Strategy for Modeling Piping and Vessel/Tank Nozzles
Strategy for Modeling Piping and Vessel/Tank Nozzles
(OP)
Good afternoon. I would like to hear some discussion on how you all are handling the evaluation of piping connecting to vessel nozzles and tank nozzles. This question comes up all the time, in nearly every project we do, because all the projects in my area have thermal and seismic loading.
The question is: how to economically handle the evaluation of applied nozzle loadings. Yes, there are special FEA tools for nozzle loading available and there are plenty of papers written on the topic. For a very small project, e.g. connecting a 30" tank vapor recovery line to the roof of an existing API 650 or 12B tank, where the total engineering budget for the project is only a few thousand bucks, one would like to have available a table of go-by allowables that one could reference after running his pipe stress model to determine if he needs to provide an expansion joint. I am thinking it is not practical to run FEA on every single nozzle in a project. In the 'old days' before FEA, how did guys evaluate nozzle loading?
The idea of course is to get the job done as fast as possible while complying with all applicable codes. Perhaps nowadays that's just what one has to do - model every nozzle in FEA since that approach could be held up as 'the expected appropriate level of care and custody as would be exercised by another reasonably competent person' (just used up all my legal weasel words) in the event of a failure and subsequent legal action.
Then of course one gets the owner's engineers saying "Boy, we done it this way for a hunnerd years and we never ripped off no nozzles or cracked no welds yet...". You'd be surprised how many times I've heard that.
In short: I suspect there has to be a shortcut method to evaluate vessel and tank nozzle loadings. If so, I have not come across it yet. Thanks in advance for your thoughts. Pete
The question is: how to economically handle the evaluation of applied nozzle loadings. Yes, there are special FEA tools for nozzle loading available and there are plenty of papers written on the topic. For a very small project, e.g. connecting a 30" tank vapor recovery line to the roof of an existing API 650 or 12B tank, where the total engineering budget for the project is only a few thousand bucks, one would like to have available a table of go-by allowables that one could reference after running his pipe stress model to determine if he needs to provide an expansion joint. I am thinking it is not practical to run FEA on every single nozzle in a project. In the 'old days' before FEA, how did guys evaluate nozzle loading?
The idea of course is to get the job done as fast as possible while complying with all applicable codes. Perhaps nowadays that's just what one has to do - model every nozzle in FEA since that approach could be held up as 'the expected appropriate level of care and custody as would be exercised by another reasonably competent person' (just used up all my legal weasel words) in the event of a failure and subsequent legal action.
Then of course one gets the owner's engineers saying "Boy, we done it this way for a hunnerd years and we never ripped off no nozzles or cracked no welds yet...". You'd be surprised how many times I've heard that.
In short: I suspect there has to be a shortcut method to evaluate vessel and tank nozzle loadings. If so, I have not come across it yet. Thanks in advance for your thoughts. Pete





RE: Strategy for Modeling Piping and Vessel/Tank Nozzles
RE: Strategy for Modeling Piping and Vessel/Tank Nozzles
As SNORGY says, both of these are included in most pipe stress programs.
My advice to most people is to skip the FEA. These canned programs still require some user awareness of the FEA tool, and most users don't have that awareness/skill. It's very easy to get very good looking wrong answers. Stick with the WRC methods or the API 650 Appendix P method.