8/9/2019 506062

1/43

FAILURE BEHAVIORO FFLAWED CARBON STEEL PIPES AND FITTINGS

M. 6. Reynolds

Approved: 5 (S. R . VandenbergProject Engineer

Approved:

6169Dev. Eng.-XXX250-MS-10170

Prepared for theU. S. Atomic Energy CommissionContract AT(04-3)-189Project Agreement 37

GEAP-10236AEC Research and

Development ReportOctober 1970

D . H. Imho ff, ManagerDevelopm ent Engineering

~-.i

This report was preparad as an account of worksponsored by the United States Government. Neitherthe United States no r the Un ited Stat es Atomic EnergyCommission, no r any of their employees, nor any oftheir contractors, subcontractors, or their employees,makes any warranty, express cr implied, or assumes any Ilegal liability or responsibility for the accuracy, com-_ _ _ ~

AT OMIC P OWER E QUIP ME NT DE P ART ME NT *GE NE R AL E L E CT RIC COMP ANYSAN JOSE, CALIFORNIA 95125

G E N E R A L @ E L E C T R I C

8/9/2019 506062

2/43

DISCLAIMER

This report was prepared as an account of work sponsored by anagency of the United States Government. Neither the United StatesGovernment nor any agency Thereof, nor any of their employees,makes any warranty, express or implied, or assumes any legalliability or responsibility for the accuracy, completeness, orusefulness of any information, apparatus, product, or processdisclosed, or represents that its use would not infringe privatelyowned rights. Reference herein to any specific commercial product,process, or service by trade name, trademark, manufacturer, orotherwise does not necessarily constitute or imply its endorsement,recommendation, or favoring by the United States Government or anyagency thereof. The views and opinions of authors expressed hereindo not necessarily state or reflect those of the United StatesGovernment or any agency thereof.

8/9/2019 506062

3/43

DISCLAIMER

Portions of this document may be illegible inelectronic image products. Images are producedfrom the best available original document.

8/9/2019 506062

4/43

LN-1(5171

NOTICEThis report was prepared as an account of work sponsored by the United StatesGovwnment. Neither the United States nor the United States Atomic Energy Com-mission, nor any of their employees, nor pny of their contractors subcontractors,or Their employees, makes any warranty, express or implied, or assumes any legalliability or responsibility for the accuracy, completeness or usefulnessof any infor-mation, apparatus, product or process disclosed, or represents that i t s use wouldnot infrinqe privately owned rights

8/9/2019 506062

5/43

GEAP-10236AE C Research and

Development Repo rtOctober 1970

F AIL URE BEHAVIO ROFFLAWE D CARBON STEEL PIPES AND F ITT ING S

M. 6. eynolds

Approved :S. R. VandenbergProject Engineer

D . H . Imh off, ManagerDevelopment Engineering

Prepared for theU. S. Atomic Energy CommissionContract AT (04-3)-1 89Project Agreement 37

I

ATOMIC POWER EOUIPMENT DEPARTMENT GENERAL ELECTRIC COMPANYSAN JOSE, CALIFORNIA 951256169-Dev. ng.-XXX250-MS-10/70

G E N E R A L @ E L E C T R I C

8/9/2019 506062

6/43

LN-1(5171

NOTICEThis report was prepared as an account of work sponsored by the United StatesGovwnment. Neither the United States nor the United States Atomic Energy Commission, nor any of their employees, nor m y of their contractors subcontractors,or their employees, makes any warranty, express or implied, or assumes any legalliabil ity or responsibility for the accuracy, completeness or usefulnessof any infor-mation. apparatus, product or process disclosed, or represents that i t s use wouldnot infringe privately owned rights

8/9/2019 506062

7/43

GEAP-10236

TABLE OF CONTENTS

1 . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 General Testing Plan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2.1 Pipes with Long Axial Part-Through Cracks . . . . . . . . . . . . . . . . . 22.2 Behavior of Pipes with Axial Through-Wall Flaws . . . . . . . . . . . . . . . 32.3 Pipes with Short Part-Through Flaws . . . . . . . . . . . . . . . . . . . 42.4 Circumferentially Flawed Pipes Under Bending Stresses . . . . . . . . . . 4

3. Experimental Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . 64. Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

4.1 Pipes with Axial Through-Wall Flaws . . . . . . . . . . . . . . . . . . . 104.2 Pipes with Axial Part-Through Flaws . . . . . . . . . . . . . . . . . . . 104.3 Pipes with Circumferential Flaws . . . . . . . . . . . . . . . . . . . . . 17 4.4 Failure Behaviorof Flawed Tees and Elbows . . . . . . . . . . . . . . . . . 17

5. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Acknowledgment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

http://0.0.0.0/http://0.0.0.0/http://0.0.0.0/http://0.0.0.0/

8/9/2019 506062

8/43

GEAP-10236

LIST O F IL L UST RAT IO NS

Figure Title Page

67

89

101112131415161718

4

Axial Fracture in Circumferentically Flawed Pipe . . . . . . . . . . . . . . . . .Flaw Configurations in Fit ting Specimens . . . . . . . . . . . . . . . . . . . .Toughness Anisotropy of ASTM A1066 Pipe . . . . . . . . . . . . . . . . . . 9

57

Typical Failure Behavior in Hydrostatic Tests of ASTM A1066 Pipe . . . . . . . 1Variation o f Pressure Load Limit with Flaw Length in ASTM A1066 Pipe with AxialThrough-Wall Flaws . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12Pressure Load Limit Versus c/R for ASTM A1066 Pipe with Axial Through-Wall Flaws . . 12Effective Critical Stress Intensity Factor for ASTM A1066 Pipes Containing AxialThrough-Wall Flaws . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13af /aa Versus Crack Half.Length. ASTM A1066 Pipe . . . . . . . . . . . . . 14Failure Diagram for ASTM A1066 Pipe. Part-Through Flaws . . . . . . . . . . . . 5Comparison of Inside with Outside Flaws . . . . . . . . . . . . . . . . . . . . 16Test Assembly. Schematic . . . . . . . . . . . . . . . . . . . . . . . . . 19Pipe wi th Through-Wall Flaw. Bending Only . . . . . . . . . . . . . . . . . . . 0Pipe with Circumferential Part-Through Flaw. Bending Only . . . . . . . . . . . . 1Fracture Appearance in Tee Tested to Failure . . . . . . . . . . . . . . . . . . 3Gross Failure Appearance. Unflawed Elbow . . . . . . . . . . . . . . . . . . . 3Gross Failure Appearance. Unflawed Tee . . . . . . . . . . . . . . . . . . . . 24Failure Appearances. Flawed Elbows . . . . . . . . . . . . . . . . . . . . . 25Failure Appearances. Flawed Tees . . . . . . . . . . . . . . . . . . . . . . 26

LIST OF TABLES

Table Title PageI Room Temperature Tensile Propertiesof ASTM A106B Pipe Used in Pressure

Load Limi t Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 Pressure Load Limits for Circumferentially Flawed Pipes . . . . . . . . . . . . . 183 Behavior of Circumferentially Flawed Pipes under Bending and Pressure Stresses . . . . 9

Pipe Fitting Test Data . . . . . . . . . . . . . . . . . . . . . . . . . . 2+/-vi-

8/9/2019 506062

9/43

GEAP-10236

ABSTRACTThe pressure load lim it for a cylindrica l shell containing a flaw is a function of cylinderradius and wall thickness, flaw dimensions, and properties of the material. The strength oflarge diameter thin-wa ll cylinders of brittle material containing axial through-wall flawscan be predicted by linear elastic fracture mechanics if co rrections for eff ect o f curvatureare made. Expressions to predic t pressure load limits in more du ctile materials becomemore empirical. Such expressions have been developed by several investigators.Pressure load lim its for cylinders containing full-length , axial, part-thr ough flaws can bepredicted by equating the effec tive stress in the ligame nt at the base of the flaw to theplastic ins tabil ity stress for the materia l. Load lim its for cylinders having short par t-through flaws are predicted from a failu re diagram in which relative burst pressure isplotted against the reciprocal of flaw length.Results of hydrostatic tests at 60F on flawed ASTM A 106B pipes in nom inal diametersranging from 4 to 12 inches are presented. Results of tests on pipes with axial through-wall flaws are compared with empirical equations. Results of burst tests on flawed teesand elbows are presented.

