Forces & Moments for Different Load Cases

Hi All,

I am doing a riser modelling analysis and run the following cases:-
(1) LC = W+P1 (weight + internal pressure, 120 barg)
(2) LC = P1 (internal pressure, 120 barg)
(3) LC = T1 (content temperature, 65 deg C)
(4) LC = W+P1+T1

I have the following observations which i do not understand.
1. The global element forces & moments for all nodes are the same for case (1) & (2).
2. The global element forces & moments for case (2) is zero for all nodes.
3. The addition of global element forces & moments for case (1) & (3) is not the same as case (4).

Please advice.

Regarding your comments:

1) Unless your pipe density is zero, this doesn't make any sense.

2) In a pipe flexibility analysis, pressure has no effect on the system (assuming you don't have expansion joints or Bourdon turned on) displacements and loads. Pressure is only used in the "Code stress" equations as a contribution to the "longitudinal pressure stress".

So if the element forces & moments are zero in case 2, and you said case 1 was the same, then I would again assume the density of the pipe material is zero.

3) The principle of "superpostion" does not necessarily apply to piping systems. See "Formulas for Stress and Strain" by Roark & Young, Section 3.2, paragraph 2. This is especially true if the system you've modeled includes non-linear boundary conditions.

1) The pipe density for the riser model is not zero. The global forces & moments for all nodes for (1) LC = W+P1 and (2) LC = W (Previously mistakenly written as LC = P1) is the same.

2) Pressure will cause end-cap effect on the expansion spool bend section and riser bottom bend section. The forces and moments should not be entirely zero for LC = P1. Please advice.

3) In my riser model, "superposition" is no longer valid. Can CAESER capture and present the force and moment results due to two separate design load case [FUNCTIONAL = W + P1 + T1] and [ENVIRONMENTAL = D1 + WAV1] for the utility check in compliance with DNV OS F101? Pelase advice.

1) Ok, that makes sense (see #2 below).

2) No it won't, there is no "end cap" effect. The assumption (as previously stated) is that pressure causes no displacements (or loads) on the piping system. Pressure is only used in the stress equations - i.e. the PD/4t term. This is why "W+P1" yields the same element forces as "W".

However, you can turn on the Bourdon Pressure option to account for the pressure strain in the pipe. You can find this option on the "Special Execution Options" dialog.

3) Yes:

LC1 = W + P1 + T1
LC2 = D1 + WAV1