Using fracturemechanics to determine the required inspection levelBy Walter J. Sperko, P.E.

Anyone who fabricates pressure pip-ing (or almost any construction) by weldingknows that various codes and customerspecifications limit the extent of cracking, in-complete penetration, incomplete fusion,undercut, reinforcement, porosity, and slagpermitted in a weld.

Codes and customers also specify themethod of detection of such defectsvisualinspection is always required, and some-times it is supplemented by radiographic,ultrasonic, magnetic particle, or liquid pene-trant examination.

Where do these requirements and crite-ria originate? How do they relate to thedesign?

The inspection requirements and accept-ance criteria found in the American Societyof Mechanical Engineers (ASME) pipingcodes were established based on the com-bined opinion of many distinguished engi-neers who sat on various committees manyyears ago.

There is no real engineering basis for theflaws that are permitted in todays pipingcodes; the acceptance criteria in the codes

How good do piping system welds really need to be?

are more weld quality control standardsthan anything. In fact, nobody couldhave established engineering-based cri-teria as recently as 20 years ago, becausethe tools that are available today werenot available then.

Today, an engineer can determinehow good a weld really needs to be byusing fracture mechanics analysis. Frac-ture mechanics traditionally has been afairly esoteric discipline, but because ofthe need to put this tool to use in thestructural steel, nuclear, petrochemical,and chemical processes industries, ithas been simplified and made moreuser-friendly.

Fracture mechanics allows an engi-neer to consider the size of a flaw at anylocation, the stress across the flaw, andthe fracture resistance (i.e., the tough-ness). Before discussing the availabletools, however, it is important to reviewa few case studies to see how a coupleof flawed piping joints behaved undersome severe conditions.

Flawed PipeJoint Failures

Piping Leaks. In 1994, this writerwas asked to help resolve a disputebetween an engineer who designed apiping system and the owner of the sys-tem. The steel piping system distributedhot water at 400 degrees Fahrenheit and600 pounds per square inch (psi) toprovide space and hot water heating ona local government-owned campus ofbuildings.

The original underground pipingsystem was found to be leaking shortlyafter its first year of operation in the1960s, and it has leaked chronically invarious locations ever since. The leakingsuperheated water provided an attractiveYellowstone-like campus environmentcomplete with smoldering geysersfor the residents and the employees.

The replacement piping system alsostarted leaking within one year ofinstallation. The essential problem withthe system was lack of control over thewater chemistry, which caused corro-sion fatigue throughout the systembut attacked the piping preferentiallyat the welds.

Figure 1 shows a view lookingdownward into a valve pit of the pip-ing system. The leaking steam is obvi-ous and typical of all the steam leaks.

Catastrophic failure never occurred,although the size of the leaks increasedover time. Regular repairs were made be-cause the rate at which make-up watercould be put into the system was lim-ited, and, when a leak occured inside abuilding, the heat from the leaking steameventually became unbearable.

Fracture mechanics analysis can ex-plain this behavior by showing that, al-though the stress was high and the flawswere large, steel is very tough at 400 de-grees Fahrenheit, and catastrophic fail-ure will not occur.

Incomplete Penetration. The sec-ond case study was on a chilled watersystem in a 60-story office building. Al-though radiography of the system wasnot part of the contract or code require-ments, the customer radiographed manywelds and found incomplete penetration,which he believed to be unacceptable.

A demonstration was conducted inSeptember 1992 to show that circum-ferential pipe butt welds can have sig-nificant weld imperfections and still beentirely suitable for typical chilled watersystem service conditions. The entire

process was videotaped for future pre-sentation to a jury because few jurorscould be expected to understand frac-ture mechanics.

The demonstration assembly wastwo 10-foot lengths of 24-inch Amer-ican Society for Testing and Materials(ASTM) A53 Grade B seamless stan-dard weight (38-inch-thick) pipe withcaps welded on one end of each piece(see Figure 2). These lengths werewelded together so that a bad weldexisted in the middle of the assembly(see Figure 3).