1. INTRODUCTIONThe objective of the Reactor Primary Coolant System Rupture Study is to determine the modes by which reactor

piping might fail, the processes leading up to the failure, and finally the numerical probabili ty that such a failure mightoccur. Two distinct kinds of failures must be recognized (1) a leak resulting from a crack or other flaw in the wallof apipe or fitting, and (2) sudden fracture failure under monotonic load as a result of a flaw in a pipe or fitting. The mechanismby which a flaw i s produced in a pipe or fit ting i s outside the scope of this report; it shall deal only with the failure of flawedstructures under monotonic load and the relationship between failure load and flaw size.

The presence of a flaw in any stressed structure, such as a pipe, may be expected to reduce the maximum loadwhich it can support. The design load for an unflawed structure i s commonly well below the value a t which generalyielding would occur. A large enough flaw in any structure can cause a failure a t a below-yield load by virtue of simpleloss of load-bearing cross section; how large a flaw is required depends on the shape of the flaw and the deformationproperties of the material. I f the material is brittle, a relatively low load may be enough to bring the local stress a t the tip of asharp flaw to the frac ture level and cause the structure to fail as a result of unstable fracture propagation. If the flawconfiguration i s geometrically simple enough that the elastic stress field about i t s tip can be determined theoretically orexperimentally, the gross stress a t which failure will occur can be estimated in terms of a single material constant. Thisconstant is the "plane strain fracture toughness", Klc, of linear elastic fracture mechanics. Within the range of applicability oflinear elastic fracture mechanics, the gross failure stress i s almost entirely dependent on material properties, flaw size andflaw configuration, and i s but slightly dependent on the size of the structure.

I f a sharp flaw exists in a ductile material, an increasing load causes plastic deformation to occur first in the region ofthe highest local stress a t the ti p of the flaw. Deformation tends to reduce the acuity o f the flaw ti p and consequently thelocal stress concentration. Hence, a larger flaw i s required to producea failure a t a given stress level in a ductile material thanin a brit tle one.

For brittle materials, the mathematics of linear elastic fracture mechanics predict that the product of flaw diameter orlength and the square of applied stress a t fracture should be a constant. This prediction has been experimentally verified forsimple flaw configurations such as a semi-elliptical surface crack in a large plate under tension normal to the crack plane, anedge crack in a plate under tension, or a through-wall crack in a plate under tension. I f the material i s somewhat ductile (bu tnot so ductile that plastic deformation extends very far from the flaw tip before failure occurs), the fracture failure stress cans t i l l be estimated by correcting the macroscopic flaw length by an amount comparable to the size of the plastic zone a t theflaw tip. Since this correction i s a function of the ratio of the fracture toughness of the material to i t s yield stress, anothermaterial constant has been introduced into the estimation of the failure stress and some of the simplicity of linear elastic

-1

8/9/2019 506062

10/43

GEAP-10236

fracture mechanics i s lost. The technique i s useful, however, for dealing with ductile materials in sections of thicknessexceeding the size of the plastic zone. For ductile materials in thin sections, linear elast ic fracture mechanics i s of limitedvalue in predicting failure stresses under monotonic loading, and empirical techniques must be used.

2. GE N E R A L TE S TI N G P LA NIn the following paragraphs, some experiments on the failure behavior of ASTM A1068 stee l pipes and fittings

containing different kinds of flaws are discussed. The use of empirical equations to predict failure pressures of flawed pipes i salso considered.2.1 PIPES WITH LONG AXIAL PART-THROUGHCRACKS

A pipe containing a long (several pipe diameters) ax ia l part-through flaw of constant depth may be consideredanalogous to a series-connected pair of cylindrical tensile elements. In this analogy, the ligament remaininga t the base of theflaw would be represented by the smaller, plastically deforming member of the pair; the unflawed portion of the pipe wouldbe represented by the larger, elastically loaded element. The load maximum for a two-element structure o f this k ind coincideswith the onset of plastic instabili ty in the smaller of the two elements. This coincidence suggests that a pipe containing a longpart-through flaw should fail when plastic instability occurs in the ligament a t the base of the flaw.

To apply this hypothesis to a pipe containing a long part-through flaw, it i s necessary to estimate the three principalstresses effective in the ligament a t the base of the flaw and the plastic instability stress for the material under triaxial stressconditions. By denoting ax ia l , circumferential, and radial stresses in a pipe under internal pressure b y u I ,u2 , and 0 3 ,respectively, the average principal stresses in the ligament can be approximated in terms of the applied nominal hoopstress oh:

and V0 3 2:- (u, + u2)2

( l a )

where :T = unflawed wall thickness,x = thickness of the ligament,v = elast ic Poisson ratio for the material.

The ratio of the nominal hoop stress based on unflawed pipe dimensions to the effective stress* in the ligament can becalculatedas a function of x/T by using the principal stress values given in Equation (1 1.

The estimate of the plastic instability stress in the ligament i s rather arbitrary. However, low carbon steels, such asASTM A106B, commonly exhibit a rather sharp tensile yield point, and it i s assumed that the effective instability stress forsuch a material i s approximately equal to i t s uniaxial tensile yield stress.

The validity of the plastic instability hypothesis has been tested with experimentally determined burst hoop stressvalues for ASTM A106B pipes containing long part-through flaws.' Ratios of burst hoop stress to uniaxial tensile yield stresshave been compared with hoop stress to effective stress ratios calculated by using the principal stress values given inEquation (1). Considering the rather arbitrary choice of a value for 0 3 and the effective instability stress, reasonableagreement was obtained over a considerable range of x/T.

1*Effective stress 5 s defined by i = (al - u2 2 + (u - O3l2 + ( 0 2 - U3l2

-2 -

8/9/2019 506062

11/43

GEAP-10236

2.2 BEHAVIOR OF PIPES WITH AXIA L THROUGH-WALL FLAWSto the crack plane i s given by the relation

The failure stress for a wide plate of brit tle material containinga through-wall crack and under uniaxial tension normal

n u 2 c = Klc2 (2)where:

u = applied stress,c = crack half-length, and

Klc = material constant.The so-called plane strain fracture toughness or critical stress intensity factor i s defined by linear elastic fracture mechanics2I f the material s not completely brittle, Equation (2) must be corrected by a term which i s a function of the ratio of Klc to theyield stress of the material. Dimensionally, the equation i s unchanged; i.e., the limiting stress that the plate will sustain variesinversely as the square root of the crack length. The fracture mechanics analysis indicates that a tensile stress componentparallel to the crack should not affect the value of the instability stress normal to the crack plane; thus, a crack in a very largediameter thin-wall cylindrical shell under internal hydrostatic pressure should approach the behavior of the crack in a widethin plate under uniaxial tension equal to the hoop stress. However, as the diameter of the shell decreases and crack lengthincreases, it i s necessary to take into account bending stresses about the crack. Anderson3 used dimensional reasoning anddeveloped an expression of the form

where:Kc = nominal stress intensity factor,UB = biaxial yield stress,

c = crack half-length,R = cylinder radius,u = membrane hoop stress a t instability, andq = "bulge coefficient" dependent on material properties and cylinder radius.

