have taken to

hydrostatic testing the pipelines. This involves filling the pipe with water, pressurising it to one and a half times its design operating pressure and using sensitive pressure sensing devices to detect any leaks.

Hydrostatic pressure testing has many drawbacks: • The pipeline has to be taken out of service. • Time is taken up making sure that all seals at valves are tight. • There are costs accrued from disposing of contaminated water. • Lines have to be dried out thoroughly after testing. • It will not measure the amount of wall thinning, therefore indicating potential future problems. • Conventional non-destructive testing is too expensive and too limited in coverage to meet the

need for quantitative non-invasive evaluation of pipelines in tank farms. Teletest® on the other hand offers many advantages: • Lengths of pipe up to 200m can be tested from one test site. • If several pipes are running in a rack, they can be tested sequentially, only changing the tool

diameter, as required. • The tests can be controlled from a vehicle standing on a roadway. • Inspection can be carried out beyond elbows. • The line can remain in use during testing. • The damage in the pipe can be graded between severe, moderate and minor. Data is provided for analysis of overall condition of pipe in a tank farm, so that particular problem lines can be identified and dealt with in a planned maintenance programme.

Table 11 Selected Pi work in tank farms

Client Application Location

ST Services 4-16" pipes in a network on a tank farm including a 1.2km, 14" stainless steel line at Eastham works.

Birkenhead, UK

ST Services Limited

ST Services needed to prove the integrity of the pipes at one of their tank farms at Eastham on the banks of the River Mersey in the UK. They contemplated achieving this by hydrostatic testing. However, this suffered from the serious disadvantages mentioned above and they therefore decided to use Teletest® to locate any areas of corrosion. Pipe diameters were 4, 6, 8, 10, 12 and 14". Some pipes were spirally welded. In the pipe racks, access was restricted by the close proximity of pipes to each other (Fig.48). Also some were insulated, some elevated, some buried and some steam jacketed.

52

Despite the challenge of these various conditions, the survey was very successful. Test ranges depended on pipe condition, geometric features, contents, etc. However, in some cases it was possible to test 200m of pipe from a single point (100m in each direction). Testing productivity was up to 500m tested in a day. Some 7km of pipe were tested from nearly 300 locations. The length of the test shots depended on the presence of flanges, pipe elbows and pipe attachments, the viscosity of the pipe contents and the stickiness of the pipe coatings. Most of the pipe was painted with an epoxy coating, which posed no problem to the guided wave propagation. Some pipe under roadways however was coated in thick sticky bitumen that reduced guided wave propagation to less than a metre. The position of corrosion was accurately located along the line so that further assessment could be carried out with visual inspection and ultrasonics. The Terminal Manager, Andy Smith, said that he was very pleased with the outcome. Cross-checking Teletest®'s findings has given confidence in the predictions. Consequently, ST Services have abandoned hydrostatic testing. Furthermore, their confidence is such that they have prioritised the pipes and extended the period between Teletest® surveys for all pipes found to be clear of indications. Jetties

Most crude oil and oil products are transported in bulk by sea and therefore stored in tanks close to the shore. Pipelines can extend a long way from the shore to the loading jetties in deep water. The pipes are generally straight and are closely spaced in racks. Because of their length expansion loops are a common feature.

Inspection is normally visual with simple digital ultrasonic thickness measurements to keep a check on the rate of pipe wall thinning. The trend in wall thinning is generally uniform. However, local corrosion can occur at accelerated rates causing leaks without affecting overall rate of wall loss. Unfortunately

Fig.48 STC Services pipe

rack in tank farm

Fig.49 Jetty lines

53

isolated corrosion ‘hot-spots’ can occur where the pipe is not accessible, inside a sleeve for example. Detection of internal corrosion requires the use of volumetric NDT methods. Conventional ultrasonic and radiographic methods are impracticable because: • Coverage is very limited. Only a 250mm wide band around a pipe may be tested at one time. One

pipe may be in excess of 1km long. • The pipe may be insulated, in which case the insulation will have to be removed. • A wide range of pipe diameters requires a wide range of scanners and other NDT fixtures. • There is limited access between pipes on a rack. • The underside of pipes on a rack is not accessible. • Pipe contents are varied, ranging from light fuel oils to viscous foodstuffs. In the case of

radiography, this can affect image sensitivity. • Automated ultrasonics and radiography are expensive.

