This application claims the benefit of U.S. Provisional Patent Application No. 61/230,879, filed Aug. 3, 2009, and U.S. Non-Provisional patent application Ser. No. 12/572,752, filed on Oct. 2, 2009, which in turn claimed priority to U.S. Provisional Patent Application No. 61/220,734, filed Jun. 26, 2009.
This invention relates generally to inspection tools designed to detect anomalies in tubing, piping and pipelines and, more particularly, to inline inspection tools employing magnetic flux leakage detection techniques.
Many installed pipelines may be inspected using the Magnetic Flux Leakage (MFL) technique, primarily for the purpose of identifying metal loss anomalies. Magnetic flux leakage has been shown to respond in predictable ways to anomalies in the wall of the pipeline as the principal axis of the metal loss anomaly and field angle are varied. Both experimental and modeling results have been used to confirm this effect, which is also widely described in the literature.
Due in part to limitations imposed by data acquisition, data storage and magnetic circuit designs, most in-line inspection tools have employed axially oriented magnetizers (see e.g., U.S. Pat. No. 6,820,653 to Schempf et al.). However, the present axial field magnetizer designs make identification and quantification of extremely narrow axial features difficult, or in some cases, impossible. For these feature classes, a solution using a magnetic field in the circumferential or transverse direction, have been marketed and placed in service over the past decade by pipeline inspection providers. However, due to the constraints of physics, the performance and accuracy of these transverse magnetic flux inspection (TFI) tools in general is less than that of axial field tools for general metal loss anomalies.
Additionally, these TFI tools typically require a minimum of two magnetizer assemblies in order to achieve adequate coverage, making it impractical or difficult to incorporate these into an existing axial MFL tool.
For those pipelines that may have extremely narrow metal loss features, or certain classes of seam weld anomalies, standard axial field tools do not provide adequate detection and quantification capabilities. In these cases, for MFL based tools, either the initial or supplemental surveys are performed using a TFI tool. While TFI tools may be capable of detecting extremely narrow anomalies and certain seam weld anomalies, they also detect all of the remaining volumetric metal loss features typically found in pipelines, complicating the process of identifying the targeted anomaly classes.
One of the earliest TFI arrangements is described in U.S. Pat. No. 3,483,466 to Crouch et al. Crouch discloses a pair of electromagnets arranged perpendicular to each other with detectors such as magnetometers or search coils positioned on each side of the magnets. Other than the use of permanent magnets and hall device-type sensors, Crouch's arrangement remains as the basis for most modern implementations. Additionally, some designs involve segmented or individual discrete magnets that, in most cases, retain the transverse or circumferential field direction. For example, U.S. Pat. No. 3,786,684 to Wiers et al. discloses individual magnets arranged in arrays oblique to the pipe axis with the fields of each array perpendicular the others. However, this arrangement limits the field to sections and areas between the poles of each individual magnet. Furthermore, the short pole spacing required for a Wiers-type implementation decreases the length of the magnetic circuit, thereby causing the tool to suffer from velocity effects, and also masks, distorts or degrades data quality at welds, dents, or other anomalies.
Other designs involve elaborate complex geometries, multiple magnetizer sections, and elaborate mechanical arrangements such as helical drives, gears and wheels designed to induce spiral or helical motion of the magnetizer section. For example, U.S. Pat. No. 5,565,633 to Wernicke discloses a mechanically complicated device for use with magnetizer sections having two or more magnetic circuits and a plethora of sensing units. In one embodiment, the magnet blocks are arranged with spirally situated parallel poles. In another embodiment, the magnet blocks are twisted pole pairs displaced axially. Both embodiments require mechanically induced rotation in order to achieve full coverage of the inner pipe surface. Similar to Wernicke, U.S. Pat. No. 6,100,684 to Ramuat discloses a substantially transverse field magnetization arrangement that involves multiple magnetizer sections and a complex arrangement of wheels to induce helical motion of the sections and achieve overlapping or full coverage of the pipe wall. U.S. Pat. No. 7,548,059 to Thompson et al. includes two skids (poles) that incorporate fixed magnets arranged in closely spaced pairs to create a nominally transverse field spiraling around the pipe. This tool—which includes a variety of moving parts such as supporting tendons, pulleys, and springs—requires much added complexity in order to be flexible enough to accommodate bends in the pipeline. Furthermore, the magnets in this arrangement induce a field between two parallel poles, forming a single closed loop circuit between the poles of the individual discrete magnet blocks.
