Various methods have been in use for the determination of stresses at the nozzle-shell junction due to external loads and moments. Some methods evaluate stress in the cylindrical or spherical shell (e.g. WRC 107 now WRC 537) while others evaluate stresses in cylindrical shells and nozzles (e.g. WRC 297). ASME Section VIII Division. 2 specifies use of WRC 107/WRC 297 or Finite Element Analysis (FEA) for determination of stresses at shell-nozzle junctions with external loads and moments on the nozzle.
Each method could yield a different result for the same loading condition and geometry and this has been recognized in comparisons made in WRC 297 with WRC 107 and FEA. Further, customized FEA software are also available for this analysis.
There still seems to be some confusion in users of these methods regarding selection of method for optimization of design. Users not familiar with Finite Element Method prefer to use calculations based on WRC 107/297.
Hang-Sung Lee, et.al. have recently (PVP 2011 – 57407) analyzed nozzle shell junctions using the Finite Element Method, compared their results with calculations to WRC 297 and made recommendations.
The work presented in this paper is not an attempt to compare individual stresses obtained by classical versus analytical methods. Instead, an attempt has been made to consolidate the results obtained by the various methods into charts to enable a user to make a preliminary assessment to ascertain under what geometrical conditions the calculations made by each of the above methods would result in overall Code acceptable stresses without the results being either overly conservative or un-conservative. This is particularly relevant to the geometries which use the graphs and charts which have been extrapolated without rigorous theoretical background in the WRC Bulletin 537. The Finite Element Method has been used as the referee method.