Wednesday, 24 May 2017

Pressure Vessels Design and Practice by Somnath Chattopadhyay Free Download PDF





Overview of Pressure Vessels


Vessels, tanks, and pipelines that carry, store, or receive fluids are called pressure vessels. A pressure vessel is defined as a container with a pressure differential between inside and outside. The inside pressure is usually higher than the outside, except for some isolated situations. The fluid inside the vessel may undergo a change in state as in the case of steam boilers, or may combine with other reagents as in the case of a chemical reactor. Pressure vessels often have a combination of high pressures together with high temperatures, and in some cases flammable fluids or highly radioactive materials. Because of such hazards it is imperative that the design be such that no leakage can occur. In addition these vessels have to be designed carefully to cope with the operating temperature and pressure. It should be borne in mind that the rupture of a pressure vessel has a potential to cause extensive physical injury and property damage. Plant safety and integrity are of fundamental concern in pressure vessel design and these of course depend on the adequacy of design codes. When discussing pressure vessels we must also consider tanks. Pressure vessels and tanks are significantly different in both design and construction: tanks, unlike pressure vessels, are limited to atmospheric pressure; and pressure vessels often have internals while most tanks do not (and those that do are limited to heating coils or mixers). 

Pressure vessels are used in a number of industries; for example, the power generation industry for fossil and nuclear power, the petrochemical industry for storing and processing crude petroleum oil in tank farms as well as storing gasoline in service stations, and the chemical industry (in chemical reactors) to name but a few. Their use has expanded throughout the world. Pressure vessels and tanks are, in fact, essential to the chemical, petroleum, petrochemical and nuclear industries. It is in this class of equipment that the reactions, separations, and storage of raw materials occur. Generally speaking, pressurized equipment is required for a wide  range of industrial plant for storage and manufacturing purposes. The size and geometric form of pressure vessels vary greatly from the large cylindrical vessels used for high-pressure gas storage to the small size used as hydraulic units for aircraft. Some are buried in the ground or deep in the ocean, but most are positioned on ground or supported in platforms. Pressure vessels are usually spherical or cylindrical, with domed ends. The cylindrical vessels are generally preferred, since they present simpler manufacturing problems and make better use of the available space. Boiler drums, heat exchangers, chemical reactors, and so on, are generally cylindrical. Spherical vessels have the advantage of requiring thinner walls for a given pressure and diameter than the equivalent cylinder. Therefore they are used for large gas or liquid containers, gas-cooled nuclear reactors, containment buildings for nuclear plant, and so on. Containment vessels for liquids at very low pressures are sometimes in the form of lobed spheroids or in the shape of a drop. This has the advantage of providing the best possible stress distribution when the tank is full. 

Pressure Vessels Design Philosophy

General overview

Engineering design is an activity to ensure fitness for service. Within the context of pressure vessel design, this primarily involves strength considerations. The ‘‘total design’’ is a topic with far-reaching ramifications. It might include aspects of fuel system design, reactor design, or thermal hydraulic design. In our subsequent discussions, the underlying philosophy, decisions and calculations related solely to the strength design are referred to the ‘‘pressure vessel design.’’ For certain pressure vessels and related equipment, preliminary design may still be governed by heat transfer and fluid flow requirements. Although the aspect of thermal hydraulic design is intricately related to the structural design, especially for thermal transient loadings, we will not be discussing them in any detail. It will be assumed that the temperature distribution associated with a particular thermal transient has already been evaluated in a typical design application. However, in these cases the designer still has to consider how the desired configurations of the vessel are to be designed from a structural standpoint and how these designs will perform their intended service. The role of engineering mechanics in the pressure vessel design process is to provide descriptions of the pressure vessel parts and materials in terms of mathematical models, which can be analyzed in closed form in a limited number of situations and mostly have to be solved numerically. Even the so-called simple models that can be solved in closed form might involve fairly complex mathematics. In a few isolated instances, intelligent applications of well-known principles have led to simplifying concepts. These concepts have generally eased the designer’s task. However, in a majority of cases, especially when advanced materials and alloys are at a premium, there is a need to make the optimum use of the materials necessitating application of advanced structural analysis. As the complexity of the analysis increases, the aspect of interpretation of the results of the analysis becomes increasingly extensive. Furthermore, a large number of these models approximate the material behavior along with the extent of yielding. As we understand material behavior more and more, the uncertainties and omitted factors in design become more apparent. The improvement will continue as knowledge and cognizance of influencing design and material parameters increase and are put to engineering and economic use.