Teletest® offers many advantages in testing jetty pipelines: • Lengths of pipe up to 200m can be tested from one test site. • If several pipes are running in a rack, they can be tested sequentially from one data gathering

position, only changing the tool size as required. • Inspection can be carried out beyond bends and elbows. • The line can remain in use during testing. • The damage in the pipe can be graded from the amplitude and density of Teletest® signals. The

data can provide information about the overall condition of pipelines on a jetty, so that particular problem lines can be identified and dealt with in a planned maintenance programme.

• Tests can be repeated exactly, so trends in corrosion damage can be quantified.

Table 12 Selected Pi work on jetty lines

Client Application Location

Shell 4, 6, 8 and 12" Jetty lines through a sea-wall at Shellhaven

Essex, UK

Vopak Three 10", eight 8" and five 6" lines at West Thurrock terminal

Essex

Vopak Terminal jetty lines at West Thurrock and Ipswich, United Kingdom Vopak called in Plant Integrity to conduct Teletest® surveys of five 8" import lines at their Ipswich terminal and nine 6, 8 and 12" import lines at their West Thurrock terminal. In each case, over 0.5km of pipe was surveyed over a two-day period. In a subsequent job at West Thurrock, Pi inspected three 10", eight 8" and five 5" jetty lines on an overhead gantry. These had a variety of contents; firewater, oils and gases. The Teletest® excitation mode (longitudinal or torsional) was chosen to suit the conditions in each pipe. All 16 pipes were tested in a two day window; a total pipe length of 1,202m. The cost of this exercise worked out at approximately €5 (£3.5) per metre of pipe inspected over 100% of the pipe wall. Jetty lines through a sea wall at Shellhaven

This programme involved the inspection of eight lines of 4, 6, 8 and 12" diameter; where they passed through the 0.8-1.0m thick sea wall between the Shell Haven Refinery and the jetty (Fig.50).

The rest of the jetty lines are inspected visually, with periodic checks on wall thickness with a digital ultrasonic gauge to track trends in wall thinning. In the sea wall this is not possible.

54

The pipes had a bitumastic wrapping, which absorbed the ultrasound, greatly reducing range. However, the pipes were accessible from both sides so that, despite the attenuation, the technique performed successfully. Performance was helped by the wrapping being old and no longer in contact with the pipe.

Power industry

Piping in power stations (Fig.51) can be divided between primary and secondary piping. The primary piping operates at high temperatures and the damage mechanisms can be very specific, e.g. creep. The pipes have to be inspected during shutdowns using special NDT techniques.

The main application for Teletest® in the power industry is in screening secondary piping, where its ability to monitor long lengths of piping is a major advantage.

Table 13 Selected work in the power industry

Client Application Location

British Energy Fifty-two 50m long, 2" Boiler tubes Wylfa Anglesey FBS/EPRI/Continental Edison Electrical conduit piping New York State

British Energy Plant Integrity Ltd has carried out a number of inspections for British Energy at their nuclear power stations.

Fig.51 Power plant

Fig.50 Pipe rack through

sea wall

55

At Wylfa, fifty 2" boiler tubes were inspected with Teletest®. Access was very restricted (Fig.52). The tubes are bent through six right-angle bends and, in order to account for the resulting non-linear attenuation in the guided waves, the equipment was calibrated on a test pipe simulating the pipes in the plant (Fig.53).

At Heysham, some secondary piping was inspected from the jetty outside the station. The 24" power plant pipework was insulated and contained oil. This reduced test ranges, but nearly 1km of line was inspected in four days. The 16 and 20" pipes were tested from only four locations. Some areas of heavy corrosion were detected. For the boiler spine inspection a series of trials were first conducted with Mitsui Babcock at which a new boiler spine was first ‘finger-printed’ in its pristine condition and then a series of slots introduced incrementally at specific positions, such as a collar and a weld. The increase in amplitude of the signals from the slots was monitored on the Teletest® A-scan. These trials were followed by actual tests at Hartlepool and Heysham power stations. Electric Power Research Institute (EPRI)

Two inspections were carried out on behalf of EPRI in the USA. The first concerned 10, 16 and 18" conduit piping under New York. This contains highly toxic electrical insulating fluids and has for some time been a cause of concern for possible leaks. Trials with Teletest® at Continental Edison’s power plant provided encouraging results, although test ranges in the sample pipes (Fig.54) were reduced because of the presence of numerous attachments.

