Suppose a +Y pipe support is modeled with a friction coefficient and the calculated operating load is less than the sustained load. As the system comes up to temperature the pipe may be sliding while the normal force is somewhere between the cold and hot loads. Are the calculated horizontal friction forces potentially too low for support design?

You have a good point.First I would never recommend that we consider the horizontal friction load from Pipe stress programme output for support design. Friction analysis is not deterministic but stochastic.It is a numerically iterative solution and can vary significantly depending on the selection of Friction stiffness. To me it is just to give the user a "feel" of how system can behave with frictional resistance.To me it is significant only from the standpoint of qualifying an equipment load ( that too if friction is helping to reduce the loads, such beneficial effect should not be considered ). Mathematically friction load at a support cannot be more than coefficient of static friction times normal load ( at condition when the pipe is about to slide).The max. normal load ( max. of all conditions ) should be multiplied with the coefficient of static friction to arrive at the value of friction load that has to be considered for support design.


Anindya Bhattacharya

Stress Analyst

Bechtel Corporation

I agree with the comments from Anindya above.
The engineer should understand how these loads are generated and consider how sensitive they are to the assumptions made.

A typical example of this is a long length of pipe in a pipe rack with an anchor in the centre.
If this pipe is symmetric either side of the anchor then the anchor loads predicted will be zero which is not the case in reality. There are many assumptions made that will increase this load such as; during startup there may be a temperature differential either side of the anchor; construction tolerances; a source of vibration at any support will reduce the real frictional force generated; etc…
If you consider the pipe on one side of the anchor only the predicted resultant axial load on the anchor from friction on each support along the rack would be very large. Most pipe rack structures could not take such loads.

The piping system should be designed with good engineering judgment and the use of norms built up over the years.