Irvine, Quirk, and Bevitt4 used experimental data from t e s t s on 5-foot-diameter pressure vessels and concluded that the loadlimit for a thin-wall cylinder under internal pressure was given by an expression of the form

u3 c2 = K (4 )where K i s a constant dependent on ultimate strength, yield strength, and the Charpy impact energy of the material. Theform of the relationship between this constant and the material properties of the material was dependent on the fracturemode observed.

fi be r and co-workers' developed an expression based chiefly on an earlier work of Folias,6 which takes into accountboth bending stresses and plastic zone correction and gives an effective critical stress intensity factor or crack toughness foraxially cracked pipes in terms of hoop stress and crack length. This expression i s of the form

where:K = stress intensity factor, ksi J in.,c = crack half-length, inches,

R t

-3-

8/9/2019 506062

12/43

8/9/2019 506062

13/43

GEAP-10236

F I n t

I n t/ 0.80 + -& 7.12 + 1.08 i)where:

F = applied tensile load,M = applied bending moment,a = ligamentdepth,t = flawdepth,c = pipe inside radius,4 = angular coordinate,

a + t = pipe wall thickness, andKI = opening-mode stress intensity factor.

Gilman' has developed an expression for the stress intensity factor Kc in a pipe containing a short circumferentthrough-wall law and under a combined pressure and bending stress:

where:c = crack half-length,P = pressure a t failure,

M = bending moment a t failure,t = pipe wall thickness,R = pipe mean radius,

Ro = pipe inside radius, andfl , 2 = correction factors for bulging.

There i s some uncertainty as to the values of fl and f2 to be used in the calculation, but appropriate values for thcrack lengths in question can be estimated from the Folias" relation with an error of probably less than 10%.

-5-

8/9/2019 506062

14/43

Figure 1. Axia l Fracture in C ircum ferentically Flawed Pipe

3. EXPERIME NTAL PROCEDURESThe work to be reported has been limited to specimens of ASTM A1066 pipe in sizes ranging from 4 to 12 inches

nominal diameter and to ASTM A2346 tees and elbows of 6-inch Schedule 80 size. Pipe specimens consisted of segments oflength greater than four diameters closed a t the ends with standard welding caps of the same material. Fit ting specimens werewelded to 1-foot segments of A1066 pipe, which were, in turn, welded to standard welding caps. Pipe specimens containingboth through-wall and internal and external part-through flaws were tested. Fitt ing specimens were limited to externalpart-through flaws in the configurations shown in Figgre 2.

Part-through flaws in pipes were cut chiefly with a 2.75-inchdiameter, 45 degree Vee milling cutter with a 0.010-inchtooth t ip radius. Earlier in the program, short part-through flaws were cut in some specimens with slit ting saws or abrasivewheels ranging from 0.015 to 0.020 inch in thickness. Full-length axial flaws have been cut with mil ling cutters ranging from0.020 to 0.188 inch in thickness.*

Axial through-wall flaws of varying lengths were cut in the centers of the specimens with a saber saw. The ends of theflaws were sharpened either by hand-broaching o a ti p radius of approximately 0.001 inch or by cyclic pressurizationof thespecimen un ti l visible fatigue crack propagation had occurred. Before welding the end caps on these specimens, each flaw wassealed with a laminated patch consisting of a layer of 1/32-inch-thick annealed aluminum next to the pipe wall and extendingapproximately 2-1/4 inches beyond the flaw in each direction, a somewhat smaller piece of 1/16-inch-thick annealed mildsteel, and finally a piece of 1/16-inch-thick Neoprene extending about 1 inch beyond the edges of the aluminum. The patchwas formed to the proper curvature and assembled outside the pipe with General Electric RTV 102 cement between metallayers and RTV 90 cement between the Neoprene and the metal. After curing, the completed patch was cemented in placewith RTV 90 cement between the Neoprene sheet and the pipe wall. While the cement was curing, the patch was held in placein the pipe by an inflated rubber bladder, and the pipe was warmed slightly wi th an infrared lamp.

The specimens were tested hydrostatically with water or water-ethylene glycol mixtures as the pressurizing fluid.Hydrostatic pressure was generated wi th an airdriven, piston-type booster pump with a piston area ratio of 300 to 1. Anelectrical contact bu il t into the end of the driving cylinder closed when the piston was fu lly retracted. Pressure applied to thespecimen was measured by a strain-gage-type pressure transducer connected through an appropriate signal conditioner to astrip-chart potentiometer recorder. The chart paper on the recorder was advanced by a solenoiddriven ratchet motoractuated by electrical pulses from the contact on the pump. It was found that the volume of water delivered per stroke of thepump was essentially constant over the pressure range covered in the tests. Since chart motion was proportional to thenumber of pump strokes, the recorder produced a plot of applied pressure against volume o f water delivered t o the specimen.I f the compressibility of water i s neglected, the chart record amounted to a plot o f load versus deformation for the specimen.In early experiments of the tes t series, specimen deformation was measured by post-yield strain gages attached in a cir-* Burst hoop stress was found t o be strongly dependent on flaw depth and rather insensitive to flaw width.

I -6-

8/9/2019 506062

15/43

GEAP-10236

. '1 -

.i'YPES OF TEE F L A W S

TOOL RVULIUT--, N r

Figure 2. Flaw Configurations n Fitting Specimens

-7-

8/9/2019 506062

16/43

GEAP-10236

cumferential direction to the mid-portion of the pipe specimen approximately 90 degrees from the flaw. After developmentof the volume recorder, strain gages were not used, and general plastic deformation of the specimen was determined by pre-and post-test measurements of pipe circumference. Unflawed pipes were tested to obtain load maxima in the absence ofdefects. ,Material

ASTM A106B seamless pipe i s made of s teel of 0.30% nominal carbon content, and i s neither isotropic nor par-ticularly uniform with respect to mechanical properties. The material used in these tests was intended to be representative ofroutine manufacturing practice; no effort was made to select it on a basis of special properties. Mechanical tests of thematerial were limited to tensile tes ts and Charpy impact fracture t e s t s with either full-size specimens or subsize specimenswhen specimen size was limited by pipe wall section thickness. Figure 3 contains data from a single length of 6-inch Schedule80 pipe which are illustrative of the toughness anisotropy of this material. The room temperature tensile properties of thematerial used and the burst hoop stress for unflawed pipe of each size are given in Table 1.

Size (inch)and Schedule

12-8010-4010-4010-8010-808-408-408-808-806 4 06-406-80(a)6-80 a16-80(b)4-1604-804-403-1603-802-1602-80

Table 1ROOM TEMPERATURE TENSILE PROPERTIES OF ASTM

A106B PIPE USED IN PRESSURE LOAD LIMIT TESTSSpecimenDirection

CLCLCLCLCLCLCCLLLLLLL

0.2% YieldStress (ksi)

45.838.938.238.448.948.850.341.344.641.243.839.745.350.037.444.243.338.045.049.452.2

UltimateStress (ksil

68.664.365.072.273.574.676.267.368.161.I62.665.566.567.370.072.868.862.964.280.273.5

C = Circumferential (a) and (b) indicate two distinct heats in this sizeL = Longitudinal

PercentElongation

31 O35.030.432.226.528.427.929.626.037.030.434.027.425.533.732.232.732.733.427.525.6

Burst HoopStress,* (ksi)

61.256.366.069.462.262.0

58.865.4

--------

57.7----

Fo r an un flawed p ipe specimen based on original dimensions.

-8-

8/9/2019 506062

17/43

G EAP-10236

250

200h2B#3

k$ 15Cwh3.eu

lo (

5c

\ W / ! pecimen O rientation I

0 100 zu u JU U-200 - 100Tem pera t ure (OF)

\

Figure 3. Toughness An isot rop y of A S T M A 106B Pipe

-9

8/9/2019 506062

18/43

GEAP-10236

4. EXPERIMEMVAL RESULTS4.1 PIPES W I T H AX I AL T HRO UG H- W ALL F L AW S

Twenty-five specimens of A106B pipe in various diameters and wall thickness with axial through-wall flaws of differentlengths were tested. Typical failure appearances are shown in Figure4. A plo t of maximum hoop stress versus flaw half-lengthfor six different pipe sizes i s presented in Figure5. I f the load maximum was not reached first, general yielding occurred a thoop stresses equal to approximately 1.I times the uniaxial tensile yield stress, as would be predicted from theory. Therewas no evidence that the behavior of fatigue-sharpened flaws differed from that of the hand-finished flaws. Data points foreach size of pipe are consistently above or below the drawn curve. The relative position of these points with respect to thecurve cannot be accounted for by the corresponding heat-to-heat variation in mechanical properties as given in Table 1. I fmaximum hoop stress i s plotted against flaw half-length divided by pipe mean radius, the spread of data points about thedrawn curve i s reduced somewhat (Figure 6). The dotted line in Figure 6 indicates the code allowable stress for ASTMA1066 pipe a t room temperature. The intersection of this line with the curve indicates that a flaw of length approximatelyequal to the pipe diameter is needed to reduce the load maximum under monotonic loading to the level of the code allowablestress.