The objective was to create a weldwith incomplete penetration about 25percent of the wall thickness in deptharound the entire circumference of thepipe.

This weld condition was verified byboth radiographic and ultrasonic in-spection. Both inspections indicatedincomplete penetration in the root,and ultrasonic inspection measured itas being between 18 and 332 inch deep.

The next step was to pressurize theassembly until it failed and observewhether the assembly failed at the im-perfect weld or elsewhere.

Engineering theory says that thestress in the pipe caused by pressure istwice as high in the circumferential or

Figure 1This photo was taken looking down into a valve pit of the piping system. The leakingsteam is typical of all the steam leaks.

hoop direction as it is in the axial di-rection. This means that, theoretically,the pipe will split down its length whenoverpressurized rather than across theimperfect circumferential weldprovid-ed the imperfection is not too large.

Fracture mechanics predicted that theincomplete penetration could be as deepas 40 percent of the pipe wall thickness,and failure would still occur in the hoopdirection rather than across the bad cir-cumferential weld.

Piping in operation has both pres-sure stress and bending stress caused bythe weight of the pipe and its contentsas it hangs between supports. Bendingstresses also result from thermal growthof a pipe that is heated or cooled whilein service. These bending stresses changeonly the axial stresses; they do not changethe circumferential stress.

In this demonstration, a small amountof bending stress was added to the pres-sure stress by supporting the 20-footassembly at each end. This resulted in abending stress of about 1,000 PSI in

the middle of the assembly, across thebad weld (see Figure 4).

This stress was compressive at thetop of the pipe. It pushed the imperfectweld together, and the stress was tensileat the bottom of the pipe, where it com-bined with the pressure stresses to pullthe weld apart.

To make the local stress across thebad weld as high as possible, the pipe wasrotated so that the deepest incompletepenetration in the bad weld was locatedat the bottom of the pipe during the test.

According to the design rules ofASME B31.1, Power Piping, and B31.9,Building Services Piping, 24-inch A53Grade B seamless pipe with a 0.375-inchnominal wall thickness may be used at amaximum pressure of about 400 PSI.Using the code safety factor of 4, thepipe should withstand at least 1,600 PSIbefore failure occurs.

Based on the actual tensile strengthof the pipe, failure was predicted tooccur between 2,000 and 2,250 PSI,depending on the exact wall thickness

of the pipe. The water temperature dur-ing testing was about 60 degrees F.

The test assembly was pressurizeduntil it failed. The burst pressure was2,250 PSI, and failure occurred in theaxial (hoop stress) direction as predictedby fracture mechanics (see Figure 5).The failure was quite dramatic, shootingwater more than 100 feet into the air.

Gross plastic deformation of the pipewas obvious during the test, mostly inthe form of swelling. The diameter of thepipe near the weld grew from 24 to 2612inches. The wall of the pipe at the pointof failure was less than 316 inch thick.

Do these results mean that weldquality in circumferential butt welds inpiping is unimportant? Absolutely not.

The stresses in piping systems arecomplex. It is possible for a piping sys-tem to have high stresses at some weldsand low stresses at others.

Are ExtraExaminationsCost-Effective?

Weld quality is a serious concernwhen stresses are high and less of aconcern when stresses are low. Thisobservation currently is not taken intoaccount by ASME codes and rarely byengineers when the extent of nonde-structive examination or the acceptancecriteria are established.

Because welds that meet radiograph-ic or ultrasonic examination require-ments cost anywhere from 25 to 200percent more than do welds that are notrequired to be so examined, radiograph-ic or ultrasonic examination of weldsthat are not highly stressed wastes timeand money unless special circumstancesjustify the added cost.