Similar to Thompson et al., the magnets used in the prior art are described as being blocks, with no reference to a supple or conformable upper surface used for the magnet block. Use of a rigid contact arrangement for the magnetic circuit degrades data quality by introducing air gaps or variable reluctance zones in the magnetic field path at dents or along welds and other upsets that may be present within the pipeline. For certain classes of features, disturbances created in the ambient field mask or otherwise distort the flux leakage signals present because of the features of interest. Any magnetic anomalies existing within dents and weld zones are of greater significance due to their presence within these zones and, as such, represent areas in which data quality is critical.
Additionally, the prior art requires the use of a large number of poles or surfaces in an intimate contact arrangement to the pipe wall surface. This arrangement can result in extremely high frictional forces or resistance to motion being experienced by the magnetizer assembly, thereby inhibiting or preventing its use in applications requiring lower friction.
As already discussed, pipeline operators are currently able to inspect many installed pipelines using the magnetic flux leakage (MFL) technique, primarily for the purpose of identifying metal loss anomalies. However, for certain classes of anomalies, the current axial field magnetizer designs used in the MFL technique make detection and quantification of extremely narrow, crack or crack-like axial features difficult or, in some cases, impossible. To enable detection and quantification of these features, alternative techniques utilizing acoustic (ultrasonic) waves have been studied or employed. These acoustic waves are typically generated by external piezoelectric transducers or electro-magnetic acoustic transducers (EMAT).
EMAT implementations are usually one of two basic types: Lorentz and magnetostrictive. Both types require an external magnetic bias field to be present. In Lorentz-type EMAT, the magnetic bias field is perpendicular to the pipe wall and interacts with Eddy current-induced paths or strains in the pipe wall. The magnetostrictive-type EMAT uses a magnetic bias field that is in the pipe wall plane, axial or circumferential, and interacts with magnetically induced strains.
It is well known in the nondestructive testing industry that magnetostriction in steel is much more efficient in generating shear horizontal (SH) acoustic waves when the magnetic bias field is at an angle with respect to the sensor coil conductors of the EMAT. This result has been verified by the inventors during initial development of an EMAT sensor array according to the invention disclosed herein. During the study it was discovered that several of the notches machined into test plates were not detectable using an axially oriented magnetic bias field. Rotating the magnetic bias field angle relative to the axis of travel and the EMAT sensor provided an increase of approximately 20 decibels in measured signal. This arrangement produced a much greater signal response compared to the electronic noise, resulting in distinct crack indications above a relatively uniform baseline.
Consequently SH wave applications using EMAT sensor coils that are set at an angle to the magnetic field, are usually superior to applications where the field plane lines are parallel to the sensor coil conductors (see e.g. DE Pat. App. Pub. No. 10/2007/0058043 assigned to Rosen Swiss AG). Detection and quantification of stress corrosion cracking (SCC) is one of the main types of anomalies targeted by this technique. In addition to SCC, which is typically axially oriented, girth welds, which are circumferentially oriented, have been known to exhibit crack-like features. Therefore, for an EMAT system to be globally effective, a method is needed that is readily adaptable for detection of both axially and circumferentially oriented features.
Prior art in-line inspection tools use annular arrays of permanent magnets to magnetize the pipe in a direction that is parallel to the axis of the pipe. To obtain the beneficial angle between the magnetic bias field and the sensor coils, the sensor coils are rotated toward the pipe axis (see e.g., Canadian Pat. Appl. No. CA 2,592,094 of Alers et al.). The SH waves impinge on the plane of the axially oriented SCC at this same angle. Therefore, SH wave reflections from SCC are detected efficiently only by receiver sensor coils that are positioned lateral to and rotated toward the transmitter coil. Also, the attenuation measurements used for detection of coating disbond use receiver coils that are positioned diagonally to and rotated toward the transmitter coils. These attenuation receiver coils are shifted circumferentially so that they are in-line with the transmitted wave. An appreciable increase in received signal amplitude is an indication a coating disbond.
There is a need for an EMAT tool that provides full coverage of the inner pipe wall surface without the need for mechanically complicated structures and produces a field that may be used with EMAT sensors to detect axially- or circumferentially-oriented volumetric features and coating disbonds.
A pipeline inspection tool made according to this invention includes at least two pole magnets arranged about an external surface of the tool body and oriented oblique to the central longitudinal axis of the tool body. A sensor array is provided between the opposing edges of the two pole magnets. The sensor array includes a line or set of sensor coils that are oriented at a different angle than the pole magnets relative to the longitudinal axis of the tool body. Therefore, the sensor array is at an angle with respect to the magnetic bias field generated by the pole magnets. The pole magnets and the sensor array may each extend the length of the tool body and have a general helical-shape. Preferably, the sensor coil sets are perpendicular to the longitudinal axis of the tool body but, depending on the type of anomaly to be detected, may be arranged parallel to the longitudinal axis of the tool body.