Fig.53 Tarong power station

Fig.52 Boiler tubes at Wylfa

56

The second inspection was at Pennsylvania Power and Light’s Susquehanna plant. The 24" emergency pump discharge and 14" emergency service water return lines were inspected. The tools had to be located tight against the concrete walls (Fig.55) to send the guided waves into the buried pipes. Test ranges of 50m (170ft) were achieved.

SPECIAL APPLICATIONS LRUT is a fledgling NDT technology and new applications are continually being uncovered. Some of the applications are summarised below.

Table 14 Selected special applications Client Application Location

TTS Japan Inspection of spherical tank support legs Japan MQS Cooperheat Reactor coils at Pharmacia and Upjohn Illinois, USA IS for Hydro-Agri Ammonia sphere support legs Le Havre,

France John Pattisson Associates Ltd for Red Funnel

18" piles at East Cowes ferry terminal Dorset, UK

TTS/JFE Cone support column on blast furnace, Fukuyama plant

Japan

Highways Agency Lamp post inspection Kent, UK INSPECTION OF SPHERICAL TANK SUPPORT LEGS

Fig.55 Twenty four inch diameter tool

adjacent to concrete wall of pit

Fig.54 Electrical conduit piping

57

The collapse of a spherical tank for storing liquefied gas was attributed, in part to corrosion of one of the support legs. This has led to general concern about the long-term integrity of the legs supporting storage spheres world-wide. There are access problems that often make it difficult to inspect sphere legs by conventional NDE techniques. The legs are often encased to give protection either from heat, in the event of a fire, or from low temperatures, should there be a spill of the liquid gas followed by rapid boil off. Conventional NDE methods would require the legs to be exposed over their full lengths. Secondly such methods would call for the installation of scaffolding to provide man access to the full leg height. Teletest® provides a means of overcoming both these difficulties as long as access to the leg can be obtained for a sufficient length for the transducer ring to be mounted. It has been used successfully to inspect sphere legs on two sites posing different challenges.

The first project was carried out by TTS, one of Plant Integrity's Japanese customers. One of the three

spheres concerned is shown on the left. In this case the legs were protected by lightweight concrete.

The following photograph shows how the coating was cut away to allow access for the Teletest® tool.

Normal full weight concrete attenuates ultrasound rapidly. However, this relatively thin light-weight

coating was less attenuative and TTS were able to 'shoot' the full 6m height of the 10" and 14"

diameter legs. The results fell into three categories - no corrosion, general corrosion and localised

corrosion.

A record showing local corrosion between 2 and 4m from the datum is shown below. Note the clear

reflection from Weld A, which is above the cladding. TTS's client expressed total satisfaction with the

inspections and monitoring of their sphere legs using Teletest® is now planned on a regular basis.

Concrete

Fig 56. Collapsed spherical tank due to corrosion

Fig 57. Teletest used to measure corrosion of tank legs

58

A

Corrosion area

Weld A

Project carried out in France The condition of 18" piles at the refurbished East Cowes terminal was determined by placing the Teletest® tool above the sea level and below the pier. A 10-12m test range reached the pile level with

the seabed, beyond which noise levels were high. However, the pile end could be detected at a further 5m range.

The integrity of lampposts has been called into question after some well-publicised instances, where they have fallen onto carriageways. The main damage stems from internal corrosion.

Fig.59 Furnace actuator

Fig 58. Teletest scan of a corroded region

59

The Teletest® tool is positioned at the bottom of the column to inspect the whole column length in one test (Fig.60). One lamppost can be tested in about 15 minutes, an important consideration when contemplating a survey of the vast number of lampposts around the country.

ON-GOING RESEARCH AND DEVELOPMENT

Plant Integrity Ltd, through its parent company TWI, is committed to research into long range ultrasonics, both to enhance performance and to develop new applications. Long range ultrasonics is a comparatively new branch of NDT with a potentially enormous variety of applications. However development is hampered by the great many wave modes that can be generated in a structure. Each has to be studied carefully to determine which is appropriate. Knowledge of the characteristics of guided waves in structures is therefore essential to the development of applications. Our understanding of the physical principles of plate wave and guided wave propagation is helped considerably by using computer models such as the finite element solver ABAQUS that can simulate wave propagation in pipes, plates, rails and even more complex shapes. The computer modelling is supported by experiments using state-of-the-art equipment. TWI’s funding for research into guided wave technology comes from a variety of sources: • An internal research and development budget • Individual Industrial Members of TWI with a specific need • Groups of Industrial members with a shared need. • Co-operative research supported by all TWI’s Industrial members • Collaborative projects with a consortium of participants supported by government funding either

from the European Commission, the UK Department of Trade and Industry, the US Department of Transportation and other similar funding bodies.