Why do these extra inspections costmore? In addition to the direct addi-tional cost of performing radiographicor ultrasonic examination, some factorsthat contribute to increased cost of mak-ing reduced-flaw welds are:

1. Fewer welders available who havethe necessary skill do the work.

2. More time used by welder to pre-pare ends and preclean.

3. More time used by welder to getperfect fit-up and alignment.

3/8 Inch

10 Feet

Good Weld

Water Inlet

Bad Weld

10 FeetAir Outlet

Good Weld24 Inches

Support Support

General Arrangement of Test Assembly

Weld Reinforcement

Weld Metal Outside of Pipe

Inside of Pipe3/32 Inch

3/8 Inch

Incomplete Penetration

Figure 2

Figure 3

4. More time used by welder tomake and prepare tack welds.

5. More time used by welder tomake root pass. (Sometimes, two weld-ers are needed.)

6. More time used by welder to getperfect layers of weld metal, includingcleaning between layers and contouringprevious layers of weld.

7. More time spent on preparing thecover pass for examination.

8. Additional supervision and/orinspection personnel to verify that thewelders are doing the job so that theexaminations pass.

Additional examinations increase afabricators or contractors labor costsfor which they should be compensat-ed. A customer who wants welds thathave been demonstrated to containfewer flaws should so specifyand bewilling to pay for what it takes tomake such welds.

In fact, the B31 Code Sections typi-cally specify something comparable tothat found in B31.1, paragraph 136.1:

The degree of examination andthe acceptance standards beyondthe requirements of this code shallbe a matter of prior agreement be-tween the manufacturer, fabrica-tor, or erector and the owner.

The codes make the engineer re-sponsible for specifying the extent andthe acceptance criteria for any nonde-structive examination, beyond thatrequired by code, that he believes arenecessary to ensure the integrity of anypiping system, and he is obligated todo that before the system is fabricatedand installed.

Uses for FractureMechanics

The stress level at any weld deter-mines the quality of weld needed, andthat, in turn, determines the extent ofinspection justified. The tool that isavailable to assist the engineer in mak-ing the correct decision about the ex-tent of inspection required and the ac-ceptance criteria is fracture mechanics.

Unfortunately, few engineers whoare involved in commercial or industri-al piping recognize that this technolo-gy is available to them, and even feweruse it in establishing the extent ofnondestructive examination to beapplied.

Fracture mechanics recognizes theconnection between the size of anyflaw, the stress across the flaw, and thetoughness of the material. Because noweldor base metal, for that matteris perfect, there is always some toler-ance for flaws.

The reason that more piping sys-tems do not fail catastrophically is thatthe stress is sufficiently low and/or thetoughness of the material is sufficientlyhigh. Most piping failures show up asleakageas was clearly the case in the400-degrees-F, 600-PSI system de-scribed earlier.

What the Codes SayThe codes recognize that every weld

has flaws, and the codes permit themto some extent in all new construction.For example:

1. ASME B31.1 and ASME SectionIII permit some slag inclusions, porosi-ty, and undercut.

2. ASME B31.3 and B31.9 permit

not only slag, porosity, and undercut,but also 112 inches of incomplete pene-tration in any 6 inches of weld lengthfor normal fluid service.

3. American Petroleum Institute

Figure 5This NPS 24 standard-weight seamlesspipe contains a bad circumferentialweld. However, as indicated by thearrow, the failure was not near the weld(under the white tape).

Bending Stress Distribution

Compressive

Tensile

Support Support

1,000 PSI

0 PSI

Figure 4In this example, a little bending stress was added to the pressure stress by supportingthe 20-foot assembly at each end.

(API) 1104 permits not only slag andporosity, but incomplete penetration upto 2 inches in 12 inches of weld length,incomplete fusion, and cold laps up to2 inches in 12 inches of weld length. Italso permits crater cracks up to 532 inchin length. Interestingly, API 1104 isnot as tolerant of undercut as the B31Code Sections.

API 1104 also has a procedure forevaluating crack-like flaws using fracturemechanics if the toughness of the steel is

known. This method includes consider-ations for fatigue and dynamic loading.

