which piping code???

[stephenebrady]

I have been involved in pipe stress analysis for several years in process industries and have used CAESAR as a design tool, and ASME B31.3 as the design code. However, I have come across some calculations for large diameter (1500mm to 3200mm Diam) piping in the Iron & Steel industry, Blast Furnace gas, coke oven gas etc. where the specified code is B31.2?
Having looked around for B31.2 I have found that it was withdrawn in the 60's but parts are incorporated into B31.8, and CAESAR has an option to used B31.8.

Is B31.3 acceptable to use for this application, or would it be more correct to used B31.8? and what are the major differences??

 

[DSB123]

stephenebrady,
Suggest you should not be using Caesar for design of large diameter piping as this is only a flexibility analysis program. With large diameter piping you have "local" effects which must be considered and Caesar does not address. Also if the piping is for Furnace gas then the ducting will probably refractory lined, another complication.

 

[stephenebrady]

CAESAR is used as a design tool and forms part of a complete package of calculations to cover the design of the main.
We make sure the flexibility analysis conforms to the rules of B31.3 and then move on to cover the finer aspects of the design.

For this particular case, assume that it is NOT large diam.

Any other thoughts on my original question?? (refratory lined or not)

 

[bvi]

The specification of B31.2 must be an error since that code covered fuel gas piping.

Either B31.1 or B31.3 would be acceptable choices, in my opinion, although B31.1 is more commonly chosen in an industrial (non-process plant) setting.

It is the Owner's responsibility to select the piping code that they wish to be used; if they have not selected a code, make a recommendation to them and get their concurrence.

 

[JohnBreen]

Hi Steve,

I think the thing that you must keep in mind is that Caesar II is a beam theory program. It calculates bending stresses as the intensified moment divided by the GROSS section modulus. For most pipe stress analysis this is appropriate (and it is what the B31 Codes prescribe).

When the wall thickness to diameter ratio is very small the "pipe" (or "duct") often sees membrane bending stresses that are larger than those that would be calculated by beam theory (especially at support locations). Also consider that "ovalization" can occure at support points and the calculated section modulus of a "round" pipe is no longer accurate. Therefore, the beam theory bending stresses would not be accurate. This really becomes an important issue if one of your "thin-wall" systems is filled with water for a leak test.

I would continue to use MOST of the guidance provided by B31.3 for your application, however you must also consider loading conditions (and local wall flexibility) that are not addressed by B31.3. Get a copy of AWWA M-11 and study it, especially the parts that discuss support. You may have to design "ring stiffeners" for some of your systems that have longer spans between supports. Take a look at some of the classic textbooks on horizontal pressure vessel design - especially the discussions of "saddle supports" and "ring stiffeners". One or more of the classic (and inexpensive) Lincoln Arc Welding Institute books also address relatively thin wall tubular structures. Also, find out about a software package called FEPIPE by Tony Paulin - it might be appropriate for you to "go a step further" in some of your analyses.

You can still get a copy of B31.2 from ASME but its status is now just "background reference". When B31.2 was withdrawn, its wisdom was mostly incorporated in ANSI Z223.1 (this is also part of the NFPA Standard). B31.8 is a "buried piping" Code and so it really does not address the systems that you describe.

You are very correct to bring up this topic for discussion. You are not the first to become concerned about the adequacy of "following the sheep" in pipe stress analysis. The B31 Codes are by design simplified methodologies and they are not intended to be cookbooks. The piping engineers is charged with the responsibility to evaluate his/her designs completely.

Thanks for posting your topic for discussion.

Best regards, John.

 

[DSB123]

JohnBreen,
Good post John. Basically what I was getting at. Local effects are more important and Caesar is not a tool to be used for this type of pipework. Doing a run on Caesar will probably say the piping is O.K. but if "local" issues are not addressed then expect failure.