A summary of on-going research projects is given in Table 15.

Fig.60 Lamppost inspection

60

Table 15. Research projects

Project Objectives Dates Funding

Sources Budget

2

Euro €

Enhancement of the LRUT method for detection of degradation in buried unpiggable pipelines

• Benchmarking the test performance against that of intelligent pigs, which currently define the industry standard for pipeline in-service inspection

• Improving the guided wave ultrasonic method capability to provide quantitative data on anomalies that require interpretation.

• Introducing a new beam focusing technique for improved sensitivity and range.

• Re-benchmarking the performance and formatting the output for maximum effectiveness in a direct assessment methodology for pipeline integrity.

2002 to 2005

US Department of Transportation PetroChem Inspection Services Pi, FBS Inc Pennsylvania State University

1,175,000

Development of phased array and swept frequency guided wave LRUT for the fitness for service assessment of pipe corrosion

• To take enhanced LRUT tests protocols (phased array, multi-mode and frequency sweep) from the laboratory and develop them into a robust site-useable package.

• To validate these techniques by trials.

• To enable wave modes (both transmit and receive) and test frequencies to be optimised.

• To improve flaw discrimination capability. • To improve the ability to use LRUT in complex pipework such as that in

compressor stations. • To improve the application of LRUT methods to pipe with bitumastic

wrapping or buried in heavily compacted soil.

2004 to 2006

US Department of Transportation, Northeast Gas Association, BP, ConocoPhillips TTS Japan

1,012,000

Development of next generation long range ultrasonic testing systems

To develop efficient hardware for testing rail and to widen the scope for the application of ultrasonic long range guided wave testing in pipe.

2001 to 2004

TWI Member Companies

560,000

Smart structural diagnostics using piezo-generated elastic waves PIEZO-DIAGNOSTICS

To develop a new structural diagnostics environment based on piezoelectrically generated wave propagation with the following key features: • a long distance wave generation technique, based on advanced, low

frequency piezoelectric transducers • an advanced signal processing system to identify location and intensity of

multiple flaws affecting elastic wave propagation. The vision for the future is to fully integrate localised monitoring with remote assessment via telecommunications technology.

2002 to 2005

European Commission

3,000,000

2 Converted from £ sterling or $ US where necessary

61

Project Objectives Dates Funding Sources

Budget2

Euro €

Condition monitoring of large oil and chemical storage tanks using ultrasonic guided wave tomography without the need to empty and clean TANKINSPECT

To develop ultrasonic guided wave (UGW) technology that sends low frequency ultrasonic waves many metres along the tank floor plates. This technology, currently in its infancy, will be developed to inspect up to 100m across the diameter of oil storage tanks. The technique will be used to locate the corrosion and defects on the floor plates of the storage tanks.

2004 to 2006

European Commission

2,000,000

Development of inspection systems for the inspection of metal pipelines buried in concrete, water or covered with coatings PIPESCAN

To develop ultrasonic guided wave technology to increase the testing range and allow up to 100m of buried or covered pipelines to be tested in a single measurement. The technique will act as a screening measure. It will be backed up by the development of a variety of advanced inspection techniques based on digital radiography, pulsed Eddy currents and alternating current field measurement (ACFM). Such techniques can be applied from outside the insulation or other thick coatings around the pipe and be capable of detecting defects under pipe coverings. In addition, to develop automated defect recognition capabilities so operator subjectivity, that is present in all current NDT techniques, is eliminated.

2004 to 2006

European Commission

2,000,000

Development of ultrasonic guided wave inspection technology for the condition monitoring of offshore structures OPCOM

• To develop the ability to detect a fatigue crack occupying only 1% of the cross sectional area 100m from the transducer assembly

• To increase the test range from 10 to over 100m by improving the signal to noise ratio through transducer design, wave mode selection and signal filtering

• To improve crack discrimination through use of multiple wave modes and frequency sweeping

• To increase performance under ‘sticky’ coatings and concrete that attenuate the waves and reduce the test range by using lower ultrasound frequencies and torsional waves instead of longitudinal wave modes;

• To locate flaws around the circumference of the tubular member by phased array focusing and by electronic rotation of focus point around the pipe.