Values of effective critical stress intensity factor or fracture toughness were calculated from the load maximum data byusing Equation (5). hese values are plotted against flaw half-length in the upper part of Figure 7. The increase in K withincreasing flaw length i s not surprising since the authors of the equation make no claim for i t s validity a t large valuesof c2/Rt. By experimentation it was found that an empirical equation of the form

greatly reduced the variation in K with flaw length. Values of K calculated by using Equation (11 ) are shown in the lowerpart of Figure 7. Again, there i s no obvious correlation between the scatter in the vaiues of K about the mean value of95.6 ks i J in. and the heat-to-heat variation in either tensile yield stress or the burst hoop stress for pipe in the unflawedcondition.

Flow stress (of) was calculated from the t e s t data with Equation (8). A plot of these values, divided by the average ofthe yield and ultimate tensile stresses, versus flaw half-length i s shown in Figure8.Comparison with Figure 7shows that thescatter of data points about the drawn curve is about the same in each case.4.2 PIPES WIT H AXIAL PART - T HRO UG H FLAWS

Results of tests on pipes containing short part-through external axial flaws are shown in Figure9. In this figure,pressure load limit or burst pressure, expressed as a fraction of t h a t for an unflawed pipe is plotted against the reciprocal ofcorrected flaw length." The upper dotted line represents the unflawed condition, for which x/T equals un ity. ** The lowercurve represents the behavior of pipes with through-wall ax ia l flaws for which x/T i s equal to zero. Load maxima for pipeswith full-length flaws (1/L 0) are plotted along the ordinate a t the left of the diagram a t heights dependent on the valueof x/T; a l l points representing load maxima for pipes having short part-through flaws l i e between the curve and the upper linea t locations dependent on the x/T value for the particular flaw. Flaws with low values of x/T l i e near the curve; high values(shallow flaws) give points near the upper line. In Figure 9, dotted lines have been drawn to give estimates of load limit valuesfor x/T equal to 1/4, 1/2, and3/4, espectively. As was the case for through-wall flaws discussed above, there is considerablescatter in the data points although the trends indicated by the dotted lines are evident. To eliminate the effect of pipe-to-pipeproperty variations, two 20-foot lengths of pipe were cut into t es t specimens. Specimens wi th both inside and outside flawswere cut from each pipe as was one unflawed control specimen. Test data from these two pipes, designated as numbers 18and 19 , are given in Figure IO, as are the data from inside-flawed specimens from another pipe. The improvement in* This correction was necessary because flaws were machined with a round cutter and were not of full depth a t the ends.**Here T represents the pipe wall thickness and x the thickness of the ligament a t the base of the flaw.

Corrected length i s length of a flaw having ful l depth a t the ends and subtending area equal to machined flaw.

-10-

8/9/2019 506062

19/43

-1

8/9/2019 506062

20/43

I I I I I I I I I I0 . 4 0.6 0 .8 1.0 2. 0 3. 0 4. 0 5. 0 6. 0Cra ck Half-Length, c (inches)R/t Rt- -iamet er Schedule R t- -V 4 40 2. 10 0.23 9. 15 0.48

A 6 40 3. 20 0.28 11.4 0. 9000 6 80 3.12 0.43 7.26 1.340 8 40 4.17 0.32 12.8 1.35X 10 40 5.22 0.36 14. 5 1.880 12 80 6.04 0.71 8. 51 4.28

Figure 5. Variation of Pressure Load Limit with Flaw Length inAS TM A 106B Pipe with Axial Through-Wall Flaws

L I I I I I0.2 0 . 3 0 . 4 0.5 1.0 2c/R

R/t Rt- -Diam eter Schedule R t-V 4 40 2. 10 0.23 9. 15 0.483.20 0.28 11.4 0.906 40

6 80 3. 12 0. 43 7. 26 1. 344. 17 0. 32 12.8 1. 358 40

X 10 40 5. 22 0. 36 14. 5 1. 886. 04 0. 71 8. 51 4. 2812 80

00

Figure 6. Pressure Load L im it Versus c/ R for AS TM A 106BPipe with Axial Through-Wall Flaws

c .

8/9/2019 506062

21/43

_ _

GEAP-10236

2 5 0 0 0 1

Modified EquationI 0 1I I I I I 10 1. 0 2.0 3.0 4 . 0 5.0

Crack Half-Length (inches)Diameter Schedule__.-V 4 40

A 6 400 6 80 Fatigued0 8 40X 10 400 12 80

0 6 80

Figure 7. Effective Critical Stress Intensity Facto r for ASTM A 106B PipesContaining Axial Through-Wall Flaws

-13-

8/9/2019 506062

22/43

GEAP-10236

* I I 1 I I I 1 1 1 I

1.2 - -A A 0 -

X o o -00 e

1.0 - x ob"0 O e 0b 0 6 . 3 . 0 0

's 0.9 -0.8 -

0.7 ---

I I I I I I I I 1 I i

DIAMETER SCHEDULE R t R/t Rtv 4 40 2.10 0.23 9.15 0.48A 6 40 3.20 0.28 11.4 0.90

80 3.12 0.43 7.26 1.340 60 8 40 4.17 0.32 12.8 1.35

40 5.22 0.36 14.5 1.8812 80 6.04 0.71 8.51 4.28

x 10

Figure 8. a f h a Versus Crack Half-Length, ASTM A 106B Pipe

-14-

8/9/2019 506062

23/43

.. .-

GEAP-10236

g 1.0En

a+ 0.2VIQ:3m

UNFLAWED, x / T = l

NUMBERS INDICATE VALUES0 0.94

0.0 v I I I 1 I0 8 0.2 0.4 0.6 0.8 1 o 1.2

I/L* (in.-1)

Figure 9. Failure Diagram for AS TM A 106B Pipe, Part-Through Flaws

-1 5-

8/9/2019 506062

24/43

GEAP-10236

kJ . O NUMBERS INDICATE VALUESbCIWs4 0.83

W3-I2 0.6

-I I0 PIPE NO. 17

L08t;s 0.4-L A PlFE NO. 18

0 IPE NO. 19SOLID POINTS OUTSIDE FLAWSOPEN POINTS INSIDE FLAW

I J0.0 I I I II/L* (in.-')

0.8 1 o 1.20.2 0.4 0.60.0

Figure 10. Comparison of Inside with Outside Flaws

-16-

8/9/2019 506062

25/43

G AP-10236

consistency compared to the data illustrated in Figure9 i s apparent. Also, it is apparent that the reduction in strengthresulting from an inside flaw i s equivalent to that from an external flaw of the same size and depth.4.3 PJPESWITH CIRCUMFERENTIAL FLAWS

Because the axial stress in a pressurized, thin-walled, cylindrical vessel i s but one-half th e circumferential stress, it maybe expected that the reduction in pressure load limi t due to a circumferential flaw will be less than that due to an axial flawof comparable size. This i s indeed the case, as seen in Table 2, which l i s t s the pressure load limits for a number of specimensof ASTM A1066 pipes containing circumferential flaws of various configurations. In each case, the specimen was tested usingwater, or a mixture of water and ethylene glycol, as the pressurizing medium. With the exception of the tests involvingbending moments, al l tes ts were conducted a t approximately 6OoF. In all specimens, except those containing the most severeflaws, the hoop stress reached levels sufficient to cause general yielding before the stress intensity in the direction normal tothe flaw reached th e instability level.

