Does anyone know what requirements are required for a formal pipe stress analysis. I know that ASME B31.3 319.4.2 states formal analysis requirements, but what I want to know is more specific requirements, for example pipe larger than DN400, pipes connected to sensitive equipments, pipes subject to slugflow, etc. These piping systems need formal pipe stress analysis.
Are there any books or guide line defining those specific requirements. Thanks

Hi Totan

Requirements? The ASME B31 Codes do not address the question you asked. I doubt that the PED will address it directly either. You will only get opinions.

As I look over to the right of this page I see that there are as of today 2405 members registered on this forum. That is potentially how many opinions you could hear. Of course, many corporations have their own requirements and these are usually well thought out and they apply to the specific plants that they operate. Any list printed in any book will be right (for some applications) and will be wrong (for other applications). I would be reluctant to make such a list.

There are parameters, as you alluded to, that should be considered.

Connections to strain sensitive equipment.
Minimum to maximum temperature range.
Minimum temperature.
Maximum temperature.
Operating temperature.
Internal pressure.
External pressure.
Pipe size and schedule.
Fabrication method - e.g., welded, flanged, etc.
Other issues - e.g., buried, double containment, etc.
Materials of construction - e.g., metallic, glass, plastic, etc.
Phase of the moon
What side of the bed Herr Luf got out of this AM...

This ought to be an interesting thread.

Regards, John

Only analyze lines that...


Are required by contract or...

are over code stress limits or over load load, sensitive equipment....

Clear??? of course not!

If it is cost plus contract then it is absolutely necessary to calculate 1/8" lines.

There is a paragraph on this topic in Process Piping The Complete Guide to ASME B31.3 by Charles Becht IV, ASME Press, Second edition is ISBN 0-7918-0217-5. Author is a member of ASME comitee for B31.3 so I assume you can follow what ever he recommends.

Hi Jouko,

That is exactly why "I would be reluctant to make such a list". Our colleague, Dr. Becht's opinions are sound and based upon years of experience in both field work and analytical methodologies. And I might add that Dr. Becht is the chairman of the B31.3 Committee. However, it is important to recognize that his opinions are HIS opinions - not the opinions of ASME or of any Code Committee.

We are all still individually responsible for the engineering decisions we make - See B31.3, paragraph 300,(b),(2).

I will ask Dr. Becht if he will comment on this thread.

Regards, John.

Gentlemen,

The B31.3 code eloquently states in a sense what I stated.... I admit a bit of sarcasm in my post but in essence if you cannot be sure whether a system is overloaded or over stressed then you should analyze... the judgement of the designer is critical.... outside of any contractural obligations of course!

Formal pipe stress, as a term used in ASME B31.3, is NOT equivalent to pipe stress analysis on the computer. Formal analysis requirements are in 319.4.2. Note that formal analysis can be by simplified methods.

Formal pipe stress is required for all lines that do not meet one of the conditions in 319.4.1, which include "judged adequate by comparison with previously analyzed systems." So judgement is permitted.

Companies have various rules of thumb as to what systems are more critical and should be evaulated by computer analysis. The PIP standards also have such criteria. These are guidelines.

Not to hijack this thread...

Jouko points to Chuck Becht's book. I find his Appendix II quite useful - Guidelines for Computer Flexibility Analysis. It's a four page checklist of things you should consider when running an analysis. It doesn't say what lines must be checked, but, if you are going to run a computer analysis, it'll give you a lot of things to consider.

It's not so much what the computer tells you, it's what you tell the computer.

I wish people would know the difference between formal stress analysis and computer analysis and good design and bad design. A lot of time and money could be saved. I trust more good pipe engineer's handwritten note on a draft drawing "Stresses and supports are OK" than 5 day stress engineer's computer calculation that shows no stress issues. For me the first option is more in line with "formal stress analysis" than the second.

I just spent couple days walking on hot steam pipes at temperatures over 30C. According to computer analysis lines are perfect. On site you see different. Pipes are walking around as most of the restrants are missing. I describe lines as bananas in the report.

Thanks for writing the book.