Re: Pipe supports

On Feb 15, 2005, at 10:02 PM, Dharmendra Hingorani wrote:

> In such cases, although Caesar analysis will allow rotations at such
> supports, but in actual practice, due to the physical arrangement used
> for such supports, rotational DOF is also restrained although the same
> was not considered in stress analysis.
>
> How, prudent and logical are such assumptions?
Pipe networks are considered continuous frames for stress analysis. For structural loads, the more restraint you apply at the frame supports, the lower the member (pipe run) stress, but the more severe the support loading. For example a simply supported beam carries half again more moment than the same span with the ends clamped. The supports for the simple span are more lightly loaded because the clamped supports have to carry end moments. Thermal loads are the exception because they're statically indeterminant and are actually caused by displacement constraints.

As a conservative rule of thumb design the supports for the maximum possible restraint conditions but size the pipe for the minimum restraint. This will require a certain amount of judgment, lest you get grotesquely conservative, and do something silly like size a hanger to take bending. It's always a good idea when you relax the restraint applied by a support, to go back and check the actual design to see if it allows for the calculated movement. You can usually do this quick and cheap. remember, too, that the as-built system may have its own ideas about how to make itself comfortable that may not jibe with your idealization.
Christopher Wright P.E. |"They couldn't hit an elephant at chrisw@skypoint.com | this distance" (last words of Gen.

.......................................| John Sedgwick, Spotsylvania