Each sensor coil set may lie 180° opposite a corresponding sensor coil set, with a portion of the opposing sensor coil sets contained within a common circumferential band of the tool. Sensor coil sets lying on a same side of the tool body are offset from one another, being generally evenly spaced apart and equidistant from the opposing edges of the oblique-oriented pole magnets. Each set of sensor coils includes at least one transmitter coil and at least two opposing pairs of receiver coils. One receiver coil in each pair may be a RD receiver coil and the other receiver coil may be a RA receiver coil. Because the sensor coil sets are rotated relative to the magnetic bias field, the receiver coils are in-line with, and have the same angular orientation as, the transmitter coil. In other words, the receiver coils are oriented parallel to the transmitter coil and do not need to be shifted diagonally or rotated with respect to the transmitter coil.
The transmitter coil transmits a tone burst or signal that impinges upon the wall of the tubular member being inspected and travels back to the receivers. The receiver coils are spaced relative to the transmitter coil so that the signal transmitted by the transmitter coil does not mask detection of the reflected signal by the receiver coils. Each receiver coil is gated to receive these reflected signals—which may be normalized—within a targeted sampling zone and detect anomalies in the tubular member. The transmitter may then transmit a second signal after the first signal has traveled a predetermined number of times around the circumference of the tubular member. Depending on the orientation of the sensor coil sets relative to the oblique-oriented magnets, the sensor array is capable of detecting wall anomalies in both the axial and circumferential direction.
It is an object of this invention to provide a magnetic flux leakage (MFL) tool that responds to a broad range of anomalies capable of generating magnetic flux leakage signals. Another object of this invention is to provide a MFL tool capable of 360° coverage of the internal pipe wall using a single magnetizer without the need for multiple magnetizer sections, magnetizers, or relative motion between the sensors or sections to achieve detection of nominally axially oriented features. It is another object of this invention to provide a MFL tool with an EMAT array that reduces the probability of missing cracks in the pipe wall and has improved sensitivity to small defects, i.e., up to 20 db increase in signal amplitude. Yet another object of this invention is to provide an EMAT array that requires a substantial decrease in RF pulser power requirements. Still yet another object of this invention is to provide an EMAT array that includes self-calibration of the transmitted signals using the receiver coils closest to transmitter coils. A further object of this invention is to provide an EMAT array that experiences less interference between transmitters caused by acoustic ring around.
Preferred embodiments of a magnetic flux leakage (MFL) tool made according to this invention will now be described by making reference to the drawings and the following elements illustrated in the drawings:
Referring first to
Referring now to
The rotation amount of the pole magnets 41, 61 depends on the amount of rotation required to achieve full coverage of the internal pipe wall surface. Going through the sequence from
Referring now to
The final configuration of oblique magnetizer assembly 20 may include any current combination of data sets, including but not limited to deformation, high level axial MFL, internal/external discrimination, inertial data for mapping, and low level or residual MFL. In one preferred embodiment of an inline inspection tool 10 incorporating oblique magnetizer assembly 20, the tool 10 includes an axial magnetizer 100 and a deformation sensing section 110 (see
Referring now to
The oblique-oriented pole magnets 41, 61 are generally at an angle β relative to central longitudinal axis 27, with angle β being different than angle γ. Because the flux lines 81 generated by pole magnets 41, 61 are generally perpendicular to the edges 42, 62 of the pole magnets 41, 61, magnetic field 80 is rotated at an angles relative to the central longitudinal axis 21 and, therefore, is at an angle with respect to sensor coil sets 94a-e. In a preferred embodiment, angle γ is about 90°, angle β is about 45°, and angles ε is about 45°.
Arranging the sensor coil sets 94a-e perpendicular to the central longitudinal axis 27 of cylindrical tool body 21 (and therefore perpendicular to the pipe axis) allows sensor array 90 to detect features in both the axial and circumferential directions. Transmitter coils 95 generate SH waves 96 that travel circumferentially around the pipe and impinge at a normal angle (perpendicular) to axially oriented cracks. Arranging the sensor coil sets 94a-e parallel to the central longitudinal axis 27 of the cylindrical tool body 21 (and therefore parallel to the pipe axis) allows sensor array 90 to detect features in the circumferential direction. Shear horizontal waves 96 are transmitted along the pipe wall in the axial direction so that reflections from transverse cracks, such as cracks in girth welds, are detected. Unlike the orientation of receivers in prior art EMAT tools, receiver coils 97, 98 do not have to be shifted diagonally with respect to, or rotated toward, the transmitter coil 95 in order to gain the benefits of having magnetic field 80 rotated with respect to the EMAT sensor coils 95, 97 & 98.