• To develop ultrasonic guided wave techniques that can inspect complex tubular members

2005 to 2008

European Commission

2,900,000

Development of a LRUT system to examine Offshore Subset Risers, Steel Catenary

• To develop a system to continuously monitor SCRs • To develop a novel LRUT technology to detect corrosion and fatigue

cracks in sub sea risers and SCRS

European Commission

2,000,000

62

Project Objectives Dates Funding Sources

Budget2

Euro €

Risers (SCRs) and Flow lines RISERTEST

• To develop ultrasound focusing techniques to distinguish between a deep narrow crack and a wide shallow one

Long Range Ultrasonic Condition Monitoring of Engineering Assets LRUCM

To develop methods of continuously monitoring the integrity of a wide range of engineering assets, e.g. • Oil/gas pipelines • Rails • Bridges • Sea/river defences To transfer these technologies through a group of national and international inspection and maintenance associations

2005-2008 European Commission

4,300,000

Autonomous Robotic System for Inspection of Mooring Chains that Tether Offshore Oil and Gas Structures to the Ocean Floor CHAINTEST

• To develop an autonomous amphibious vehicle that will carry out inspection tasks on chains above and below sea level

• To develop a cleaning system that will remove marine growth • To develop a vision system that will measure chain link dimensions • To develop novel NDT sensors to detect fatigue cracks using ACFM,

LRUT and resonance methods

2005-2007 European Commission

2,000,000

LRUT in the Food Processing Industry FOOD

• To develop LRUT techniques for inspecting small diameter pipes as used in the food processing industry

• To develop methods of inspecting heat-exchanger tubes

2007-2010 British Government

1,850,000

NDT Inspection of Inaccessible Electrical Wiring in Aging Aircraft AWARE

• To develop novel NDT techniques and delivery systems designed to increase confidence in the integrity of aircraft wiring

• To ensure that aircraft design lives are achieved or extended • Complimentary techniques will be developed as a global integrated

solution. • Key innovations will be the use of ultrasonic guided waves and

enhanced time domain reflectometry, combined with robotic delivery of sensors to critical buried locations

2007-2010 British Government

1,850,000

Total 24,647,000

Further information about some of these projects is given below:

63

Enhancement of the long-range ultrasonic method for the detection of degradation in buried unpiggable pipelines

Total budget: €1,175,000 (US $1,450,000 or Sterling £800,000) Duration: 3 years This project is funded by the US Department of Transportation with in-kind participation from Pi, PetroChem Inspection Services, FBS Inc and Pennsylvania State University. Its purpose is to develop the Teletest® system beyond a screening tool to one that provides quantitative information about corrosion. Pi has developed a 2

nd generation 24-channel flaw detector.

This enables the transducer array to be divided into octants, each of which is separately addressable. Thus the Teletest® tool as can be used as a phased array, enabling the guided waves to be focused. Figure 61 is a finite element visualisation of such focusing.

Figure 62 shows unfocused and focused A-scans and demonstrates that focusing greatly improves sensitivity. A flaw, which is on the borderline of detectability using unfocused ultrasound becomes

quite clearly distinguishable when focusing is employed.

Fig.61 Ultrasound focusing in straight pipe

Unfocused Focused

0

50

100

150

200

250

300

0.0

1.3

2.5

3.8

5.1

6.3

7.6

8.8

10.1

11.4

12.6

13.9

15.1

16.4

17.7

18.9

20.2

21.5

22.7

24.0

25.2

26.5

27.8

range, ft

signal

Initial pulse Pipe end

0

50

100

150

200

250

300

0.0

1.3

2.5

3.8

5.1

6.3

7.6

8.8

10.1

11.4

12.6

13.9

15.1

16.4

17.7

18.9

20.2

21.5

22.7

24.0

25.2

26.5

27.8

ra nge , ft

signal

De fect s igna ls

In itia l pu ls e P ipe end

Fig.62 Unfocused and focused A-scans

64

Focusing can also be used to enhance inspection of pipe with elbows (Fig.63).

These enhancements will make the system sensitive to the loss of cross-sectional area as percentage of pipe wall in a given octant rather than over the whole pipe circumference, thus enhancing overall sensitivity. The variable focusing of a phased array will also discriminate between several dispersed shallow flaws and an isolated deep one. Finally, intervening elbows and branches will not restrict the range and sensitivity of the test.

