| Applies To | |||
| Product(s): | AutoPIPE | ||
| Version(s): | All | ||
| Environment: | N/A | ||
| Area: | Combinations | ||
| Original Author: | Bentley Technical Support Group | ||
| Date Logged & Current Version | June. 2016 10.01.00.09 | ||
Problem:
An AutoPIPE model in ASME B31.8-2012 code. The model cover a pipeline section with a valve station. Considering that E2 = -E1:
Expect to obtain the same results for following code combinations:
- GRTP2+E1
- GRTP2-E2
AutoPIPE results are not as expected (they are very different!), WHY?
Solution:
Linear Analysis:
For a linear analysis, you will get the same results for “GRTP2+E1” and “GRTP2-E2(2)”. You can check that by unchecking the Non Linear checkbox for analysis set 2 and reviewing the output report:
Example:
Non-Linear Analysis:
Expect to get different results when the direction of a load has changed for a non-linear analysis. Why?
Support gaps in different directions (left/right, forward/backward, or up/down) may act differently when the load direction is reversed. For example, a straight pipe running in Global X direction having a guide support with 2” gap in the left and right direction. Support that the pipe moved 1” in the left direction for operating loads (GRPT2). Suppose that now E1 is applied forcing the pipe towards local right: The pipe has 3” gaps which it can move. Now suppose E2 is applied having same magnitude as E1, but opposite direction: E2 will only have 1” gap before the support is closed. Both the support forces and pipe forces will be different for E1 and E2 in this case, which will result in different stresses. Non-linear analysis hence is a more realistic approach, which yields better results.
Consider soil springs applied to a skewed pipe (pipe is not running parallel to any Global axis) and we run non-linear analysis. The up and down soil springs act independently, but also has different resistance and yield displacement values. Also, the earthquake load is applied in the global direction. Now consider that the earthquake load tending to move the pipe in global X direction is acting as shown below (E1 in positive global X, E2 in negative Global X):
Where the Up and Down (red soil springs) are the vertical soil springs in with different resistance and yield displacements. The Hor (Green soil spring) is the Horizontal Soil Spring. You can see that even when the magnitude of E1 is the same, the displacements that it would create are different based on the direction in which the load is applied and due to the differences in the soil spring resistance and yield values. The results for the pipe moments would be different for E1 and E2, which will lead to different stress results.
The considerations above are for very simple cases, and with the model that you sent there would be these and other considerations in action which would essentially lead to difference in stress results when direction of load changes for non-linear analysis. This is a more realistic approach and covers the reversing nature of loads like earthquake.
See Also
"Code Compliance" sub-report using Results> Output Report