The CAESAR II buried modeler will compute the appropriate axial stress in the fully restrained buried pipeline segments as well as the partially restrained buried pipeline segments that are consistent with the design code equations. However, in order to accomplish this, the appropriate input specifications / settings / commands need to be activated /invoked, and vary somewhat with the different code check options.

The primary items to check are as follows:

1) Make sure that the appropriate specification is activated such that the buried modeler knows which segment(s)are buried (are to be restrained by soil springs) and which segments are not. If this is not properly accomplished, the axial stresses reported by CEASAR II will be erroneous.

2) Most code equations use the outside diameter for computing stresses in pipelines. Check to see if this option is properly specified. By default CAESAR II will use the inside diameter, and the stresses computed by CEASAR II will diverge from the code equations that use the outside diameter.

3) Based on experience, it is always "safe" to invoke the "Bordon Effects" option for buried pipeline analysis, especially when evaluating transition areas into above ground scraper trap locations.

When visually interrogating the axial stress report, look for where the axial stress becomes effectively "constant" from one node to the next. This is a good indication that this reach of pipe is "fully restrained". This relatively "constant" axial stress is the stress to check against the code equations. You will also find that this reach of buried pipe is remote from buried bend locations and above ground/below ground pipeline transitions. In this manner, if your manual check against the code equations for restrained pipe axial stress does not closely match the CEASAR II report results, either the pipe is indeed not fully restrained (since it is too close to buried bends or transitions) or something else is causing CAESAR II not to recognize that the pipe is buried and is restrained by soil springs (i.e. refer to items 1 thru 3 above).

If you cannot figure out the problem, it is always prudent to model a simple buried pipe problem consisting of a long straight section of buried pipe (at least 3 to 4 km) at a constant cover depth that protrudes straight out from the ground (with no bends) at each end for 6 inches or less, and is flanged or capped at both ends. Both ends are specified free to move. You can then experiment with different internal pressure and delta T load combinations. You will find that there is a distance from each end where the pipe is only partially restrained and exhibits nodal displacements, and then reaches a point called the "virtual anchor" where the pipe becomes fully restrained by the soil axial friction springs and there is no nodal displacements at these fully restrained nodes. The axial stress at these fully restrained nodes is the stress to check the code restrained pipe equations against.

If you are unable reach a reasonable comparison of the code equation check against the CAESAR II results for the restrained axial stress, then there is something still amiss with your CAESAR II input information / specification / settings in the buried modeler. However, if you reach a reasonable comparison of axial stress results, then your input specification / settings in the CEASAR II buried modeler are correct, and you can repeat the same settings in your more complex pipeline configuration model.

The simple problem described above has a closed form solution approximation, as described by P. J. Schnackenberg, "Pipeline Rules of Thumb Handbook". In order to closely match this closed form solution analysis against the CEASAR II simple model described above, you will need to specify a very large stiffness for the CAESAR II buried modeler axial restraint soil spring such that the full ultimate axial soil restraint is mobilized at the slightest pipe movement.

Ken Nyman