0


EDITORIAL

J. Pressure Vessel Technol. 103, 1 (1981) (1 page);   doi:10.1115/1.3263364
FREE TO VIEW
Abstract

RESEARCH PAPERS

J. Pressure Vessel Technol. 103, 2-9 (1981) (8 pages);   doi:10.1115/1.3263367

The use and operational experience of forged high-pressure monobloc vessels is reviewed. It is noted that such vessels fall mainly into two categories: firstly those like stirred autoclaves for use in the continuous polyethylene process which are subjected to at the most two or three thousand cycles of pressure in a lifetime; and secondly, those such as isostatic compaction vessels which may be subjected to more than a hundred thousand cycles of pressure. It is concluded that the design procedures for one category are not sufficient for the other. Review of the performance of polyethylene autoclaves suggest that the design procedures which are based on well-established fundamental principles are more than adequate to ensure a safe design. With isostatic compaction vessels there is the added requirement to design against fatigue, and in particular to ensure adequate strength in the transition region between the threaded portion of the bore required for the screwed plug and the main bore, as it is mainly in this region that calamitous failures have been initiated. Some thoughts on some of the requirements for a code of practice for forged high-pressure vessels are put forward.

J. Pressure Vessel Technol. 103, 10-15 (1981) (6 pages);   doi:10.1115/1.3263358

Liquid sloshing in pool due to three-directional earthquake ground motion is analyzed. The liquid pool is represented by a rigid annular circular cylindrical tank. Analytical and numerical solutions are presented and their limitations are discussed. For a given seismic ground excitation time-history, the free surface and the pressure and velocity fields in the pool are calculated by superposition of modal responses. The results show that container vertical acceleration is of secondary importance in determining the free surface displacement, but has a major effect on the pressure load on the container boundary.

J. Pressure Vessel Technol. 103, 16-19 (1981) (4 pages);   doi:10.1115/1.3263363

A general procedure is presented for a more accurate, rapid and economical computation of response from higher frequency modes, when the modal superposition time history or response spectra method is used for dynamic analysis of structures subject to uniform, translational seismic excitation. The procedure utilizes special amplification characteristics of earthquake loading for frequencies not less than a certain assigned value and certain properties of structural modal characteristics. The upper bound for the assigned value of frequency is usually 33 cycles/s. This procedure requires computation of only those frequencies that are lower than the assigned value, and other modal properties associated with these lower frequencies. For rigid structures (i.e., structures with frequencies not less than the assigned value), this procedure reduces itself down to the familiar static analysis.

J. Pressure Vessel Technol. 103, 20-26 (1981) (7 pages);   doi:10.1115/1.3263365

This paper discusses a hybrid numerical method to calculate homogeneous equilibrium two-phase flows in one space dimension. A finite difference method is used for field solution while the method of characteristics is employed for treating boundary conditions. The boundary method is a new method that uses an integral procedure for precise specification of boundary values leading to an accurate overall solution. No extraneous information, or nonphysical approximation, is required as often is the case in most other attempts where the boundary error was the cause of difficulties and was detrimental to the inaccuracy of the overall solution and the eventual numerical instability due to the large gradient in phase distribution commonly existing near the boundaries. Procedures for many types of boundary conditions commonly encountered in reactor system analyses have been formulated and discussed. Sample problems have been calculated for two-phase water/steam mixture and single-phase nitrogen gas. The results are presented to demonstrate the degree of accuracy attainable from the hybrid method. The basic method discussed may be generalized to apply to more general nonequilibrium two-phase flow models with unequal phase velocities provided real characteristics exist.

J. Pressure Vessel Technol. 103, 27-32 (1981) (6 pages);   doi:10.1115/1.3263366

A modal superposition method for the nonlinear dynamic analysis of a structure subjected to multiple support motions is presented. The nonlinearities are due to clearances between the components and their supports. The finite element method is used to derive the equations of motion with the nonlinearities represented by a pseudo force vector. The displacement response may be divided into two parts: elastic deformation and rigid body motion. The presence of rigid body motion necessitates the inclusion of the higher modes in the transient analysis. The modal superposition method is used to analyze the dynamic response of one loop of the nuclear steam supply system. This loop has nonlinear supports and is subjected to nonuniform seismic excitations at the supports. It is shown that the computational cost of the modal superposition method is lower than that of the direct integration.

J. Pressure Vessel Technol. 103, 33-42 (1981) (10 pages);   doi:10.1115/1.3263368

Severe hydraulic transients may be produced in the intermediate heat transport system of a liquid-metal-cooled breeder reactor by a sodium/water reaction in a failed steam generator. The PTA-2 computer code has been developed to analyze these transients, including the effects of plastic deformation of the piping and cavitation on pulse propagation. Comparisons are shown between PTA-2 predictions and the results of two experiments performed at Stanford Research Institute. Piping was plastically deformed by a pressure pulse traversing the experimental system, and the resulting rarefaction waves caused cavitation to occur. Excellent agreement between computation and experiment was obtained for dynamic strain and pressure distributions and the location and duration of the cavitation. The effect of elbow curvature on pulse propagation was found to be masked by pipe plasticity effects.

