Not sure what you mean by a radial load, but I think not. Nozzle pressure thrust acts along a line generally perpendicular to the nozzle exit or throat area, i.e. an axial load.
Thanks for your response. Yes, we are talking the same direction. Forces along the axis of the nozzle are termed radial loads in WRC-107 since they act on the shell toward the center or axis of the vessel.
This is an old debate, that never came to a clear issue.
In my opinion these are the facts:
- most client specifications do not explicitly state whether pressure end loads are included in their standard nozzle loads; I suspect they are normally (but silently) included
- as far as primary stresses are concerned, the verification of nozzle reinforcement takes care of nozzle end loads; so, if pressure is the only source of stress and design is per Div.1 (no check of secondary stresses required) there is no need of WRC calculations
- when other loads on the nozzle exist (here a difference should be made for loads of thermal origin, but this is a completely different discussion) or secondary stresses need be evaluated (Div.2), then the pressure end effects should be included in WRC calcs, as this will be a source of bending in the vessel wall near the nozzle; for this calculation the actual geometry of nozzle (including any reinforcement) should of course be used, but also the longitudinal and circumferential stresses in vessel wall should be included with no intensification (plainly PD/2t or PD/4t, t being the design thickness at the nozzle location)
- if I remember correctly, BS5500 had a different approach on this point: pressure stresses are intensified by a suitable coefficient, and I suppose this was to include the effect of local bending: in that case the end load should be excluded from nozzle loads.
External radial loads on a nozzle as a result of pressure only do no exist since the vessel and attached piping are in a state of equilibrium. This will be true if the connections are rigid (welded, flanged, etc). Exceptions would be expansion joints which are free to move radially. Slug loads (2 phase flow) may produce external loads and requires special consideration.
Just curious: If you take a free body diagram at a section cut through the nozzle, say just before the flange, I would argue that at that cut the nozzle neck has a longitudinal stress which equals P*R/2*t. This longitudinal stress, in units of force per unit area could be multiplied by the cross sectional area of the cut to develop a load which is longitudinal in the nozzle... or radial to the vessel shell.
If you state that these radial loads due to pressure do not exist, how do you explain the presence of longitudinal stress in the nozzle neck, or does the longitudinal stress also not exist?
Primary long and circ stresses in the nozzle and the shell obviously exist but are contained by the nozzle reinforcement. For this reason you don't want to be doing WRC calc's on manways. Primary stresses in the shell are part of the WRC analysis to satisfy PL = 1.5Sm max. and PL+Pb+Q = 3Sm max.
Thanks everyone for your input. I'll ponder your responses. As a side note, I believe that Caesar II has an option in its WRC-107 module to allow the inclusion of pressure thrust.
Pressure thrust is a physical load on the nozzle and should be included in WRC 107 type calculations. It is no different than any other axial piping load. It causes local bending stresses in the shell, which are different that the hoop stresses caused by internal pressure. However, it is not a load on the equipment foundation, because it is balanced by the load on the vessel wall opposite the nozzle.
For most machinery, I have found that the pressure thrust load is not included, as the major concern is overall distortion and misalignment, not nozzle stress. There are exceptions.
Thanks very much for your valuable input. For the benefit of others COADE has published on their website a few informative articles on this subject in their Mechanical Engineering News. See July 2001 page 9 and June 1997 page 10. From their articles it can be seen that pressure thrust should generally be included in WRC-107 calculations unless the piping system has restraints in place to counter the force.
First, pressure thrust for nozzles is the cross sectional area of the nozzle(in^2) X the pressure (psi) to give a load in lbs. This is the loading that gives you longitudinal stress in the nozzle neck (or the vessel itself). Does this load translate into an "external radial load"? I think not, and here's why. All pressure thrust loads are equal and balanced in each direction by the stressing of the metal. Think of it this way....if you lined up all the manways on one side of a vertical vessel, would the vessel itself fall over? Of course not. You will have produced longitudinal stress in each nozzle neck, but the pressure thrust loads balance out. A stress is produced, but not an external load. This is where the confusion is coming from.
It is an external load on the nozzle, a very real one. Vesselanon, how do you think the load gets transferred to the vessel to balance against the pressure thrust load on the opposite wall? It is NOT a load on the foundation, as it is balanced, unless there are expansion joints in the system that do not include hardware that carries the pressure thrust forces across them.
To help you better understand the subject of nozzle thrust and how it may or may not be influenced by attached piping and why it exists around openings in general, I would suggest that you view the information COADE has published on their website. See their Mechanical Engineering News, July 2001 page 9 and June 1997 page 10.
