Flaw Signature Repeatability in OCTG and Drill Pipe Inspection

Key words:  Repeatability, Solid State Magnetic Sensor, Electroacoustic

Definitions:
Ultrasonic Transducer -
An electroacoustic or magnetoacoustic device containing an element for converting electrical energy into  acoustical energy and vice versa.

Solid State Magnetic Sensor - Small semiconductor element directly responsive to both the magnitude and direction of the magnetic field.

Flaw Signature Repeatability
Flaw signature repeatability in any nondestructive testing method is dependent upon the reliability of the test method employed, the test hardware and the operator. Recently there has been increased interest shown by major oil company representatives in the repeatability of  Electromagnetic Inspection (EMI) and A-SCAN ultrasonic inspection (UT).  Volumes  have been written on both of these nondestructive testing methods.  The following information reveals the physical differences and operational characteristics of  the UT transducer and solid state magnetic sensor.  Repeatability is greatly affected by each method’s basic sensor design, size and functionality.  In addition, environmental adaptation also becomes an important aspect since inspection of OCTG & drill pipe may be required in cold, if not freezing, conditions.

Electromagnetic Inspection  (EMI)
Flux leakage EMI sensors are employed to locate flaws  in Oil Country Tubular Goods (OCTG) and drill pipe.  An active electromagnetic field is induced and EMI sensors are passed across the outer surface of the tube.  Discontinuities, both ID and OD, are detected.  Two common sensors used in EMI of OCTG and drill pipe are the search coil and the solid state magnetic sensor (SSMS).  Disadvantages of the search coil (a wire wound induction coil) have been enumerated.1  The solid state magnetic sensor, conversely, brings finite control to the electromagnetic inspection system.  An extremely wide range of frequencies from defects are sensed by the SSMS,  providing exacting and repeatable signals.  Flaw signal repeatability is optimized.  The solid state magnetic sensor is extremely robust and can be employed for many years; greatly reducing the average cost of this electromagnetic transducer.

The linearity and frequency response of the solid state magnetic sensor also allow their use in locating wall loss.  100% surface coverage is provided.  Both ID and OD full length and localized wall loss is rendered by solid state magnetic sensors.  EMI wall loss detection eliminates the gamma radiation tool still used by some pipe inspection service companies.  Radiation may now be retired from its roll as an integrated part of antiquated multi-station EMI tubular inspection systems.  Search coils are not able to render wall loss signals or provide repeatability to high frequency detects.

When using solid state sensors to inspect OCTG for flaws, there is no special interpretation to defect signal renditions.  “Solid state signals,” the NDT Handbook tells us, “ …. are directly proportional to the actual magnitudes of the magnetic flux density B and have uniform sensitivity over a wide frequency range, including the lowest possible test frequency (DC).  Conversely, search coils respond differently for each defect frequency component they encounter.”2

Inspection rates are very rapid using EMI.

One other notable advantage to SSMS is their small size.  Each sensor is about one-eighth (1/8) the size of the typical ½” ultrasonic transducer.  By comparison, this small foot print permits greater resolution to defects than in those systems using electroacoustics.  Resolution is the sensor’s ability to separate fractional component parts of the pipe and make these discretely visible to the inspector.  Multiple solid state magnetic sensors (providing up to 100 individual tests in time of flight) may be used together around the pipe’s circumference.  No fluid coupling of any kind is required between the solid state magnetic sensor and the pipe’s surface.  Sensors may also be used in a non-contact configuration.  The geometry of flaw location to sensor position is not a factor when using solid state sensors.

Another important consideration is the absence of a helix when using  electromagnetic inspection.  The tubular does not need to spin when inspected by electromagnetics.  100%  surface coverage with matched circumferential sensitivity is provided when solid state magnetic sensors are used as flaw detection devices.

A-Scan Ultrasonic Testing
Ultrasonic testing has been around for many years. According to Robert E. Green, Jr. of Johns Hopkins University “…..it was Floyd Firestone’s development of the Supersonic Reflectoscope in 1942 that led to the practical instruments and techniques used in the United States today.”  And to a large extent not much has changed in ultrasonic transducer manufacturing.  The NDT Handbook tells us,  “ The piezoelectric element parameter values supplied by the manufacturer are usually based on an average value derived from sampling several batches of piezoelectric materials.  The nominal values are not sufficiently precise for use in modeling the performance of Transducers.”3 In short, the actual value of the transducer in MHz can be different from that which is marked on the outside of the element’s housing.