To study the failure of circumferentially flawed pipes under bending stresses, four pipes were tested under bendingloads. The experimental configuration used for testing specimens under combined bending and pressure loading i s shownschematically in Figure 11. In each case, the bending load was applied a t a point approximately 33 inches from the flawlocation. The hydraulic cylinder, of a 3.25-inch piston diameter, was actuated by an electrical pump in conjunction with aspring-loaded pressure relief valve to maintain the bending load a t any desired constant value.

Test data for pipes tested under bending loads with or without simultaneous pressure loads are given in Table 3.The condition of the specimens (6-6388 and 6-64-88) after tes t i s shown in Figures 12 and 13. Both specimens failed

by sudden propagating fracture once the load limit was reached. After a bending moment of approximately 6.6 X 10pound-inches was applied to specimen 6-63-86, the load was removed to inspect the ends of the 4-inch through-wall flaw. Aplastic displacement (COD) of approximately 0.10 inch had occurred a t the ends of the flaw. Upon reloading, failure occurredsuddenly around the ful l circumferencea t an applied moment of 6.50 X 10 pound-inches. The fracture surface was found t obe perpendicular to the maximum principal stress, wi th minimal shear lips. In each case, the applied load a t the moment ofbrittle fracture was grossly in excess of what would be permitted by design codes.

For comparison with data from axially flawed pipes, approximate elastic crit ical stress intensity factors were calculatedfor the pipes tested under bending loads, using either the Harris or the Gilmang relation, depending on flaw type. I f thetoughness anistropy of the pipe shown by the Charpy tests i s considered, the Kc values of 133 and 147 ksifln. for thecircumferential through-wall flaws i s not inconsistent with the value of 96 ksi &n obtained for axial through-wall flaws. Thebehavior of the pipes with complete part-through flaws is clearly not consistent with the Harris formula, as would beexpected, considering that necking preceded failure on the side of the pipe wall opposite the flaw,4.4 FAILURE BEHAVIOR OF FLAWED TEES A N D ELBOWS

Test procedures for flawed tees and elbows were identical to those used for straight segments of pipe. Test data andresults for the specimens tested are given in Table 4. Pressure load limits ape expressed as a fraction of the burst pressure of anunflawed fi tting in conjunction with the short segments of pipe welded to it. Wall thicknesses of tees and elbows are greaterthan in matching pipe of the same schedule in order to compensate for geometrical stress concentration. In hydrostatic testsof an unflawed te e and an unflawed elbow a t 6OoF, failure in each case initiated as a shear fracture in a crotch of the fittingand propagated into the attached pipe as an axial bri ttle fracture, as shown in Figures 14, 15, and 16.

Figures 17 and 18 show resultant failures from testing flawed elbows and tees, respectively. Under each photograph i sgiven the failure pressure, P/Po as a fract ion of that fo r an unflawed fitting, the flaw length, L,measured a t the pipe surface,the average wall thickness, T, a t the flaw location, and the fraction, x/T, of wall thickness remaining a t the base of the flaw.I f failure pressure data for flaws of types A and C in elbows (see Figure 2) are compared with pipe data by interpolationin Figure 9, it i s seen that somewhat deeper flaws are required to produce a given reduction in failure pressure in elbows thanare required in straight pipe segments. There are two reasons for this: the slightly greater wall thickness in the elbow and thedifference in principal stress ratios. By use of strain gages on unflawed specimens it was found that the circumferential toaxial elastic stress ratio was approximately 1.5 in the location of a type A flaw and 1.7 in the location of a type C law,compared to 2.0 in the wall of a straight pipe.

Analysis of the flawed tees on a basis of either linear elastic fracture mechanics or th e unstable plastic ligament conceptwas not attempted because of diff iculties in defining the stress fields about the flaws. It was found that tees were quitesensitive to flaws of types J and K and (in the absence of bending stresses) almost completely insensitive to flaws of type L.

-17-

8/9/2019 506062

26/43

SpecimenNumber

6-20-8

6-21 8

6-22-8

6-23-8--.90

6-24-8

6-25-8

6-26-8

8-204

6-29-1 8

Size andSchedule

6 in. - 80

6 in. - 80

6 in. - 80

6 in. - 80

6 in. - 80

6 in. - 80

6 in. - 80

8 in. -

6 in.

40

80

OutsideDiameter

6.66

6.66

6.67

6.66

6.66

6.66

6.66

8.66

6.68

WallThickness, T

(in.)0.434

0.429

0.429

0.435

0.432

0.436

0.428

0.321

0.449

Table 2PRESSURE LOAD LIMITS FOR

CIRCUMFERENTIALLY FLAWED PIPES

"T = Wall thicknessx = Ligament thickness

**Expressed as frac tion of value for an unflawed pipe.

BurstPressure

(ksi9.79

8.60

4.81

9.20

7.74

8.20

9.42

4.275

8.58

PressureLoad

Limit**1 oo

0.98

0.55

1 oo

0.88

0.93

-

0.76

-

FlawConfiguration

0.020-in. Milledslot, outside

0.020-in. Milledslot, outside0.020-in. Milledslot, outsideThrough-wall,0.050 in. wide

0.020-in. Milledslot , outsideThrough-wall,0.050 in. wide0.020-in.Milledslot, outside

0.020-in. Milledslot, outside45" Vee, inside

FlawLength

(degrees)360

360

360

34

360

68

360

360

180

x/T*0.72

0.55

0.18

0.00

0.38

0.00

0.52

0.39

0.33

RemarksGeneral yielding priorto axial failure

General yielding prior t ocircumferential failureCircumferential failure

Split ax ially aftergeneral yieldingGeneral yielding prior tocircumferential failureSplit axially aftergeneral yielding

Type-304 stainless steel,22% circumferentialstrain a t failureGeneral yielding,circumferential failureSpl i t axially andcircumferentially

8/9/2019 506062

27/43

G AP- 10236

\

/

FLAWI YOKE 4SSEMBLY~ P E C I M E N

/

\

Specimen Number

Test Temperature, 'FFlaw Lengthx/T(a)Burst Pressure, ksiBurst Moment, I O 5 Ib-in.Kp, ks if i. (b)KB, ksi&n. (b)Kc, ksi6.b)Flaw Type

Figure 11 . Test Assembly, Schematic

Table 3BEHAVI O R O F C I RC UM F EREN T I AL L Y F L AW ED P IPES

UND ER BEND ING A ND PRESSURE STRESSES6-22-8

60360'0.184.810

37.900.020-in. s l i t

37.9(4

6-27-887442.7'0.06.005.49

40.692.2

1330.045-in. s l i t

6-288889360'0.706.526.60

17.740.758.40.020-in. s l i t

- RAME8 x 8 x 3/8 I -BEAM

6-63-887568.6"0.006.500

147147

0.045-in. s l i t

6-64-870

360'0.47506.620

64.664.6(C45' ve

(a) x = Ligament thicknessT = Wall thickness

(b) Kp + KB = Kc.These subscripts refer to pressure, bending, and critical, respectively.

(c) Low values because of inappropriate formulation for K (see text) for360degree flaws.

-19-

8/9/2019 506062

28/43

GEAP-10236 I

Figure 12. Pipe with Through-Wall Flaw, Bending Only

-20-

8/9/2019 506062

29/43

GEAP-10236

Figure 13. Pipe with C ircumferential Part-Through Flaw, Bending Only

-21

8/9/2019 506062

30/43

SpecimenNo.