In the maintenance and repair codes:

1. British Standard PD-6493 pro-vides a generic method for evaluation offlaws in any welded structure.

2. ASME Section XI, Inservice In-spection of Nuclear Power Plant Com-ponents, Code Case N-494, provides acomplete method for evaluating flaws inpiping. This code case was written for

Class 1 piping, and it should be appliedto other piping systems only by knowl-edgeable parties.

3. The Metals Properties Council(MPC) of the Welding Research Coun-cil has developed software for analysisof flaws in welds in vessels and piping.

4. API is in the final stages of issu-ing RP-579, Recommended Practice forFitness for Service, which will containmethods for evaluating flaws in pipingand vessels.

All of these references and materialsare available to the public, except for theMPC software, which is available onlyto members of MPC.

Although the maintenance and re-pair codes are not written for new con-struction, the concepts and methodolo-gies they use certainly are reasonable foruse in determining the need to performnondestructive examinations beyondthose required by code.

To this end, the writer has prepareda concept table that relates service con-ditions to the extent of inspection (seeFigure 6). This table recognizes thatthe stresses in piping systems come fromdifferent sources (pressure, dead weight,thermal expansion, hydraulics, etc.) andthat the extent of inspection should berelated to the nature of these loads andto their magnitude, as well as to the na-ture of the fluid being handled.

This approach also recognizes thata visual inspection of pipe welds at fit-up (after the joint has been cut,beveled, and tack-welded together butbefore the root pass is put in) goes along way toward ensuring adequateweld quality. In fact, the most cost-effective inspection method is to havesomeone, in addition to the welderwho will be making the weld, inspectthe fit-up.

Taking the StepsPiping system designers should

evaluate their systems in terms of thefluids being handled, the stresses inthe system (especially at changes indirection), the number of pressure andtemperature cycles that the system will

Weld TypeService

Longitudinal Circumferential Branch FilletCondition Groove Groove GrooveHoop stress > 80% of the RT or UT N/A N/A N/AallowableHoop stress > 50% and MT or PT N/A N/A N/A 80% of the allowableSustained stresses due to N/A RT or UT VT fit-up and MT or PTcombined pressure, Backgougedead weight, or other by ownernon-self-limiting loads> 80% of the nonintensifiedallowableDynamic loads that will RT or UT RT or UT VT fit-up and MT or PTexceed 80% of the Backgougeallowable stress by ownerToxic service (when specified RT or UT RT or UT VT fit-up and MT or PTby the owner) Backgouge

by ownerWelds subject to more than N/A RT or UT VT fit-up and MT or PT7,000 thermal cycles and the Backgougecalculated stress range at the by ownerweld is > 80% or the allowableSustained stress caused by N/A Visual examination at Visual examination at N/Apressure + dead weight and fit-up by owners or fit-up by owners orany other non-self-limiting independent inspector independent inspectorloads > 50% but not > 80% of engaged by owner engaged by ownerthe nonintensified allowable stress at temperatureSustained stress caused by N/A Visual examination at Visual examination at N/Apressure + dead weight and fit-up by independent fit-up by independentany other non-self-limiting inspector engaged inspector engagedloads > 25% but not > 50% of by contractor by contractorthe nonintensified allowablestress at temperatureSustained stress caused by N/A Visual examination at Visual examination at N/Apressure + dead weight and fit-up by contractors fit-up by contractorsany other non-self-limiting in-house inspector in-house inspectorloads < 25% of the nonintensified allowable stress at temperature

Relating Service Conditions to Examinations

General Notes:(A) All welds shall be given a visual examination in addition to the type of examination specified.(B) RT - radiographic examination; UT - ultrasonic examination; MT - magnetic particle examination;

PT - liquid penetrant examination; VT - visual examination.(C) Surface examination (MT, PT, or VT) shall apply to the outside surface and inside surface, where readability

accessible, of longitudinal groove welds.(D) Weld types apply to welds in pipe and fittings.