Sensor coils 95, 97 & 98 may be mounted on a suitable mechanism such as a spring loaded pads (not shown) that keep the coils 95, 97 & 98 in close proximity to the inside diameter of the pipe. The transmitter coils 95 induce SH guided waves 96 in two circumferential directions around the pipe. The receiver coils 97 detect reflections from stress corrosion cracks (SCC) and serve as the calibration receivers. Receiver coils 98 detect the SH guided waves 96 that propagate from the transmitter coils 95 in the circumferential direction. The characteristic features of these detected signals, such as amplitude and time of arrival, can be used to detect features such as coating disbond, corrosion and SCC.
The receiver coils 97, 98 are placed at a predetermined distance from transmitter coil 95 so that signal responses are detected by receiver coils 97, 98 but not affected adversely by the initial electronic excitation pulse. Each transmitter coil 95 in a set 94a-e is grouped with two receiver coils 97, 98 on each side. Sensor array 90 preferably includes the requisite number of transmitter coils 95 and receiver coils 97, 98 in order to provide overlapping coverage of SCC and coating disbond detection. In one preferred embodiment, each of two sensor arrays 90—arranged opposite one another and for use in a 24-inch diameter pipe—included five transmitter coils 95 and 20 total receiver coils 97, 98.
Each transmitter coil 95 when fired causes SH guided waves 96 to propagate to both to the left and to the right of the coil 95 and around the circumference of the pipe. The receiver coils 97, 98 closest to the active transmitter coil 95 are first sampled in time (gated) to receive the outgoing waves 96 and then gated at a longer predetermined time delay, preferably on the order of 50 and 90 microseconds for a 24-inch diameter pipe, to detect reflections from SCC. These reflections are from targeted sampling zones “Z” located between the RD receiver coils 97 and a predetermined distance “D” past the RA receiver coils 98 so as to maximize coverage and minimize interference. The reflection signals are normalized, i.e., divided by the outgoing signals detected in the RD receivers 97 to provide continuous calibration of the signal reflections.
By way of example, considering a 24-inch pipe and a target axial sample spacing of 6 mm (0.24 in.), a pulse rate of 390 Hz will yield an axial resolution of 5.1 mm (0.20 in.). This pulse rate allows the SH wave 96 to travel approximately 4.25 times around the pipe circumference before the second pulse or tone burst is fired. Consequently, the remnants of the first pulse are between the receiver coils 97, 98 and therefore have no affect on the receiver coils 97, 98 located on the opposite side of tool body 21 within that circumferential ring at the sampling time interval (gate).
The SH waves 96 are still within the receiver gates during the third tone burst, after the wave 96 has traveled about 8.5 times around the pipe. Using an attenuation factor of 0.8 in 2 feet of travel (a factor determined from lab experiments), a tone burst transmitted at 100 percent full scale has an amplitude of less than 0.3 percent when it arrives at the receiver coils 97, 98 located on the opposite side of the cylindrical tool body 21. This amount of noise is usually negligible compared to other sources of noise, e.g., thermal electronic noise, which can be as much as 3 percent of full scale.
Coating disbond is detected in the targeted sampling zones Z between RD receiver coils 97 and RA receiver coils 98 which are located in-line with the transmitter coils 95. Coating disbond detection may be accomplished by computing the ratio of the gated receiver signals. Ratios that are above a set threshold indicate a lack of coating or disbond on the pipe in a particular zone 99.
In studies conducted by the inventors, a sensor array 90 made according to this invention has shown the following benefits over the prior art:
Additional configurations are possible, depending upon the pipe diameter, with differing numbers of pole magnets 41, 61, sensor coils 95, 97 & 98 and sensor arrays 90. For circumferential detection, for example, the sensor array 90 would be rotated at an oblique angle γ relative to the pipe axis, still being located within the angular magnetic biasing field 80. In addition to SCC and crack-like features, these configurations could respond to features such as coating disbonds and metal loss. The resulting system may also be used as an EMAT-only system or combined with any of the various other technologies available in in-line inspection tools, including but not limited to MFL, Deformation, Caliper, and Mapping.
