During piping installation and erection, we found that some of our prefab pipe spools where too short, or misaligned with the mating connections. The spools under consideration are very small diameter (1"-2"), small bore piping (80S), and therefore quite rigid.To assemble the flange (i.e. inserting a gasket and tightening the bolts) our technicians have had to use a whole lot of tools like jimmy's, tension bands etc. to make sure to get a good alignment of the flanges and their bolt holes, before tightening the bolts. The amount of force required was significant.
As a starter, I know the misalignment is the issue and needs to be solved (by cutting and re-wleding the spool to fit), rather than what Im about to ask here .. however, what Im trying to do with AutoPIPE is model this effect of the pulling force on the flanges, and review the impact which this pre-imposed stress has on the code allowable stresses for the various operating cases.
What would be the correct way to do this in AutoPIPE? Currently, my flanges are modelled as rigid pipe ends, with flange inserted (yellow ring), which with an anchor (which models the flange of the equipment to which the spool connects).
Thanks in advance.
Been messing around with this problem and came up with an idea. Would the following method be a good way to model the issue?
At the point where I modelled the flange and anchor, I insert an Imposed Support Displacement, with displacements equal to the misalignment in x, y, z, Rx, Ry and Rz. Case to combine with is GR, since the flange is pulled towards the mating/connecting flange of the equipment during installation, i.e. in the cold (gravity) condition.
Would this be correct?
Hubert,
AutoPIPE cut-short / cut-long analysis may not work in the current case as you seem to have displacements / rotations in directions other than the axial direction, otherwise that could have been your first choice.
Imposed support displacements are applied to the base of a support. Any imposed support displacement would displace the base / ground by the prescribed amount, exerting a force on the piping node which depends on the support stiffness in the direction of the displacement. It could potentially work if you know the amount of misalignment, however it may not be a true representation of pre-stressed pipe like cut-short analysis.
Could it be possible to confirm the results of imposed support displacement modeling by comparing the pipe forces result from the program with the force required to align the pipe (if you have recorded that force)?
Regards,
Bilal Shah
Hi Bilal,
Cut-short was my first guess as well, but I quickly found it wouldn't give a proper representation of the issue. ISD (Imposed Support Displacement) seems to work from a visual point of view, i.e. if I plot the deflection for the gravity case it accurately shows how the pipe 'sits' in gravity. The pipe will move at the node under review, equal to the amount of misalignment (which makes sense, as that is what I modelled). Stress/strain point of view is what I know need to determine to see if this is the way to go.
The flange is a 1" B16.5 Cl 2500 316 flange. Misalignment is about 40 mm down, 10 mm sideways (i.e. axial) and slightly less than 1 degree rotation. The stresses for the code combinations, when modelling the ISD, go up to 250% for the sustained case, opposed to ~75% for the case with no ISD (i.e. perfectly aligned flanges).
I have no idea how much the force is. However as we have the piping system in our fabrication shop, I helped the construction guys a hand during disassembly. I found that I was able to push the pipe spool in the right direction for about 50% of the misalignment, but not completely. Reviewing the froce in the pipe within AutoPIPE for the gravity case I find a resulting force Fr of ~2 kN. This seems to match the order of magnitude I was pushing.
I therefore conclude this method seems feasible, and that ISD's also work for anchors. An alternative could be to replace the anchor at the node with a series of supports which model an anchor (thus restricting all 6 DOF's), and then apply the ISD to that node/support.
Hi, I have a similar problem. (Pleas correct me if i misundestood)....Summarizing what yo did to "model this effect of the pulling force on the flanges, and review the impact which this pre-imposed stress has on the code allowable stresses for the various operating cases" was the following:
1-Run your model unconnected to the flange (the flange node wiht 6 DOFs) considering only gravity
2-Check the defelection due to gravity is similar o equal to the measured misalignment.
3-Run a new model with an anchor in the flange node considering only gravity
4-The loads on the anchor are an estimte of the load needed to aling.
Is this correct?
In my case (it was some time ago, so Im not 100% sure anymore), I had a bracket pretty close to the flange. Therefore, deflection due to gravity was not an issue that much for my case, hence I didnt model that. Not sure if that would matter for your case.
I created a flange with an anchor at the flange (I usually do this to be conservative). This resembles the nozzle of my equipment to which I connect. I then added an Imposed Support Displacement, which I combine with my GR load case. I then insert the deflection in this Imposed Support Displacement and evaluate the results.
Hope this helps.