Phased array and swept frequency guided wave LRUT techniques for fitness-for-purpose assessment of pipe corrosion

Total budget: €1,000,000 (US $1,250,000 or Sterling £700,000) Duration: 2 years This project is funded by the US Department of Transportation, the Northeast Gas Association, BP, ConocoPhillips, TTS (Japan) and TWI. It will be carried out in collaboration with Pennsylvania State University, FBS Inc and NGA. Its purpose is to take the techniques developed in the foregoing project and to validate and ‘bed them down’ into a robust site useable inspection method for use in the direct assessment of pipeline integrity. In addition to phased array focusing, new Teletest® software enables tests to be done at a range of frequencies in an automated ‘frequency sweep’. Thus the sensitivity can be rapidly optimised for any given pipeline condition.

Next generation systems

Total budget: £380K Duration: 2 years, finishing 2005 TWI Internal Funding Much of the original development in guided wave ultrasonics involved the use of finite element computer models to study the propagation of various guided wave modes along a pipe and their interaction with discontinuities. This work was essential to optimise the hardware design.

Fig.63 Ultrasound focusing beyond an elbow

Direction of

wave propagation

65

This modelling is continuing in the development of the next generation of hardware to widen the scope of LRUT applications. The project has used a three dimensional finite element technique to visualise the wave modes and calculate the dispersion curves for wave propagation in a rail (Fig.64).

Other models have shown the interaction of a guided wave in a pipe with a pipe branch (Fig.65).

SMART structures

Total budget: £2 million Duration: 3 years, finishing 2005 This project is funded by the EC Commission. SMART structures that are able ‘to look after themselves’, and give warning of any impending failure are part of the future in maintenance. This project aims to develop a new structural diagnostics environment based on piezoelectric generated wave propagation with the following characteristics. • A long distance wave generation technique based on advanced, low frequency, piezoelectric

transducers • An advanced signal processing system able to identify location and intensity of multiple flaws

affecting elastic wave propagation. The vision for the future is to fully integrate localised monitoring with remote assessment via telecommunications technology. It will allow real-time monitoring of a structure 24 hours a day. One can envisage real-time decision-making on infrastructure integrity over the Internet.

Fig.64 Guided wave

propagation in a rail

Fig.65 Model of wave

propagation past a branch

66

The technology aims to enhance the sensitivity of long-range ultrasonic techniques by using information contained within the waveform of the transmitted pulses of ultrasound in a manner similar to ultrasonic spectroscopy. Pre-stressed low frequency piezoelectric actuators that can operate in resonant, static or quasi-static modes transmit the ultrasound while novel PVDF piezo-electric sensors detect the ultrasound.

Loaded large diameter storage tanks

Total budget: £1.4 million Duration: 2 years, finishing 2006 This project is funded by the European Commission The aim of the project is to develop the ultrasonic guided wave method to send low frequency ultrasonic waves along the floor plates of tanks up to 100m in diameter in order to detect corrosion and other defects. A tomographic technique is envisaged with transducers placed around the tank on the annular plate outside of the tank wall. The transducer transmitting and receiving modes will be sequenced to provide a cross-sectional picture of signals from features on the tank floor.

Integrated inspection of pipes

Total budget: £1.4 million Duration: 2 years, finishing 2006 This project is funded by the European Commission. Teletest® is basically a screening tool. The tests must be supplemented with other NDT methods to measure and evaluate the corrosion. This project will integrate Teletest® with other advanced NDT techniques for evaluating indications without removing any of the thick coatings that may cover the pipe. These methods do not require contact with the pipe surface and include digital radiography, pulsed eddy currents and alternating current field measuring techniques. Moreover these techniques will incorporate automatic defect recognition capabilities, so that the operator subjectivity that is detrimental to many NDT methods is eliminated.

Offshore oil production platforms and wind turbine towers

Total budget: €2.9m Duration: 3 years, finishing 2008 This project is funded by the European Commission. The project will develop long-range guided wave ultrasonics for testing large offshore structures such as oil and gas production platforms and wind turbine towers. The system will be sensitive to fatigue crack growth as well as corrosion. The guided waves will be propagated from transducers mounted on the structure and detected by sensors dispersed over the structure at permanently fixed positions. Because of the complexity of the signals, an automated defect recognition capability will be built into the system.

SUMMARY

Teletest® is a new and potent tool in the armoury of inspection techniques used for monitoring the structural integrity of pipes, pipelines and pipework. Pi, with support from TWI, is at the forefront in developing long range ultrasonic technology for new applications, vital in the maintenance of infrastructure of oil and gas supplies, petrochemical plant, power generation and transportation.