While an EMAT tool that includes an oblique magnetizer and helical sensor array has been described with a certain degree of particularity, many changes may be made in the details of construction and the arrangement of components without departing from the spirit and scope of this disclosure. An EMAT tool according to this disclosure, therefore, is limited only by the scope of the attached claims, including the full range of equivalency to which each element thereof is entitled.
Number | Name | Date | Kind |
---|---|---|---|
2219708 | Kruse | Oct 1940 | A |
3483466 | Beaver et al. | Dec 1969 | A |
3786684 | Wiers et al. | Jan 1974 | A |
4330748 | Holden | May 1982 | A |
4675604 | Moyer et al. | Jun 1987 | A |
4691572 | van den Berg et al. | Sep 1987 | A |
4789827 | Bergander | Dec 1988 | A |
4797613 | Wentzell | Jan 1989 | A |
4851773 | Rothstein | Jul 1989 | A |
4857851 | Anderson et al. | Aug 1989 | A |
4893077 | Auchterlonie | Jan 1990 | A |
4909091 | Ellmann et al. | Mar 1990 | A |
4952875 | Adams et al. | Aug 1990 | A |
5233297 | Lara | Aug 1993 | A |
5256966 | Edwards | Oct 1993 | A |
5419206 | Kamioka et al. | May 1995 | A |
5454276 | Wernicke | Oct 1995 | A |
5532587 | Downs et al. | Jul 1996 | A |
5565633 | Wernicke | Oct 1996 | A |
5574223 | Kiefer | Nov 1996 | A |
5581037 | Kwun et al. | Dec 1996 | A |
5619423 | Scrantz | Apr 1997 | A |
5777469 | Hockey et al. | Jul 1998 | A |
6009756 | Willems et al. | Jan 2000 | A |
6087830 | Brandly et al. | Jul 2000 | A |
6100684 | Ramaut | Aug 2000 | A |
6404189 | Kwun et al. | Jun 2002 | B2 |
6456066 | Burd et al. | Sep 2002 | B1 |
6628118 | Amini | Sep 2003 | B1 |
6820653 | Schempf et al. | Nov 2004 | B1 |
6967726 | King et al. | Nov 2005 | B2 |
6975120 | Amini | Dec 2005 | B2 |
7084623 | Imamoto et al. | Aug 2006 | B2 |
7143659 | Stout et al. | Dec 2006 | B2 |
7256576 | Mandziuk et al. | Aug 2007 | B2 |
7356421 | Gudmundsson et al. | Apr 2008 | B2 |
7358721 | Narishige et al. | Apr 2008 | B2 |
7548059 | Thompson et al. | Jun 2009 | B2 |
7683611 | Burkhardt et al. | Mar 2010 | B2 |
7782048 | Sawawatari | Aug 2010 | B2 |
8089273 | Hoyt | Jan 2012 | B2 |
20040095137 | Kwun et al. | May 2004 | A1 |
20040207395 | Sarfaty et al. | Oct 2004 | A1 |
20040217759 | Burkhardt et al. | Nov 2004 | A1 |
20040221652 | Flora et al. | Nov 2004 | A1 |
20040232909 | Imamoto et al. | Nov 2004 | A1 |
20050072237 | Paige et al. | Apr 2005 | A1 |
20060027022 | Flora et al. | Feb 2006 | A1 |
20060158181 | Shoji | Jul 2006 | A1 |
20070222436 | Gao et al. | Sep 2007 | A1 |
20070229066 | Narishige et al. | Oct 2007 | A1 |
20080092672 | Gibson et al. | Apr 2008 | A1 |
20080215257 | Stripf et al. | Sep 2008 | A1 |
20090078048 | Alers et al. | Mar 2009 | A1 |
20090139337 | Owens et al. | Jun 2009 | A1 |
20090158850 | Alleyne et al. | Jun 2009 | A1 |
20090193899 | Panetta et al. | Aug 2009 | A1 |
20100117635 | Hoyt | May 2010 | A1 |
20100199767 | Ganin | Aug 2010 | A1 |
Number | Date | Country |
---|---|---|
2592094 | Jul 2006 | CA |
102007058043 | Jun 2009 | DE |
59058358 | Apr 1984 | JP |
60080760 | May 1985 | JP |
60080760 | May 1985 | JP |
62067447 | Mar 1987 | JP |
WO 2006048290 | May 2006 | WO |
WO 2006069684 | Jul 2006 | WO |
Number | Date | Country | |
---|---|---|---|
20100327858 A1 | Dec 2010 | US |
Number | Date | Country | |
---|---|---|---|
61230879 | Aug 2009 | US | |
61220734 | Jun 2009 | US |