According to experts, “No commercial (search) unit is perfect, given the best designs known.  Also, no existing transducer is as good as the idealized transducers  for which the theory of beam divergence has been worked out”4

In addition,  “ Selection of the transducer face diameter can have significant effects on ultrasonic test results.”  “In the near field or Fresnell field, the amplitude of echo signals from discontinuities can vary widely and lead to misinterpretation of discontinuity size or location.  Considerable caution should be exercised when interpreting test indications from the near field.  The length of the near field zone should be routinely calculated  to avoid interference with test procedures and signal interpretation.” Consequently, surface breaking fatigue cracks which often appear during the drilling process are difficult for the ultrasonic transducer to detect and adequately express at their true values using  present day A-Scan technology.  This is especially problematic when calibration of electroacoustic elements are in the 5% range or greater of nominal wall thickness.

UT Beam Divergence (Spread)

Definitions:
Beam spread -
“The divergence of the sound beam as it travels through a medium”.5
In physics - “the total amount of flux escaping an infinitesimal volume at a point in a vector field, as the net flow of air from a given region.”
Vector - “a quantity possessing both magnitude and direction, represented by an arrow the direction of which indicates the direction of the quantity and the length of which is proportional to the magnitude.”
Diverge - “to move, lie or extend in different directions from a common point; branch off ”.6

“The process of beam divergence, also called beam spreading and ultrasonic defraction, leads to two distinct effects.  With the transducers’ axis as the cylinder axis, there is wave amplitude and concomitant” (acting together) “beam energy outside the cylinder defined by the transducer’s active area.  In actuality there are many lobes in the wave field.”7

“All Ultrasonic beams diverge, even focused beams.  Focusing tends to concentrate the beam at a focal point but the point is never ideal.  The focusing element or curvature only tends (to be inclined but not succeed totally)  to overcome the natural divergence of the beam.”8

Consequently, using a standard half inch (1/2”) UT  transducer, if after calibration a 1/16” hole does not pass through the center of the transducer, the resultant amplitude will be different and not repeatable.  This change in amplitude may not alert the UT unit operator to the  flaw’s presence even though it would normally be cause for rejecting the part under inspection.

According to the NDT Handbook,  “The first limitation and correction factor coming from  beam divergence arises from the fact that the energy in the beam from a transducer does not remain in a cylinder of the same diameter as the active element of the  transducer, but rather starts in a cylinder and then after some distance spreads into a cone.”  “This beam spread reduces the intensity of the wave impinging on a discontinuity in a nondestructive testing application and lowers the resulting back reflection amplitude.”9

The NDT Handbook further teaches, “Because of the factors that affect the height of an A-scan sized from a discontinuity, it is often difficult to determine with precision the actual size of a discontinuity.  Ultrasonic tests are more qualitative than quantitative in this respect”.10

Helix
Any helix produced by the rotating of the test object in relationship to the UT transducer will automatically induce non-repeatability to flaws.  This is especially critical when one  considers ultrasonic beam divergence since most ultrasonic evaluations are based on a single pass over the test objects’ surface.  Even a tight helix will produce negative results.  Multiple passes can produce flaw signal renditions which vary as much as 50% or more.  The addition of extra transducers does not absolutely guarantee detection at proper amplitude.  In the worst scenario the defect will be missed entirely.

Concerning Signal Error Using Wheel Transducers to Inspect OCTG
"Wheel transducers consist of a plastic tire filled with coupling fluid under pressure.”  If the inspection parameters such as rotating the drill pipe creates a wheel bounce or if couplant is not  meticulously applied, then the resultant indications from the ultrasonic device may be of no value.  “The tire (itself) creates reflections that need to be discriminated from the significant reflections of the test object.”11  In short, there may be an attenuation of a true signal created by the plastic tire which causes the defect signal to be misinterpreted or missed all together.  This is especially critical on the near surface.  In addition, any helix produced by the wheel will not allow 100% coverage of the pipe’s inspectable surface.  The bounding of the transducer when tracking along a rotating tubular induces noise into the test results, often appearing as defects.  Wheel transducers cannot be compressed during inspection.  Pressure applied to the wheel subverts its calibration as the distance to the cross section changes.

Regarding Near Field (Surface) Fatigue Cracks
A small fatigue crack may be missed altogether due to its position immediately under the transducer.  How this can happen is related to an ultrasonic term called the diffraction unit.  According to the NDT handbook, “A diffraction unit is a distance determined by dividing the square of the transducer radius by the wavelength of the sound wave.  For the early echoes in an echo pattern, when the sound wave has not yet traveled one diffraction unit, the measurements are said to be in the transmitter’s near field and interference effects between the waves launched from the edges and center of the transducer can cause larger apparent  losses, as much as 2 or 3 dB echo.”12 Consequently, any signal from a surface breaking fatigue crack may be interfered with and not reported as a flaw.  Note:  This negative operating characteristic when added to a wheel transducer’s signal error exacerbates the difficulty of reporting fatigue cracks on the pipe’s surface; especially on the fly.

