Dear all,

During the stress analysis of jacketed piping, the jacket pipe will be welded to the core pipe with the help of a cap or a blank or even a reducer (Partial jacketing Systems.)Since there is a difference in temperature between the core pipe and jacketed pipe, there shall be big axial force on the weld junctions because of the differential expansion (where the jacket pipe is welded to the core pipe). The weld strength is to be checked for these axial loads.

If we find the weld strength is inadequate for the axial force calculated, what are the ways of qualifying the same. Is increasing the welding thickness the only option ? or there are some other methods of doing the same. Please suggest.

Thank you all. Have a great day ahead.

If the core pipe weld is inadaquate you cannot qualify it. It's no good. Presumingthat you are interested in ending the jacket at a flange, you could avoid welding to the core pipe by using a reducing (jacket) flange, similar to a slip-on flange. The bore is of core pipe size but the bolting is jacket pipe size, the jacket being welded to the back of the flange. These flanges are available commercially.

If that's no good you could consider using a bellows in the jacket pipe.

Thanks Moverz,

I also had the same idea. Usually the difference in temperature between jacket pipe and core pipe will be in the range of 10Deg C to 30 Deg C. But in one of our case the difference is 110Deg C. I found this to be impossible to solve. let's check with the process guys regarding the temperature difference. Most of the complicated problems are easily solved this way.

By the way while modeling the this kind of jacketed pipe in caesar should we have anchors on the jacket pipe with cnodes to core pipe at both ends (at both weld joints) or anchor on the jacket pipe with cnode to core pipe at one end leave the other end free (I refer to the straight portion of the pipe)

Thanks and have a great day ahead.

Damien,

For details of jacketed pipe analysis, particularly for checking the stresses at the weld mentioned by you, in my opinion, a finite element analysis is required.

If you have ANSYS or ABAQUS you can try in any of this way:

1) Use shell element based modelling. This will give you an estimate of the stresses in the model.You can estimate the stresses in the weld by use of FSRF.

2) Use submodelling technique to evaluate the stresses in the weld where 3D Elements will be used for analysis of the location of interest based on displacements taken from remainder of the shell based model or use a shell to solid coupling where you again use 3D elements to model location of interest and shell elements for the remaining part. Both these techniques will minimize the computational effort required for a full 3D model.


3) There are some excellent discussions in a recently concluded webinar by PRG on Fatigue( presenter Chris Hinnat). Look for details of fatigue considerations in this webinar.

Hope this helps.

Regards