HI SuperG,

You need to look at a primer for fracture mechanics.  Learn about "notch toughness" and critical crack lengths for various materials at various stresses.

Regards, John.

I recently did a search on Google for Pipe fracture mechanics and came across several technical papers on failures of pipelines that covered this subject. Try the National Transport Safety Board Website. Sorry I dont have the address.

Pat Leavers at Imperial College has done a lot of work on this for gas and water mains pipes.  I'm sure he'll be able to answer all of your questions.

Check out any of a number of sources related stress crack corrosion.  I'm in the natural gas industry and pretty much nobody worries about this issue unless you have one of the following:

a) a pipeline above 1000 psig (steel)
b) a pipeline operating above 40% SMYS
c) a pipeline with low toughness steel (i.e., highly "brittle")

The analysis is not trivial.  It is usually done with tons of statistics.  Such as:

1) What is the size and orientation of the maximum defect?  At what confidence interval?
2) How many pressure cycles will the pipeline see in it's lifetime (or time until potential failure)
3) Type of fluid and potential energy when under pressure

Good luck.  I've started on this in the past and found work-arounds so I didn't have to solve it.

I know that API 579 "Fitness-For Service" (published January 2000) has an extensive section dealing with Assessment of Crack-Like Flaws.  These assesments include all kinds of formula for determining whether a flaw will propogate in the operating environment.  You might want to get your hands on this document to see if there is anything in there that will help you with your investigation.  Be warned, this is the type of document that you will probably have to sit down and study for quite some time to fully utilize.  I have not personally used the section on cracks, but we have used this document for some interesting purposes such as accepting misfabricated piping components that had extremely long lead times.

Hope this helps.

What you have to do is consider the worst possible defect size and where it’s likely to occur.  For example one practice is to assume that one run of weld in a multi run joint has no side fusion.  You then have to decide what the toughness of the material is; an approximate correlation can be gained from impact tests.

You need British Standard BS7910, which will answer your question.

Regards
John   post@gowelding.com  www.gowelding.com