J. Pressure Vessel Technol. 103, 43-49 (1981) (7 pages);   doi:10.1115/1.3263369

The propagation of fluid transients through elbows is studied. A set of one-dimensional governing equations for the propagation of pressure pulses in an inviscid compressible fluid contained in a thin-walled naturally curved elastic tube is formulated and solved by two different techniques. For continuous waves, reflection and transmission coefficients for elbows are determined numerically by considering periodic waves in an assemblage of straight and curved tubes. For pulse propagation, the method of characteristics is employed to solve the assemblage problem. An experimental arrangement for pulse studies is described and experimental results are compared with numerical results from the method of characteristics.

J. Pressure Vessel Technol. 103, 50-58 (1981) (9 pages);   doi:10.1115/1.3263370

A technique for numerical integration of the finite difference (matrix) formulation of the unsteady heat transfer equation has been applied to the thermal stress analysis requirements of ASME B&PV Section III, Article NB-3650. This technique, with its properties of unconditional solution stability, has been incorporated into a new computer program, TRANS2A, which has been designed totally around the needs of the stress analyst. To be of maximum aid to the analyst, in addition to the necessary output of thermal gradients (ΔT2 and ΔT2 ) and average temperatures (Ta and Tb ), TRANS2A provides a complete set of thermal stress histories and tables of thermal stress extrema. Values of the thermal stresses are output at maxima of the thermal gradient terms (with or without adjacent sections), in addition to the extrema of the secondary and secondary plus peak stresses and time of occurrence. Each solution is performed for a set of seven general and three optional stress indices. The process allows a strict and simple data interface to the combined stress analaysis computation without excessive approximations. Data may also be stored so that sections need not require repeated analyses. All computational output, from the detailed heat transfer solution to the stress summaries, may be requested or deleted at the option of the analyst. For generality, TRANS2A includes a complete set of temperature-dependent material properties for all current piping materials and a complete set of fluid properties for water, steam, and sodium. Fluid transient data are input using phase and temperature, and a choice of four flow rate specifications. Accepted heat transfer correlations for laminar and turbulent flow in liquids and gases are included, with smoothing at two-phase excursions. Samples of the TRANS2A benchmark problems are included, with discussions on data interface and sensitivity for erratic fluid transients.

J. Pressure Vessel Technol. 103, 59-65 (1981) (7 pages);   doi:10.1115/1.3263371

The theory of tearing-mode crack-growth instability as developed by Paris and his associates is briefly outlined. Then the theory is applied to study the stability of a deep crack in a pressure vessel nozzle. The finite-element method is used to analyze the structure, which is idealized as a two-dimensional elastic-plastic body. Based on the theory and the results of the finite-element calculation, the stability (or instability) of the crack is predicted, as a function of pressure in the vessel. A detailed examination of the solution near the crack tip is made to verify that the necessary conditions for J-controlled crack growth are present.

J. Pressure Vessel Technol. 103, 66-75 (1981) (10 pages);   doi:10.1115/1.3263372

The effect of a girth-weld-induced residual stress field on the linear elastic fracture mechanics of a thin-walled pipe is examined. The procedure for using the residual stress distribution to compute KI and KII for a circumferential crack which is growing radially is described. In addition to the two-pass girth weld, stress intensity factors are computed for a residual stress distribution in a flat plate and for a hypothetical residual stress state in a second thin-walled pipe. The computed stress intensity factor for the flat plate geometry and its residual stress distribution are compared with a solution from the literature as a check on the computational procedure. The through-the-thickness residual stress distribution due to the two-pass girth weld is similar to a half-cosine wave. For purposes of comparison, the hypothetical through-the-thickness distribution selected for the second pipe is similar to a full cosine wave. The stress intensity factor is presented as a function of crack depth for a crack initiating on the inner surface of the pipe. The redistribution of residual stresses due to crack growth is also shown for selected crack lengths. The study shows that residual stress-induced crack growth in pipes can be significantly different from that in flat plates due to the possibility of locked-in residual bending moments in the pipe. These locked-in moments can have effects similar to externally applied loads and can either promote or restrain crack growth. A residual stress distribution is illustrated in which crack growth, if initiated, would continue through the entire wall. Also, a residual stress distribution is illustrated for which the crack could arrest after a certain amount of growth.

