Tips July - September 2004

Note on Branch Connection Pressure Design

Recently there was some discussion regarding the branch connection pressure design rules in B31.3, Para. 304.3.3. Some differences were noted between the rules (also referred to as the area replacement rules) as manifested in the current equations in B31.3-2002 and Para. 104.3.1 of B31.1-2001 including the 2002 Addenda. The discussion ended with the conclusion that there should be no difference between the B31.1 and B31.3 equations except in the unique nomenclature used by each.

B31.1 has recently spent a considerable effort to try to make its area replacement equations correct, and they can be assumed to be correct (unless the ASME editors have messed them up -- a situation possible to encounter).

A way to check branch connection pressure design rules is to review the equations so as to assure the designer that there is an adequate wall thickness at the end of plant life, i.e., from the run and branch piping remove the manufacturer's under thickness tolerance, remove the corrosion allowance (on the inside) and the remaining metal must satisfy the area replacement rules. That is, the amount of remaining metal removed that is required for pressure design must be at least replaced within the reinforcement zone. If one follows this concept the designer can always be assured of meeting the B31.1 or B31.3 requirements.

In addition, there should be no difference between the rules or equations for branch connection pressure design in any B31 book (or for nozzles in any ASME boiler and pressure vessel code), with the exception that some B31 books consider either manufacturer's underthickness tolerances or corrosion allowances or both as negligible, e.g., B31.4 and B31.8.

Author: Mr. Ron Haupt, P. E., of Pressure Piping Engineering ( is a member of several piping code committees (B31, B31.1, B31.3, BPTCS, and others). He consults with us in the capacity of Nuclear QA Manager.

To Include or Not to Include Hanger Stiffnesses?

In the 1970s, performing pipe stress analysis on mainframe computers was expensive (about $1000/run). So, pipe stress analysts would run a Thermal (T1) case alone first. Then, they would run a deadweight (DW) case with rigid supports at chosen hanger locations. Using the support loads and the hanger travel, analysts would select appropriate variable spring hangers from hanger manufacturer catalogs. To save on computing costs, they would not update the global stiffness matrix [K] with the hanger stiffnesses from the newly selected hangers because they would have to reanalyze the model which would cost more money. Many industrial plants built 30 or 40 years ago in the USA and presumably in other countries have pipe stress analyses done in this manner. These plants have archived such analysis reports that do not include hanger stiffnesses (as part of the global stiffness matrix [K]).

In later years, as and when the analysts in these plants needed to reanalyze those piping systems, they had to use modern programs like CAEPIPE. But, before they accepted the new results, they sometimes liked to verify that the new results (from CAEPIPE ) matched the results from the old reports generated by the mainframe pipe stress analysis programs. This is the reason why CAEPIPE provides the option "Do not include (or include) hanger stiffnesses." It helps the engineers compare results from CAEPIPE with the old archived reports.

Today, with cheap PC computing power, there is no reason why hanger stiffnesses should be excluded from analysis. In fact, including hanger stiffnesses provides a more accurate picture of system behavior. As such, we recommend that you "include hanger stiffnesses" in every analysis. Due to the limited utility of this option, we plan to remove the "Do not include hanger stiffnesses" option from CAEPIPE in the near future, and include hanger stiffnesses for every analysis.

Bourdon Effect Straight Pipe

In continuation to our tip on pressure expansion, we have included here a technical verification of this feature. As the previous article says, the expansion due to internal pressure was usually ignored in metallic pipes previously, but if you are using non-metallic pipes or have a high pressure system, the effect can be significant. So, it is recommended that you always include this option during model analysis.

Download the document here
Download the CAEPIPE model here
(Right click, and select "Save link Target As..." and make sure you type in .mod for the extension).