An Evaluation of ASME’s “Design by Analysis” Guidelines

Abstract

ASME’s Boiler and Pressure Vessel Codes have a history of over one hundred years. The codes have been evolving over time with continuous revisions, improvements and refinements. A major milestone has been the incorporation of “Design by Analysis (DBA)” guidelines about fifty years back (for instance Sec. VIII, Division 2). These were introduced as it was recognized that the prevailing Design by Rules (Section VIII, Division 1) tended to be somewhat over-conservative. The essence of DBA guidelines consists of evaluating the elastic stresses at critical locations and checking the same against the allowable. The allowable happen to functions of the nature of stress distribution and the nature of load. A given stress could be of membrane, bending or peak category and also be either primary or secondary. At the time of appearance of the DBA guidelines, the state of the art of stress analysis was not well advanced and the finite element method was just getting developed. As of today, however, the finite element method has reached a high level of maturity and is very widely used. The latest edition (2010) has recognized this and it contains modeling and post-processing guidelines applicable to FE analysis. This edition also recommends the use of one of three possible approaches. The first is the elastic analysis and classification and categorization of stresses with guidelines regarding how to deal with two- and three-dimensional situations. The other two options are provided to take care of situations wherein the categorization process may lead to either uncertainty or ambiguity. These involve nonlinear analysis either by way of Limit-Load method or Elastic-Plastic Stress Analysis. In either approach the analyst will look for the loads at which there is an onset of gross plastic flow. In the present paper an attempt is made to evaluate the latest DBA guidelines from design application point of view. The purpose is to assess the limitations of the elastic analysis approach. Studies are undertaken to focus typically on the following aspects: 1. Two dimensional problems involving symmetry or axisymmetry. There are situations in which the “bending” stresses are liable to be misinterpreted. 2. Three dimensional problems with emphasis on the assessment of bending stress as categorization in 3D situations is a real challenge 3. General situations involving the secondary stresses. The allowable stress limit for secondary stress is somewhat arbitrary and perhaps very conservative. The studies tend to suggest that the nonlinear route is to be adopted as it is reliable and accounts for many uncertainties associated with the elastic approach.