The safety demands within the nuclear industry have accelerated studies on pressure vessel material behavior and advanced the state of the art of stress analysis. For instance, the nuclear reactor, with its extremely large heavy section cover flanges and nozzle reinforcement operating under severe thermal transients in a neutron irradiation environment, has focused considerable attention on research in this area which has been directly responsible for improved materials, knowledge of their behavior in specific environments, and new stress analysis methods. High-strength materials created by alloying elements, manufacturing processes, or heat treatments, are developed to satisfy economic or engineering demands such as reduced vessel thickness. They are continually being tested to establish design limits consistent with their higher strength and adapted to vessel design as experimental and fabrication knowledge justifies their use. There is no one perfect material for pressure
vessels suitable for all environments, but material selection must match application and environment. This has become especially important in chemical reactors because of the embrittlement effects of gaseous absorption, and in nuclear reactors because of the irradiation damage from neutron bombardment. Major improvements, extensions and developments in analytical and experimental stress analysis are permitting fuller utilization of material properties with confidence and justification. Many previously insoluble equations of elasticity are now being solved numerically. These together with experimental techniques are being used to study the structural discontinuities at nozzle openings, attachments, and so on. This is significant because 80 percent of all pressure vessel failures are caused by highly localized stresses associated with these ‘weak link’ construction details. It is therefore apparent that the stress concentrations at vessel nozzle openings, attachments, and weldments are of prime importance, and methods for minimizing them through better designs and analyses are the keys to long pressure vessel life. Control of proper construction details results in a vessel of balanced design and maximum integrity. In the area of pressure vessel design there are important roles played by the disciplines of structural mechanics as well as material science. As  mentioned earlier, we try to provide a description of pressure vessel components in terms of mathematical models that are amenable to closedform solutions, as well as numerical solutions. The development of computer methods (sometimes referred to as computer-aided design, or CAD) has had a profound impact on the stress and deflection analysis of pressure vessel components. Their use has been extended to include the evaluation criteria as well, by a suitable combination of post processing of the solutions and visual representation of numerical results. In a number of cases advanced software systems are dedicated to present animation that aids the visualization and subsequent appreciation of the analysis. A number of design and analysis codes have been developed that proceed from the conceptual design through the analysis, sometimes modeling the nonlinear geometric and material behavior. Results such as  temperatures, deflections and stresses are routinely obtained, but the analysis often extends to further evaluations covering creep, fatigue, and fracture mechanics. With the advent of three-dimensional CAD software and their parametric, feature-driven automated design technology, it is now possible to ensure the integrity of designs by capturing changes anywhere in the product development process, and updating the model and all engineering deliverables automatically. Pressure vessel designs that once averaged 24 hours to finish are completed in about 2 hours. Such productivity gains translate into substantial savings in engineering labor associated with each new pressure vessel design. The typical design of a pressure vessel component would entail looking at the geometry and manufacturing construction details, and subsequently at the loads experienced by the component. The load experienced by the vessel is related to factors such as design pressure, design temperature, and mechanical loads (due to dead weight and piping thermal expansion) along with the postulated transients (typically those due to temperature and pressure) that are anticipated during the life of the plant. These transients generally reflect the fluid temperature and pressure excursions of the mode of operation of the equipment. The type of fluid that will be contained in the pressure vessel of course is an important design parameter, especially if it is radioactive or toxic. Also included is the information on site location that would provide loads due to earthquake (seismic), and other postulated accident loads. 

In assessing the structural integrity of the pressure vessel and associated equipment, an elastic analysis, an inelastic analysis (elastic–plastic or plastic) or a limit analysis may be invoked. The design philosophy then is to determine the stresses for the purpose of identifying the stress concentration, the proximity to the yield strength, or to determine the shakedown limit load. The stress concentration effects are then employed for detailed fatigue evaluation to assess structural integrity under cyclic loading. In some situations a crack growth analysis may be warranted, while in other situations, stability or buckling issues may be critical. For demonstrating adequacy for cyclic operation, the specific cycles and the associated loadings must be known a priority. In this context, it is important for a pressure vessel designer to understand the nature of loading and the structural response to the loading. This generally decides what type of analysis needs to be
performed, as well as what would be the magnitude of the allowable stresses or strains. Generally the loads acting on a structure can be classified as sustained, deformation controlled, or thermal. These three load types may be applied in a steady or a cyclic manner. The structure under the action of these loads may respond in a number of ways 