J. Pressure Vessel Technol. 103, 76-84 (1981) (9 pages);   doi:10.1115/1.3263373

Finite element stress analysis has been performed to determine the effects of two O.D. notch configurations in a cylinder subjected to internal pressure, or containing autofrettage residual stress. The effects on the residual stresses were determined by simulating these stresses with equivilent temperature loads. The results show that the deeper of the two notch cofigurations is far more severe resulting in a maximum stress concentration factor of 6.6. The shallower notch has a maximum stress concentration factor of 3.7. An additional result is that by introducing notches in autofrettaged cylinders a significant amount of the residual stresses are relieved which indicates that smaller applied pressures can be applied before yielding occurs. The results also show that the possibility of O.D. initiated fatigue failure is greatly increased.

J. Pressure Vessel Technol. 103, 85-93 (1981) (9 pages);   doi:10.1115/1.3263374

A 152-mm-thick cylindrical test vessel fabricated of A533, grade B, class 1 steel, was pressurized to failure at −23°C. The vessel contained a fatigue-sharpened notch adjacent to a half-bead weld repair that had not been stress relieved. Residual stresses and fracture toughnesses were determined before the pressure test by measurements on a prototypic weld, and fracture predictions were made by linear elastic fracture analysis. Predictions agreed well with test results, demonstrating the important influence of high residual stresses on fracture behavior.

J. Pressure Vessel Technol. 103, 94-103 (1981) (10 pages);   doi:10.1115/1.3263375

Analytical and experimental studies of the dynamic response of a system with geometric and material nonlinearity are described. The dynamic excitation consists of sinusoidal and impulsive base acceleraton. The dynamic system, which is representative of many practical cases involving mechanical equipment and piping systems, consists of a cantilever beam with a gapped support at the free end. The material nonlinearity considers both the effect of yielding and the effect of strain rate on the initial yield level. The analytical studies are performed by transforming the continuous system to an equivalent single-degreee-of-freedom system and using numerical techniques to solve the resulting equation of motion. Experimental studies are conducted and responses are measured on the same dynamic systems. Critical comparisons are made between the calculated and measured responses.

TECHNICAL BRIEFS

J. Pressure Vessel Technol. 103, 104-107 (1981) (4 pages);   doi:10.1115/1.3263356

An approximate method for calculating the elastic stresses at the crotch of a tee branch pipe connection under internal pressure is presented. The analaysis is based on shell theory together with stress multipliers for the peak bending stress in the fillet. A table of factors for rapid calculation is given.

J. Pressure Vessel Technol. 103, 107-111 (1981) (5 pages);   doi:10.1115/1.3263357

Elastic stress analysis of a right angle tee branch pipe connection of two pipes of identical diameter and thickness connected through 45-deg chamfer corner sections is developed for internal pressure loading. Stresses in the crotch portion of the vessel are determined. These results are presented in the form of a table of factors useful for rapid calculation of approximate values of the peak stresses. The existence of a structurally optimum size of chamfer is demonstrated.

J. Pressure Vessel Technol. 103, 111-115 (1981) (5 pages);   doi:10.1115/1.3263359

Design of connections of pipes and pressure vessels on the basis of a calculated maximum elastic stress often proves to be too conservative in the case of ductile materials. Elastic-plastic analysis by the finite element method proves to be too costly. This paper presents an alternative method which reduces the calculations to those of a rotationally symmetric shell subjected to axisymmetric loading. Using this approach approximate elastic-plastic deformations on the meridian passing through the crotch of a tee branch connection of cylindrical shells of equal diameter and thickness are determined. The method is limited to cases of the normal intersection of very thin shells of identical diameter, thickness, and material and to internal pressure loading. Numerical results for the intersection of two shells of R/t equal to 100 are given for an elastic-perfectly plastic material satisfying the von Mises yield condition.

PRESSURE VESSEL AND PIPING CODES

J. Pressure Vessel Technol. 103, 116-119 (1981) (4 pages);   doi:10.1115/1.3263360

Paragraph UG-39 (c) of ASME Section VIII, Division 1, 1980 Edition states: “Flat heads that have an opening with a diameter that exceeds one-half of the head diameter or shortest span, as defined in UG-34, shall be designed as a flange in accordance with the Rules for Bolted Flange Connections given in Appendix 2.” The application of Appendix 2 to such configurations is subject to various interpretations, accordingly, rules have been written for the specific case of a single, large central opening (nozzle) in integral flat heads. The new rules will appear in the Winter 1980 Addenda; for configurations not provided for the designer is referred to U-2 (g).

J. Pressure Vessel Technol. 103, 119-124 (1981) (6 pages);   doi:10.1115/1.3263361

Numerous revisions, updates and advances have been incorporated in the ASME Section VIII Code on “Pressure Vessels.” All of them have been published in “Mechanical Engineering” as required by ANSI accredited organization procedures and a few have also been the subject of specific technical papers. The objectives of this paper is not the duplication of these already published works but rather a highlighting of the more significant or, in some cases, subtle revisions that have been incorporated in Section VIII since the 1974 edition. By way of introduction, the basic design philosophies of Division 1 and 2 are also outlined.

DISCUSSIONS

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In