EL-1-8EL-1A-8EL-2A-8EL-3A-8EL-18-8EL-2B-8EL-1C-8EL-2C-8EL-3C-8T- 1 8T-15-8T-2J-8T-3J-8T-1 K-8T-2K-8T-3K-8T-I L-8T-2L-8T-3L-8(a)See Figure 2

TypeEllEl lEllEllEllEllEl lEllEllTeeTeeTeeTeeTeeTeeTeeTeeTeeTee

FlawType(a)

noneAAABBCCC

noneJJJKKKLLL

GEAP- 10236

MaximumFlaw Length

(inches)

03.256.4 16.3 18.25

11.598.667.945.690

5.694.256.251.951.752.455.56.04.0

Table 4PIPE FITTING TEST DATA

BurstPressure

(ksi1

8.686.564.706.475.866.095.217.054.708.886.336.985.846.306.704.888.859.009.18

AverageFlaw Length

(inches)03.16.36.27.5

10.77.97.24.905.03 O5.21.71 o1.34.75.O

29

PIPo1 oo0.760.540.740.670.700.600.8 10.541 oo0.7 10.790.660.7 10.750.551 oo-1 oo1-00

XK1oo0.390.300.630.290.5 10.350.630.281 oo0.780.450.450.730.700.430.770.550.43

T*av0.4530.4900.4630.4440.4430.4470.47 10.4450.4230.6550.6450.6220.6380.6120.64 10.5590.6460.6450.685

-22-

http://0.0.0.0/

8/9/2019 506062

31/43

GEAP-10236

Figure 14. Fracture Appearance in Tee Tested to Failure

Figure 15. Gross Failure Appearance, Unflawed Elbow

-23-

8/9/2019 506062

32/43

G AP-1'0236

Figure 16. Gross Failure Appearance, Unflawed Tee

-24-

8/9/2019 506062

33/43

I

GEAP- l

E L-1A-8EL-IC-8E L-2 8-8E L-2A-8

EL-1 8-8SEE TABLE 5-1

IEL-2C-8 i EL-3C

EL-3A-8

Figure 17. Failure Appearances, Flawed Elbow

-

8/9/2019 506062

34/43

I I I

._-I

8/9/2019 506062

35/43

GEAP-10236

5. SUMMARYIt has been found that while the instability behavior of carbon steel pipes containing axial through-wall flaws cannot be

predicted accurately by unmodified linear elastic fracture mechanics relationships, it is not diff icult to arrive a t empiricaexpressions which predict the failure of such pipes with acceptable accuracy, a t least for ASTM A1068 pipe a t 60F in sizesfrom 4-to-12-inch nominal diameter.

Pressure load lim it data for pipes containing short, part-through axial flaws are presented in the form of a "failurediagram," in which burst pressure relative to the unflawed condition i s plotted against the reciprocal of flaw length. Fromsuch a diagram the approximate burst strength for a pipe with a flaw of any given length and depth can be estimated.

"Axial" Le., parallel to flow lines) part-through laws in elbows were found to behave much as axial part-through flawsin pipes if differences in circumferential to axial stress ratios were accounted for in the comparison. No exact analysis wasattempted in the case of flawed tees, but in-place laws in the crotch of a tee or ax ia l flaws in the side of a te e were found tocause marked reductions in strength.

-27

8/9/2019 506062

36/43

GEAP-10236

REFERENCES1.2.3.4.5.6.7.8.9.

10.

Reactor Primary Coolant System Rupture Study, Quarterly Progress Report No. 70,July-September 7967, DecemberIrwin, G. R., Fracture, Handbuch der Physik, Springer-Verlag, Berlin, 1958, Vol. 6, pp. 551-590.Anderson, R. B., Fracture Mechanics of Through-CrackedCylindrical Pressure Vessels, American Society for Metals,Metals Park, Ohio, 1965 (Technical Report No. W44-65).Irvine, W. H., Quirk, A.. and Bevitt, E., Fast Fracture of Pressure Vessels: An Appraisal o f Theoretical and Ex-perimental Aspects and Application to Operational Safety, J. Brit . Nucl. Energy SOC., January 1964, p. 31.Eiber, R. J., Hein, A. M., Duffy, A. R., and Atterbury, T J., Investigation o f the Initia tion and Extent of Ductile PipeRupture, Battelle Memorial Institute, Columbus, Ohio, January 1967 (BMI-1793).Folias, E. S., An Axial Crack in a Pressurized Cylindrical Shell, Int. J. Fracture Mechanics, 1, 104 (1965).Eiber, R. J., Hein, A. M., Duffy, A. R., and Atterbury. T J , nvestigation of the Init iation and Extent of Ductile PipeRupture, Battelle Memorial Institute, Columbus, Ohio. November 1968 (BMi-1853).Harris, D. 0.. Stress Intensity Factors for Hollow Circumferentially Notched Round Bars, Lawrence RadiationLaboratory, University of California. Livermore, Calif., April 1966, (UCRL-14859).Gilman, J. D., Stress Intensity Factor for a Circumferential Through-Wall Crack in a Straight Pipe, January 1968,Folias, E. S., On the Prediction of Failure in Pressurized Vessels, College of Engineering, University of Utah, Salt LakeCity, Utah, January 1969, (Report UTEC-CE-006).

1967, (GEAP-5554). p. 34.

(GEAP-5557).

ACKNOWLEDGMENTThe writer extends his appreciation t o G. H. Henderson, who conducted the experimental tests.

-28-

8/9/2019 506062

37/43

GEAP-10236

3 DISTRIBUTIONAerojet GeneralSacramento PlantSacramento, California 95801Attn. Dr. F.J. Climent

Engineering Division

Advisory Committee on Reactor SafeguardsDr. Spencer H. BushConsultant to the DirectorBattelle Memorial InstitutePacific Northwest LaboratoryRichland, Washington99352

Advisory Committee on Reactor SafeguardsMr. Harold Etherington84 Lighthouse DriveJupiter, Florida 33458Advisory Committee on Reactor SafeguardsDr. William L. Faith2540 Huntington DriveSan Marino, California 91108

Advisory Committee on Reactor SafeguardsDr. Chester P.SiessDepartmentof Civil Engineering3129 Civ i l Engineering Bldg.University of IllinoisUrbana, Illinois 61801Advisory Committee on Reactor SafeguardsDr. Stephen H. HanauerProfessor of Nuclear Engineering606 Dougherty HallUniversity of TennesseeKnoxville, Tennessee 37916

Advisory Committee on Reactor SafeguardsDr. Joseph M. HendrieNuclear Engineering DepartmentBrookhaven National LaboratoryUpton, New York 11973Advisory Committee on Reactor SafeguardsDr. HerbertS. sbinDepartment of Chemical EngineeringUniversity of MinnesotaMinneapolis, Minnesota 55455

1 Advisory Committee on Reactor SafeguardsDr. Harry 0. Monson, Senior EngineerLaboratory Director's OfficeArgonne National Laboratory9700 South Cass AvenueArgonne, Illinois 60439Advisory Committee on Reactor SafeguardsDr. Arlie A. O'Kelly2421 West Rowland AvenueLittleton, Colorado 80120

1

Advisory Committee on Reactor SafeguardsDr. David Okrent, Senior PhysicistLaboratory Director's OfficeArgonne National Laboratory9700 South Cass AvenueArgonne, Illinois 60439Advisory Committee on Reactor SafeguardsDean Nunzio J. PalladinoCollege of EngineeringThe PennsylvaniaState University101 Hammond BuildingUniversity Park, Pennsylvania 16802Advisory Committee on Reactor SafeguardsDr. William R. StrattonLos Alamos Scientific LaboratoryP. 0. ox 1663Los Alamos, New Mexico 87544Advisory Committee on Reactor SafeguardsDr. CarrollW. ZabelDirector of ResearchUniversity of HoustonCullen BoulevardHouston, Texas 77004

1

'

1 Mr. Raymond F. FraleyExecutive SecretaryAdvisory Committee on Reactor SafeguardsU.S. Atomic Energy CommissionRoom 1034-HWashington, D.C. 20545Argonne National Laboratory9700 South Cass AvenueArgonne, IllinoisAttn: Mr. Paul G. Shewmon