Plastic coupling blocks known as delay lines, buffer rods or stand-offs may be used to minimize near field zone problems.  A sonic delay line should be long enough so that multiple reflections do not appear in the A-scan signal trace ahead of the back surface reflection signal from the test object.  Such multiple reflections can mask discontinuity signals from the test object.  A–Scan UT often uses angled, thin wedge shaped coupling blocks to help intercept flaws.  However, it is not practical to test with different angle beam UT transducers to explore each discontinuity.  Consequently, a flaw not at the correct angle of the coupling block used can be missed.  Geometry plays a critical roll in UT inspection of OCTG due to the curved OD encountered.13 UT signal amplitudes are subject to interpretation.

Regarding  UT inspection Oil Country Tubular Goods

Transducer Orientation
According to the NDT handbook, “Not all discontinuities are radial nor are they parallel with the centerline of the pipe or normal to the transducer beam.  As a result, the inspector’s technique (interpretation) becomes critical to the success of the test.  The (ultrasonic) transducer beam generally cannot be perfectly oriented to all discontinuities so the scan unit most often is used as a locator, providing the best possible information about the size, orientation and axial location of the indication.”14 Consequently, certain flaws which do not intersect the transducer perfectly may be reported at a greatly reduced amplitude or missed completely.

Nonuniform or nonrepeatable UT results can lead to inspection error when evaluating OCTG and drill pipe.
The NDT Handbook continues, “With this information,” (about a suspected flaw) “the inspector uses a magnetic yoke, ultrasonic testing unit and transducer, boroscope, file, grinding wheel, thickness gauge and dial depth indicator to physically determine the extent of the discontinuity.  There are no master reference standards (test rings) at this time against which all other calibration standards may be checked, as in some other areas of ultrasonic testing.  Using the same testing unit and transducer on different reference standards (thought to be the same), amplitude differences as much as 7 dB have been recorded.  This means that company A (with their reference standards) could reject good material and company B could accept rejects.  Ultrasonic testing of oil field tubular goods is not as routine as other tests (like electromagnetics) because of the variation of reject potential for any given depth of discontinuity, dependant on the wall thickness in its area.  This variation, coupled with the non-radial and off-axis orientation of the discontinuities, makes ultrasonic testing much more complex and demanding than many other applications.”15  In short, ultrasonic inspection can present vagaries in defect signals based simply on the depth of the flaw.  Deep cracks can be averaged out and not detected; shallow near surface cracks can go undetected all together.

Couplant
The operating range of ultrasonic equipment is limited, especially when water is used as a couplant.  Water is the least expensive couplant but is not operable in weather below 0° C.  Many other couplant types will also react unfavorably to freezing temperatures. There are limitations to types of couplant, thickness of same (which can provide untrue flaw detection information) and as applied to shear wave is limited to solids or very viscous liquids.  These fluids must be removed from the pipe before further processing.  Water is not the best choice for shear wave ultrasonic inspection since bubbles in water may be seen as defects thereby providing untrue information about the drill pipe’s actual condition.16  Signal error due to couplant loss during inspection is common.

Important points to remember when choosing the Ultrasonic testing method:
a)  A coupling fluid of the same type must be applied to the reference standard and the pipe’s surface.
b)  Transducer beam width can cause near field (surface) inspection to be inaccurate.
c)  Length to depth signal averaging of transducers can occur.
d)  Slow inspection speed.
e)  Less than 100% surface coverage.
f)  Extremely rough surfaces may prevent effective sound coupling.
g)  Any helix degrades repeatability.

 

Summary
End-users of tubulars inspected by EMI or UT systems must be aware of intrinsic limitations and complexities of these nondestructive testing devices.  When relying on either or both of these testing methods from pipe inspection service companies, a dynamic demonstration on a customer supplied pipe reference standard is strongly recommended.  When evaluating these nondestructive testing methods, ensure that the signal signature repeatability  is optimum.  Signal replication from man made defects should look nearly identical.  The reference standard must be inspected in its four quadrants to prove repeatability.  This is an imperative since any defect in the standard seen at less than rejectable amplitude, (after calibration) is a warning flag.  Defects in tubulars can be anywhere in the pipe’s cross section and along its entire length.  Any nondestructive system which treats known flaws in a reference standard differently will do the same to a real defect.  The possibility of returning injurious anomalies back into the well is not an option.

 

Reference:  Endnotes

1, 2
Edens, C. Wade.  “Electromagnetic Inspection:  Wall Loss and Flaw Location in Oil Country Tubular Goods”, presented at the 1991 American Society for Nondestructive Testing, Petroleum Industry Inspection Technology II topical  conference, Houston, Texas, June  25-27, 1991.

3 - 5, 7 - 16
Nondestructive Testing Handbook, Second Edition, Volume Seven, Ultrasonic Testing, American Society for Nondestructive Testing, 1991.

6 The Random House College Dictionary, 1982 Random House, Inc.