Figure 6

Reprinted with permission from the Mach/April 1999 issue of PRACTICAL WELDING TODAY 833 Featherstone Road, Rockford, IL 61107815-399-8700 Fax:815-399-7279 Web Site: www.fmametalfab.org E-mail: info@fmametalfab.org

experience, and any unusual service oroperating conditions.

Establishing the typical stressthroughout the system and finding thehigh-stress locations are the first steps inusing a fracture mechanics approach todetermining the required weld quality.

The second step is to approximate orspecify the toughness of the material ateach set of operating conditions. Theseinclude normal operating conditions,start-up conditions, and any excursions.

The last step is to determine, throughfracture mechanics and a suitable safetymargin, the largest flaw that can be tolerat-ed before catastrophic failure will occur.

If the permissible flaw is largeenough that the pipe will leak before itfails catastrophically (similar to the firstcase study), there is little justification foradditional nondestructive examination,provided leakage of the fluid can be tol-

erated (it is not flammable, explosive,toxic, etc.). If leakage is not tolerable,then additional nondestructive examina-tions that reduce the largest expectedflaw size must be specified.

If examinations beyond those specifiedby code are necessary, the designer shouldspecify the additional requirements in thepurchase order. The designer shouldunderstand that the customer will have topay not only for the nondestructive exam-ination but also for the cost of makingwelds that have fewer and smaller flawsthan might otherwise be allowed.

The work that is currently being donein the maintenance and repair codes mayeventually filter into the new-construc-tion codes, and it is likely that somedaythe extent of nondestructive examina-tion and the applicable acceptance crite-ria found in those codes will be relatedto service conditions.

Until then, the industry must de-pend on the education, experience, andjudgment of design engineers to deter-mine the level of weld quality requiredin piping systems. lWalter J. Sperko, P.E., is President ofSperko Engineering Services, Inc., 4803Archwood Drive, Greensboro, NorthCarolina 27406, phone 336-674-0600,fax 336-674-0202, e-mail sperko@asme.org.Sperko is also the technical consultant tothe NCPWB. Sperko Engineering pro-vides engineering consulting in weld-ing, metallurgy, corrosion, and qualityassurance with specialization in pip-ing/pressure vessels.

Although Sperko is a member of severalASME and AWS committees, any codeinterpretations in this article should beconsidered his personal opinions and notthe official opinion of ASME or AWS.

ADVANTAGES OF MEMBERSHIP IN THENATIONAL CERTIFIED PIPE WELDING BUREAU (NCPWB)

As a member of the NCPWB, you will receive more than 100 Welding Procedure Specifications (WPSs) and BrazingProcedure Specifications (BPSs) that your company can adopt without going through the travail of qualifying the WPSsor BPSs. Writing and qualifying WPSs and BPSs is a trial and error process that is both time consuming and expensive.

The NCPWB WPSs and BPSs are reviewed and accepted by the Hartford Steam Boiler Inspection and InsuranceCo., which has verified that they are in conformance with ASME Boiler and Pressure Vessel Code,

Section IX, Welding and Brazing Qualifications.

Membership in the NCPWB also permits the interchange of welders who have been qualified by other NCPWB mem-ber contractors. This allows your company to employ a welder whose qualification record is on file at any one of theNCPWB local chapters without further qualification.

Membership in the NCPWB also permits the interchange of welders and brazers who have been qualified by otherNCPWB member contractors. This allows your company to put to work immediately any new welder whose qualifi-cations are on file at any local NCPWB office without further testing.

The cost of NCPWB membership is a one time initial fee plus minimal annual dues. The cost of qualifying one welderor brazer or the cost of writing and qualifying one WPS or BPS far exceeds the annual dues fee. This has the potentialfor tremendous cost savings that you can pass on as reduced project cost increasing the business for your company.

To learn how your company can become a member of the National Certified Pipe Welding Bureau (NCPWB), call(301) 869-5800.