7/29/2019 Exploring Methods for Measuring Pipe Weld

1/3

JUNE 200748

Plasticity in the vicinity of a crack tipintroduces complex stress-strain distribu-tions and nonuniform material flow asso-ciated with ductile fracture processes. Themost accepted criteria regarding thesecomplex behaviors are the crack tip open-ing displacement (CTOD), J-integral, andlately, crack tip opening angle (CTOA).Crack tip opening angle is considered acomputationally attractive operationalparameter. It provides an alternative tothe J-integral criterion and shows prom-ise as a fracture criterion for resistance tocrack growth and arrest of unstable duc-tile cracks (Refs. 1, 2).

The CTOA is based on the crack open-ing displacement (COD) ductile fracturecriterion widely used in the 1960s and1970s for fracture assessment of thick-

walled pressure vessels (Ref. 3). Moreover,the CTOA test was developed to describethe crack growth process for crack propa-gation analyses in metal sheets, thin ma-terials with low to average crack tip con-straint. Currently, the CTOA criterion isa widely accepted material property usedto characterize fully plastic fracture.

The CTOA can be directly measuredfrom the crack opening profile, related tothe geometry of the fracturing structureas illustrated schematically in Fig. 1 andgiven in the following expression:

wherer= 0.5 to 1.5 mm behind the cracktip and, if several measurements are per-formed, its mean value is used.

In practice, the CTOA is often meas-ured directly from images of the test sam-ples using optics to measure the angleformed by the fractured surfaces just ad-

jacent to the crack tip, rather than usinga displacement gauge. This approach al-lows for the use of a specimen with alonger ligament for crack growth than atypical CTOD specimen, and the long lig-ament specimen design is thought to beimportant for better understanding ofcrack growth in full-sized structures. Thus,crack growth through a weld section canbe investigated in a more appropriate way.

A CTOA-based design criterionagainst crack propagation is usually writ-ten in the form:

CTOAmax < CTOAc,

where CTOAmax is a measure of the max-imum crack driving force calculated froma knowledge of the dimensions, materialproperties, and operating conditions, and

CTOAc is the resistance of the material tocrack growth.

Thus, the use of the CTOA criterionin an engineering critical assessment(ECA) approach can lead to a safe pre-diction of unstable crack propagation. Forexample, the propagation of an unstablerunning crack from a flaw detected in apipeline girth weld can be evaluated withthe knowledge of the resistance, CTOAc,of the welded girth pipeline section.

CTOA Specimen andTest Setup

A modif ied double-cantilever beam(MDCB) specimen (Fig. 2) was used toconduct the CTOA test (Ref. 2). TheMDCB specimen is designed primarily toprevent bending, which has been experi-enced in both standard and tapereddouble-cantilever beam (DCB) speci-mens. The modified specimen exhibits thefollowing characteristics: A large width, thickness, and ligament

provide a large plastic zone. High constraint in the test section is pro-

moted by two thicker loading arms. Thisserves two purposes. First, nonnegative,longitudinal strains can be achieved,and second, the loading is predomi-nantly in tension with only a small shearcomponent.

The test section does not restrain thefracture transition modes.

The test section is flat near the crack tipfor ease of CTOA measurement.The MDCB configuration and dimen-

sions are depicted in Fig. 2. The large in-plane dimensions of the specimen (200 100 mm) and the long ligament allow rel-atively large amounts of stable crackgrowth.

The machined notch (1.6 mm height)is 60 mm long (measured from the load-line of the specimen to the notch tip).

The specimen is loaded using a pair ofthick plate grips that are bolted to thespecimen Fig. 3. The loading plates in-

CTOACTOD

r=

2tan

2

1

Exploring Methods forMeasuring Pipe Weld

Toughness

Ph. P. DARCIS, J. D. MCCOLSKEY, C. N. MCCOWAN, and T. A. SIEWERT(siewert@boulder.nist.gov)are with NIST, Materials Relia-

bility Division, Boulder, Colo.

A look at the development of a new damagemechanics method for assessing the intrinsic fracture

properties of welded pipeline steel

BY Ph. P. DARCIS, J. D. MCCOLSKEY, C. N. MCCOWAN, AND T. A. SIEWERT

7/29/2019 Exploring Methods for Measuring Pipe Weld

2/3

49WELDING JOURNAL

crease the constraint in the gauge sectionof the thin flat specimens. Two cylindricalpins provide free rotation of the grippedassembly (specimen plus loading plates)during the experiments Fig. 3. The longuncracked ligament and the loadinggeometry provide the condition of stablecrack extension in the specimen ligament,similar to that of the real structure.

To facilitate the CTOA measurement,

a fine square mesh, with a spacing of 1 1 mm was lightly etched by laser on theface of each specimen.

CTOA Measurement

Digital images of the propagatingcrack tip are captured using a cameramounted on an xyz-stage Fig. 3. Thecamera and stage are controlled using apersonal computer and image analysissoftware to acquire images during theCTOA test. The captured images have asize of 2048 1536 pixels, which resultedin a resolution of about 32 m per pixel,in the configuration typically used for ourtesting. Images are acquired and stored,along with time, load, and displacementdata, by the software as the crack propa-gates across the test section of the MDCBCTOA specimen.

The CTOA is determined with soft-ware that requires the operator to tracethe profile of the crack tip, and then markdata points along the upper and lower gridlines bounding the crack, as shown in Fig.4. A series of algorithms is then used tocalculate the CTOA from the outline(profile) of the crack tip.

In each image, the CTOA is measuredby utilizing data from the crack profile

within the range of 0.1 to 1.5 mm from thecrack tip, as prescribed by the ISO and

ASTM draft standard (Ref. 4). As shownin Fig. 5, this method never uses the cracktip, or the region immediately adjacent toit in the calculation of CTOA. Variouspairs of points from the crack profile dataare used to define best-fit lines, and theintersection of these lines is used to cal-culate the CTOA.