1

-29-

8/9/2019 506062

38/43

GEAP-10236

Argonne National Laboratory9700 South Cass AvenueArgonne, I llinoisAttn: Or. P. LottesArgonne National Laboratory9700 South C a s AvenueArgonne, IllinoisAttn: Dr. C.E. DickermanArgonne National Laboratory9700 South C a n AvenueArgonne, IllinoisAttn: Dr. R.O. lvinsArgonne National Laboratory9700 South C a s AvenueArgonne, IllinoisAttn: Dr. S. FistedisArgonne National Laboratory9700South Cas AvenueArgonne, IllinoisAttn: Dr. R.C. VogelArgonne National Laboratory9700 South Cas AvenueArgonne, IllinoisAttn: LMFBR Program OfficeArgonne National Laboratory9700 South Cass AvenueArgonne, IllinoisAttn: Mr. A. AmorosiArgonne National Laboratory9700 South C a s AvenueArgonne, IllinoisAttn: Dr. L. BakerAtomic Energy CommissionDivision of Reactor Development and TechnologyWashington, D.C. 20545Attn: Col. R.L. EdnieAsst. Director for Army ReactorsAtomic Energy CommissionDivision of Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. M.A. Rosen

Asst. Director for Plant EngineeringAdvisory Committee on Reactor SafeguardsMr. Harold C. Mangelsdorf78 Knollwood RoadShort Hills, New Jersey 07078

1

1

1

1

1

1

2

1

1

1

1

Atomic Energy Commission 1Division of Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. M.J. Whitman

sst. Director for Program Analysis1Atomic Energy Commission

Division of Reactor Development and TechnologyWashington, D.C. 20545Attn: Dr. E.E. Sinclair

Asst. Director for Reactor TechnologyAtomic Energy CommissionDivision of Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. A. Giambusso

Asst. Director for Project Management

1

1Atomic Energy CommissionDivisionof Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. E.E. Kintner

Asst. Director for Reactor EngineeringAtomic Energy Commission 1Division of Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. J. W. Crawford

Asst. Director for Engineering StandardsAtomic Energy CommissionDivision of Compliance, Region I V10395 West Colfax AvenueDenver, Colorado 80215Attn: Mr. John W. FloraAtomic Energy CommissionDivision of ComplianceWashington, D.C. 20545Attn: Mr. L.Kornblith, Jr.Atomic Energy CommissionDivision of Operational SafetyWashington, DE . 20545Attn: Mr. H. GilbertAtomic Energy CommissionDivisionof Reactor StandardsWashington, D.C. 20545Attn: Mr. E.G. CaseAtomic Energy CommissionDivision of Reactor StandardsWashington, D.C. 20545Attn: Mr. M.Bolotsky

1

10

-30-

8/9/2019 506062

39/43

GEAP-10236

Atomic Energy CommissionDivision of Reactor StandardsWashington, D.C. 20545Attd: Mr. A.B. Holt

Atomic Energy CommissionDivision of Reactor StandardsWashington, D.C. 20545Attn: Mr. R. WaterfieldAtomic Energy CommissionDivision of Reactor StandardsWashington, D.C. 20545Attn: Dr. G. Burley

Atomic Energy CommissionDivision of Reactor StandardsWashington, D.C. 20545Attn: Mr. R. lmparaAtomic Energy CommissionWater Projects BranchDivision of Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. W.H. Layman

Atomic Energy CommissionNaval Reactors BranchDivision of Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. R.S. BrodskyAtomic Energy CommissionDivision of Reactor Development and TechnologyWashington, D.C. 20545Attn: A.J. Pressesky

Atomic Energy CommissionDivision of Reactor Developmentan d TechnologyWashington, D.C. 20545Attn: Mr. S.A. SzawlewiczAtomic Energy CommissionDivision of Reactor Development and TechnologyWashington, D.C. 20545Attn: Mr. R.R. Newton

Atomic Energy CommissionDivision of ProductionWashington, D.C. 20545Attn: Mr. George 6. Pleat

1

1

1

1

1

2

7

5

1

1

-31-

Atomic Energy Commissionc/o Gulf General Atomic, Inc.P.O. Box 608San Diego, California 921 12Attn: Mr. Russell H. BallAtomics InternationalP.O. Box 309Canoga Park, CaliforniaAttn: Dr. H. MorewitzBabcock & Wilcox CompanyWashington Operations Office1725 IStreet, N.W.W%shington,D.C. 20006Babcock & Wilcox CompanyP.O. Box 1260Lynchburg, VirginiaAttn: Mr. Robert WascherBattelle Memorial Institute505 King AvenueColumbus, Ohio 43201Attn: Dr. D.N. SundermanBattelle Memorial Institute505 King AvenueColumbus, Ohio 43201Attn: Dr. D.L. MorrisonBattelle Memorial Institute505 King AvenueColumbus, Ohio 43201Attn: Mr. S. PaprockiBattelle Memorial Institute505 King AvenueColumbus, Ohio 43201Attn: Mr. A.R. D u mBrookhaven National LaboratoryUpton, Long Island, New York 11973Attn: A.W. CastlemanUniversity of CaliforniaInstitute of Engineering ResearchBerkeley, California 94704Attn: Prof. V.E. SchrockHans A. MaurerCommissionof the European CommunitiesRue De La LO1 2001040 Brussels, Belgium

1

2

1

1

2

1

2

1

1

1

1

8/9/2019 506062

40/43

GEAP-10236

Canoga Park Area OfficeP.O. Box 591Canoga Park, California 91305Attn: mr. R.L. Morgan RDT Senior Site Rep.Combustion Engineering, Inc.Nuclear DivisionP.O. Box 500Windsor, Connecticut 06095Attn: Mr. M.F. Valerino

Chicago Operations Off ceAtomic Energy Commission9800 South Cass AvenueArgonne, Illinois 60439Attn: Mr. D.M. GardinerDouglas United NuclearRichland, WashingtonAttn: Mr. John RichesHarvard Air Cleaning LaboratoryHarvard University665 Huntington AvenueBoston, Massachusetts 02190

IIT Research Insti tute10W. 35th StreetChicago, Illinois 60616Am: Dr . T.A. ZakerIIT Research Institute10 W. 35th StreetChicago, Illinois 60616Attn: Mr. E.V. GallagherIdaho Operations OfficeAtomic Energy CommissionP.O. Box 2108Idaho Falls, Idaho 83401Attn: Mr. D. Williams

Liquid Metal Engineering Centerc/o Atomics InternationalP.O. ox 309Canoga Park, California 91304Attn: R.W. DickinsonLos Alamos Scientific LaboratoryP.O. Box 1663Los Alamos, New Mexico 87544Attn: Mr. J.H. ,Russel, K Division

1

1

1

1

1

1

1

2

1

1

-32-

MPR Associates, Inc.1140 Connecticut Avenue, N.W.Washington, D.C. 20036Attn: Mr. T. Rockwell Il l

Chairman AIF Safety Task ForceNational Bureau of StandardsWashington, D.C. 20545Attn: Dr. C. MuehlhauseNaval Ordnance LaboratoryWhite OakSilver Spring, MarylandAttn: Mr. James ProctorNorth Carolina State UniversityDepartment of Mechanical EngineeringRaleigh, North Carolina 27607Attn: Prof. M.N. OzisikNuclear Fuels ServicesWest Valley, New York 14171Oak Ridge Operations OfficeAtomic Energy CommissionOak Ridge, Tennessee 37830Attn: Mr. W.L. SmalleyOak Ridge National LaboratoryNuclear Safety Information CenterP.O. Box YOak Ridge, Tennessee 37830Attn: Mr. Joel BuchenanChemical Technology DivisionOak Ridge National LaboratoryP.O. Box YOak Ridge, Tennessee 37830Attn: Mr. D. FergusonChemical Technology DivisionOak Ridge National LaboratoryP.O. Box YOak Ridge, Tennessee 37830Attn: Mr. R. BlancoOak Ridge National LaboratoryP.O. Box YOak Ridge, TennesseeAttn: HTGR Safety Program Off iceOak Ridge National LaboratoryP.O. Box YOak Ridge, TennesseeAttn: Mr. W. B. Cottrell

