Dear all,

We are attempting to Provide "hydrostatic" data to our structural department from our pipe model. The Structural Department requests that forces imparted by the "fluid alone" at each support location during a hydrostatic test. We would like to take into account non-linearity effects during the hydrostatic test.

Our proposals are:
Load scenario 1
1. WW+HP+H (hydro test)
2. WNC+P1+H (dead load)
3. L1-L2 <--- "fluid alone load"

Load scenario 2
1. WW-WNC + H "fluid alone load" <-- where hangers are set to rigid

Which case is right, or is there a better scenario?

Also, does load scenario 2 carry any non-linearity responses if the model has non-linear items?

Do the added items in load scenario 1 (HP, P1) impart non-linearities that are valid?

super thanks,
mervyn

I'd stay away from "scenario 2". Go with "scenario 1", it is much easier to control and explain.



Loads are not "non-linear", only certain boundary conditions. So, "scenario 2" has a certain static response which is a function of the defined boundary conditions. Your model may or may not have non-linear boundary conditions that affect the final result. "HP and P1" don't impart non-linearities.

Richard,

I believe I understand the concept of non-linear boundary condtions. When referring to "non-linear items", I suppose I mean "non-linear boudary conditions", such as gaps, restraints with friction, limits, etc.

The model does have "non-linear boundary conditions". However, my question is when do these boundary conditions come into play?

I can see that during an expansion situation where the pipe grows and moves due to a temperature change, these boundary conditions would definitely play a role.

However, in sustained cases such as W + P, how does this affect non-linear boundary conditions? I would expect P loads and HP loads to have minimal affect on restraints altogether.

PS. could you explain why:
"I'd stay away from "scenario 2". Go with "scenario 1", it is much easier to control and explain" - why is scenario 1 easier to control? I'm having a tough time explaining why P1 should be accounted for during a hydrotest:

Load scenario 1
1. WW+HP+H (hydro test)
2. WNC+P1+H (dead load)
3. L1-L2 <--- "fluid alone load"

Nonlinear boundary conditions come into play depending on the loads applied to the system. When you subtract one load case from another you see the effects of nonlinear boundary conditions. OPE and SUS cases do not see nonlinear effects themselves because they are starting from the neutral position (middle of the gaps, resting on the +Y supports, etc.). But combination cases can see the nonlinear effects because the two load cases being combined can be behaving differently (pipe in contact on different sides of the gaps, pipe lifting off a support in one load case and not in the other, etc.). I think this is why scenario 2 is not a good idea, because it is starting from the neutral position. Scenario 1 is a difference between two load cases that may be restrained differently due to the nonlinear restraints.

Be sure you set the hanger stiffness to Rigid on both load cases in your scenario 1. Pressure won't make any difference in the loads on your restraints, only longitudinal pressure stress is affected. But you won't care about the stress in your combination case, so it doesn't matter if you use WNC+P1+H or WNC+HP+H in your second load case.