This method uses an algorithm that se-lects pairs of points along the crack pro-

file. The pairs of points are used to deriveseries of CTOA(i)values as follows:

where i is the distance between the twopoints located at the position i, andri isthe distance between two locations i = 0and i Fig. 5.

The mean of the individual CTOA(i)

data obtained from a single picture is de-

fined to be the CTOA of the growingcrack, related to the crack extension a:

CTOA Results for a PipelineSteel and Girth Weld

Crack tip opening angle test specimenswere extracted from an X100 high-strength pipeline steel. The test specimens

were oriented to produce crack growthalong the longitudinal axis of the pipe andperpendicular to a girth weld Fig. 2.The experiments were conducted undercrack opening (mode I) loading and quasi-static conditions, at a low strain rate (inthe range of 0.002 to 0.02 mm/s) underdisplacement control. Figure 6 plots theCTOA results from more than 115 imagescaptured from two CTOA tests.

In Fig. 6, the initial CTOAc is high

(around 60 deg), and it rapidly drops as

CTOAn

CTOAa i a

i

n

=

=

1

( )1

CTOAri a

i

i

a( ) =

0 (rad),

Fig. 1 Definition of CTOD and CTOA measurement schemes.

Fig. 2 MDCB specimen with a weld section exhibiting a heat-affected zone (HAZ).Dimensions given in millimeters.

Loadingpins

Plategrips

Specimen

Digitalcamera

Fig. 3 CTOA test setup.

7/29/2019 Exploring Methods for Measuring Pipe Weld

3/3

JUNE 200750

the blunting at the fatigue precrack tip de-creases and crack growth begins. TheCTOAc approaches a constant value atabout 5 mm. In this region the fractureorientation changes from being perpen-dicular to the loading axis to having anangle of about 45 deg to the axis of load-ing (shear fracture mode). This is referredto as a transition from flat-to-slant frac-ture, and typically occurs at a crack length

of about 1.5 times the specimen thickness.In this transition region, the CTOAc ap-proaches a constant value, which is asso-ciated with steady-state crack growth; itis this CTOAcvalue that is used to char-acterize the resistance to crack growth fora material.

The results in Fig. 6 show that thesofter heat-affected zone (HAZ) hashigher fracture resistance than the basemetal (HAZ CTOAcvalue of 9.9 1.5 degcompared with 4.2 0.5 deg for the basemetal). In fact, when the crack propaga-tion reaches the weld interface, the crack

propagation stops for an instant as the

CTOAc increases, then the crack jumpsinto the weld metal. The same phenome-non occurs as the crack reaches the inter-face on the other side of the weld, andthen the crack jumps into the base metal.The results also show that the weld zoneexhibits a slight decrease in fracture re-sistance, as compared with the base metal.The CTOAcvalue decreases from 4.2 0.5 deg to 2.7 0.5 deg as it leaves the basemetal and enters the weld.

Summary

The fracture toughness behavior ofX100 pipeline steel through a girth weldsection was investigated using a modified

double cantilever beam specimen. A testtechnique for direct measurement of thesteady-state CTOA was presented. Opti-cal imaging was used to record the uni-form deformation of a crack edge on aspecimen surface. The CTOA was deter-mined during steady-state crack growthby a direct measurement method usingsoftware developed for this study. The re-sults demonstrate a slight improvementof the fracture resistance in the HAZ anda slight decrease in the weld metal. Cracktip opening angle is a very promising andconvenient fracture criterion for the as-

sessment of ductile fracture resistance inbase metals and welds. Further study is

warranted on a variety of matching andmismatching welds and base metals.

Acknowledgments

The support of U.S. Department ofTransportation, Pipeline and HazardousMaterials Safety Administration, is grate-fully acknowledged.

BP Exploration is also gratefully ac-knowledged for contributing the pipeline

steel.

The efforts of Hans Windhoff, Olym-pus Soft Imaging Solutions Corp., helpedto develop the software used in this study.

References

1. Newman, J. C. Jr., and James, M. A.2001. A review of the CTOA/CTOD frac-ture criterion why it works! In 42ndAIAA/ASME/ASCE/AH/ASC Structures,Structural Dynamics, and Materials Con-ference and Exhibit. Paper No. AIAA-200-1324, Seattle, Wash.: 1042 to 1051.

2. Hashemi, S. H., Gay, R., Howard, I.C., Andrews, R. M., and Yates, J. R. 2004.Development of a laboratory test tech-nique for direct estimation of crack tip

opening angle. In 15th European Confer-ence of Fracture, Stockholm, Sweden.

3. Harrison, J. D., Dawes, M. G.,Archer, O. L., and Kamath, M. S. 1979.The COD approach and its application to

welded structures. InElastic-Plastic Frac-ture. ASTM STP, Vol. 668. Landes, J. D.,Begley, J. A., and Clarke, G. A., editors,

American Society for Testing and Materi-als: 606 to 631.

4. Schwalbe, K.-H., Newman, J. C. Jr.,and Shannon, J. L. Jr. 2005. Fracture me-chanics testing on specimens with low con-straint-standardisation activities within

ISO and ASTM.Engineering Fracture Me-chanics 72: 557 to 576.

Fig. 4 Crack profile traced manually forCTOA determination.

Fig. 5 Method for determining the CTOA.

Fig. 6 CTOAcresistance curve for X100steel through a weld section.

Change of Address?

Moving?

Make sure delivery of your WeldingJournal is not interrupted. Contact theMembership Department with yournew address information (800) 443-9353, ext. 217;smateo@aws.org.