1

1

1

1

1

1

1

1

1

2

4 c

8/9/2019 506062

41/43

_ _ _ _

GEAP-10236

Oak Ridge National LaboratoryP.O. Box YOak Ridge, TennesseeA th : Mr. P. RittenhousePacific Northwest LaboratoriesP.O. Box 999Richland, Washington 99352Attn: Mr. G. RogersPacific Northwest LaboratoriesP.O. Box 999Richland, Washington 99352Attn: Mr. J.C. SpannerPacific Northwest LaboratoriesP.O. Box 999Richland, Washington 99352Attn: Dr. J. BatchPacific Northwest LaboratoriesP.O. Box 999Richland, Washington 99352Attn: Mr. R. NightingalePacific Northwest LaboratoriesP.O. Box 999Richland, Washington 99352Attn: Mr. L. SchwendimanPacific Northwest LaboratoriesP.O. Box 999Richland, Washington 99352Attn: Mr. E.R. Astley, Mgr. FFTFIdaho Nuclear CorporationP.O. Box 1845Idaho Falls, Idaho83401Attn: Mr. Curt HaireIdaho Nuclear Corporat ionP. 0. Box 1846Idaho Falls, Idaho83401Attn: Mr. S. 0.JohnsonIdaho Nuclear CorporationP.O. Box 1845Idaho Falls, Idaho83401Attn: Mr. H. L. CoplenIdaho Nuclear CorporationP. 0. Box 1845Idaho Falls, Idaho 83401A m : Mr. 0. F. Brockett

1 Idaho Nuclear CorporationP. 0. Box 1845Idaho Falls, Idaho 83401Attn: Water Reactor Safety Program

Office, Mr. G. 0. Bright, Mgr.1Richland Operations OfficeP.O. Box 500Richland, Washington 99352Attn: Mr. C. RobinsonRichland Operations OfficeP.O. Box 500Richland, Washington 99352Attn : Mr. A.S. WaterhouseRichland Operations OfficeP.O. Box 500Richland, Washington99352Attn: Mr. A. BrunstadSan Francisco Operations Off ceAtomic Energy Commission211 1 Bancroft WayBerkeley, California94704Attn: Mr. C.V. BacklundSavannah River LaboratoriesE.I. duPont deNemours and CompanyAiken, South Carolina 29802Attn: Mr. A.H. PetersTRW Inc.TRW Systems GroupOne Space ParkRedondo Beach, California90278Attn: Or. 0. E. Langmuir

1

1

1

1

TRW I nc.TRW Systems GroupOne Space ParkAttn: Mr. S.M.ZiviWestinghouse Electric CorporationAtomic Power DivisionPittsburgh, Pennsylvania 15230Attn : Mr. R.A. Wiesemann

* Redondo Beach, California 902782 P.O. Box355

Westinghouse Electric CorporationP.O. Box 355Pittsburgh, Pennsylvania 15230Attn: Dr. D. Fletcher

1 Atomic Power Division

8/9/2019 506062

42/43

GEAP-10236

Westinghouse Electric CorporationAtomic Power DivisionP.O. Box 355Pittsburgh, Pennsylvania 15230Attn: Dr. E.'Beckjord

Roger W. StaehleMetallurgy DepartmentOhio State UniversityColumbus, Ohio

Westinghouse Electric CorporationP.O. Box 19218Tampa, Florida 33616A m : Mr. A. LohmeierSouthern Nuclear Engineering, Inc.P.O. Box 10Donedin, Florida 33528Attn: Mr. Gilbert BrownHolmes & Narver, Inc.828 South Figueroa St .Los Angeles, California 90017Attn: B. ShimizuMr. Edward T. WesselResearch and Development CenterWestinghouse Electric CorporationBeulah Road, Churchill BoroPittsburgh, Pennsylvania 15235Dr. William E. CooperTeledyne Materials Research303 Bear Hill RoadWaltham, Massachusetts

Mi. Ralph JonesDivision of Reactor DevelopmentU.S. Atomic Energy CommissionWashington 25, D.C.Mr. H.K. MarksRoom 2N83Department of the NavyWashington, D.C.H. Thielsch140 Shaw AvenueCranston 5, Rhode IslandKnolls Atomic Power LaboratoryP. 0. Box 1072Schenectady, New York 12301Attn: Dr. Robert A. Barnes

1

1

1

1

1

1

1

1

1

1

1

-34-

Mr. F.M. MoschineWestinghouse Electric CompanyAtomic Power DepartmentP.O. Box 355Pittsburgh, Pennsylvania 15230Dr. P. L. PfenningwerthBettis Atomic Power LaboratoryP.O. Box 79West Mifflin, Pennsylvania 15122

1

1

1Professor C.E. TaylorDepartment of Theoreticaland Applied MechanicsUniversity of IllinoisUrbana, IllinoisMr. E. Beauchamp-NobbsU.S. Marine Engineering LaboratoryAnnapolis, MarylandU.S. Atomic Energy CommissionDivision of Technical Information ExtensionP.O. Box 62Oak Ridge, TennesseeU.S. Atomic Energy CommissionDivision of Reactor LicensingWashington, D.C. 20545Attn: S.S. PawlickiUSAEC Site RepresentativeGeneral Electric CompanySunnyvale, California 94086Attn: Joel Levy, Senior Site Rep.Mr. B.L. GreenstreetP.O. Box YOak Ridge National LaboratoryOak Ridge, TennesseeMr. F.J. W i t tP.O. Box YORNL - Oak Ridge, TennesseeCommonwealth Edison CompanyDresden Nuclear Power StationRural Route 1Morris, Illinois 60450Attn: H.K. HoytCommonwealth Edison CompanySystem Mechanical and Structural Engineer72 West Adams StreetChicago, Ill inois 60690Attn: N.A. Kershaw

1

1

1

1

2

3

8/9/2019 506062

43/43

United Kingdom Atomic Energy AuthorityReactor Materials LaboratoryWigshaw Lane, CulchethWarrington, Lancs.EnglandAttn: R.W. NicholsAtomic Energy of Canada LimitedChalk River Nuclear LaboratoriesChalk River, Ontario, CanadaDr. George PonMr. Robert D. WylieDepartment of Materials EngineeringSouthwest Research Institute8599 Culebra RoadSan Antonio, Texas 78228

O.A. Keilermanlnstitut Fur ReaktorsicherheitDer Technischen UberwachungsVereine, e.V.5 KOLN 1 Glockengasse 2West GermanyC. A. G. PhillipsU.K.A.E.A. Safeguards DivisionAuthority Health & Safety BranchRisley, Warrington, LancashireEnglandDivision of Reactor Development & TechnologyU.S. Atomic Energy CommissionWashington, D.C. 20545Attn: J. R. HunterMr. T. R.MageeWestinghouse Electric Co.PW R Systems DivisionP.O. Box 355Pittsburgh, Pen nsy Ivan a 15230

GEAP-10236

1 Reactor Materials BranchMetallurgy DivisionNaval Research LaboratoryWashington, D.C. 20545Attn: Mr. L. E. SteeleDepartment of Material Science & EngineeringHearst Mining BuildingUniversity of CaliforniaBerkeley, CaliforniaAttn: Mr. William W. Gerberich

1United Engineers and Constructon. Inc.1401Arch StreetPhiladelphia, Pennsylvania 19105Ann: Mr. John Crowley

Argonne National Laboratory' 9700 South Cass AvenueArgonne, IllinoisAttn: Mr. Craig Cheng

AEG TelefunkenAEC Hochhaus Sued

1 6 FrankfudMain 70West GermanyAttn: Mr. Dieter Ewers

Asst. Director, Instruction DivisionU.S. Atomic Energy Commission9800 South Cass AvenueArgonne, Illinois 60439A m : R. M. Moser

1 Chicago Operations Off ce

1