  • When the response is elastic, the structure is safe from collapse when the applied loading is steady. When the load is applied cyclically a failure due to fatigue is likely; this is termed failure due to high cycle fatigue.
  • When the response is elastic in some regions of the structure and plastic in others, there is the potential to have an unacceptably large deformation produced by both sustained and deformation-controlled loads. Cyclic loads or cyclic temperature distributions can produce plastic deformations that alternate in tension and compression
  • and cause fatigue failure, termed low cycle fatigue. Such distribution of loads could be of such a magnitude that it produces plastic deformations in some regions when initially applied, but upon removal these deformations become elastic, and subsequent loading results in predominantly elastic action. This is termed shakedown. Under cyclic loading fatigue failure is likely and because of elastic action, this would be termed as low cycle fatigue.
  • When the sustained loading (due to bending or tension) is such that the entire cross-section becomes plastic, gross collapse of the structure takes place.
  • Ratcheting is produced by a combination of a sustained extensional load and either a strain-controlled cyclic load or a cyclic temperature distribution that is alternately applied and removed. This produces cycling straining of the material which in turn produces incremental growth (cyclic) leading to what is called an incremental collapse. This can also lead to low cycle fatigue. 
  • Sustained loads in brittle materials or in ductile materials at low temperatures could result in brittle fracture, which is a form of structural collapse. 

Structural Design Criteria

Modes of failure

Two basic modes of failure are assumed for the design of pressure vessels. These are: (a) elastic failure, governed by the theory of elasticity; and (b) plastic failure, governed by the theory of plasticity. Except for thick-walled pressure vessels, elastic failure is assumed. When the material is stretched beyond the elastic limit, excessive plastic deformation or rupture is expected. The relevant material properties are the yield strength and ultimate strength. In real vessels we have a multiaxial stress situation, where the failure is not governed by the individual components of stress but by some combination of all stress components. 

Theories of failure

The most commonly used theories of failure are:

Maximum principal stress theory
Maximum shear stress theory

Fatigue assessment of pressure vessels

Fatigue has been recognized as a major failure mode in pressure vessels, and specific rules for its prevention appear in design codes. Stated simply, fatigue failure is caused by the cyclic action of loads and thermal conditions. In many design situations, the expected number of cycles is in millions and for all practical purposes can be considered as infinite. Accordingly, the concept of endurance limit has been employed in a number of design rules. Endurance limit is the stress that can be applied for an infinite number of cycles without producing failure. However, the typical number of stress cycles rarely exceeds 100,000 and frequently only a few thousand. Therefore, fatigue analysis requires somewhat more involved concepts than just the endurance limit. Fatigue refers to the behavior of material under repeated loads, which is distinct from the behavior under monotonically applied loading. There is a progressive localized permanent deformation under fluctuating loads that culminates in cracks and complete fracture after a sufficient number of cycles. The fatigue process itself occurs over a period of time. However, failure may occur suddenly and without prior warning, in which case the damage mechanisms may have been operating since loading was first introduced. This period of time is often  eferred to as the usage period. The fatigue process appears to initiate from local areas that have high stresses. These highly stressed regions are due to abrupt changes in geometry leading to high stress concentrations, due to temperature differentials, imperfections, or the presence of residual stress. The failure takes place when the crack after repeated cycling grows to a point at which the material can no longer withstand the loads and a complete separation occurs. Metallurgical defects such as a void or an inclusion often act as sites for fatigue crack initiation. The fatigue process consists of crack initiation, crack propagation, and eventual fracture. Another way to look at the process is to postulate it in terms of initiation of microcracks, coalescence of these microcracks into macrocracks, followed by growth to unstable fracture. Fatigue has been classified as one of high cycle and low cycle. High cycle fatigue involves very little plastic action. The low cycle fatigue failure involves a few thousand cycles and involves strains in excess of yield strain. Fatigue damage in the low cycle has been found to be related to plastic strain and fatigue curves for use in this region should be based on strain ranges. For the high cycle fatigue cases, the stress ranges can be used. The procedure of using strain amplitude as a function of number of cycles forms the cornerstone of fatigue analysis. The design curve is based on strain-controlled data. The best-fit curves were reduced by a factor of 2 on stress and 20 on cycles to account for environment, size effect, and scatter of data. The basic elements of the fatigue evaluation in pressure vessels rests on the use of maximum shear theory of failure (Tresca criterion), with the assumptions of linear elastic behavior along with the use of Miner’s rule for estimating the cumulative effect of stress cycles of varying amplitude. Fatigue failure typically occurs at structural discontinuities which give rise to stress concentration. The stress concentration factors are generally based on theoretical analysis involving statically applied loads. These are directly applicable to fatigue analysis only when the nominal stress multiplied by the stress concentration factor is below the material yield strength. When it exceeds the yield strength, there is a redistribution of stress and strain. For sharp geometries, using values of stress concentrations obtained elastically leads to underprediction of fatigue lives when compared with the actual test data. In fact under no circumstances does a value of more than 5 need be applied, and it is observed that the value of these factors do not vary with the magnitude of cyclic strain and associated fatigue life. In the design of pressure vessels, the stress concentration which is really the strain concentration – is limited to 5 and for most discontinuities such as grooves and fillets no more than a value of 4 is used