Hot sustained stress due to well displacement

[Folmer]

After doing some searching on the web I found some information about hot sustained stress in a piping system. This can be analysed by removing the lift-off supports from the system and then checking the sustained stress.
For a "regular" piping system this works as lift-off is mainly due to the piping expanding. However I am currently working on a piping system where there is a large displacement due to well growth (250 mm) which causes lift-off in the hot condition.

The sustained stress can now be checked in two ways, the question I have is which is more realistic?
1) Put an imposed displacement on the wellhead connection and add this to the gravity case
or
2) Remove all the lift-off supports as you would normally do.

The first gives significantly higher stresses than the second. But is this realistic, or unnecessarily conservative?

I am working with AutoPIPE V9.05

 

[BigInch]

Can't say about Autopipe.

Method 2 sounds like that there is some growth allowed from a deep virtual anchor, maybe 1000 meters below ground level, such growth, being only partially restricted, is reducing axial stress somewhat.

Method 1 sounds like there is an assumed rigid anchor nearby, not a deep virtual anchor, and your imposed displacements translate into direct stress.

Not having details of the model, I'm not sure what might be more realistic. I would expect a deep modeled virtual anchor is more realistic. I am always after my guys to remove these artificially simplistic imposed anchors (just because the pipe stress programs won't work without an anchor) and try to model true growth over virtual anchor distances and the accompanying stress reductions that follow.

Independent events are seldomly independent.

 

[Folmer]

To clarify the problem a bit, I'am analysing the topside piping system of a gas production wellhead platform. Due to the temperature of the produced gas the x-mass tree connections may move upwards as much as 250 mm.

At the moment I have modelled this connection as a rigid anchor with thermal anchor displacment. Due to this displacment there are several supports with lift-off so a hot sustained run is necessary. When using approach 1, as described above, obiously the deformed situation (x-mass tree displacment lifting the piping plus gravity) is very different to that when using approach 2, (the system deforms under gravity only).

As the hot sustained run is supposed to look at the sustained stress in the hot deformed situation, I am inclined to think that approach 1 is more realistic, but as the stresses are significantly higher I am worried that it that approach might be too conservative.

 

[BigInch]

If you have determined the displacements due to both gravity and thermal expansion accurately, method 1 is the correct way to model the system. But it depends on precisely how you have determined those displacements. If the attached piping which you are about to model is rigid enough to influence those displacements, then those displacements may not have been as accurately determined as you thought.

Independent events are seldomly independent.

 

[C2it]

You could have a look on the Intergraph / Caesar II forum. Hot sustained stress has many posts there.

 

[DSB123]

Simples. The thermal displacement of the Xmas tree is caused by temperature therefore it's effects should be included in the stress range check (i.e. check stress against SA). For the sustainded stress check (hot) ignore the thermal displacement but remove all supports at which the pipe will lift off due to the thermal growth.