Pressure Thrust in PV Elite and FEA 3

[Mohit_singh]

Hi All
I am currently designing a vessel in which I have to go for FEA of a nozzle as its failing using WRC 107. Now for avoiding Rigid body motion we apply a pressure thrust in Nozzle flange while doing FEA. But when when we design the same nozzle using PV Elite , We are not including Pressure Thrust in design Calculation of the same. I am confused that shall I include it in FEA or not. In both cases I have Piping Loads Included.

 

[TGS4]

If you do not add the pressure thrust, how will your model have the correct longitudinal stress in the nozzle neck?

 

[XL83NL]

Include it, yes. You shall.

 

[cbPVme]

What do you mean it is not included in the PVElite calculations? There is a checkbox to include the pressure thrust in the WRC 107 input form.

 

[jtseng123]

No, you don't add pressure thrust in both WRC107 or FEA.

Pressure thrust is part of the "external load' on piping routing system that pipe stress engineer must include it when running pipe stress analysis.
This external "pressure thrust" will then distribute through out the piping looping, and depending on how many fixed supports, spring hanger supports and guides, the axial force contributed by the pressure thrust acting at the nozzle to shell junction may only be a small % of the full pressure thrust (PxA). This has been explained in the pipe stress analysis literature.

When you have untied expansion joint, a full pressure thrust will be acting at the nozzle to shell junction. But again, it is pipe stress engineer to analyze and tell you the load at the junction. You don't need to worry about pressure thrust.

By adding pressure thrust, you will unnecessarily overdesign your nozzle.

 

[TGS4]

jtseng123 - for the WRC 107 (superseded by WRC 537) evaluation, I agree that adding an additional external load is not required.

However, regarding the FEA, of one does not add an axial load (typically by means of a negative pressure or traction on the nozzle end), then the longitudinal stress in the nozzle will be incorrect. Therefore, adding such loads is mandatory. Whether you choose to add it as a traction or a single external force that would be additive to the external axial force from the attached piping, it is required. So, in that regards, your advice is incorrect. Unless you have some specific references from the "literature" that you noted above...

 

[jtseng123]

TGS4,
I don't think there is any literature stating you will need to add pressure thrust when running FEA. Pressure thrust will be distributed through the piping system, and reaction force will occur at the anchor points, which will be the junction. This reaction force will be only some % of the full pressure thrust.

What was done using FEA is to have the actual piping loads acting at about 12" away from the junction. That way, the full nozzle stress will be calculated. 12" is about the face of flange in majority of cases. Pipe stress engineer will be able to give the full 6 loads at that location, which shall include the reaction force from pressure thrust.

 

[TGS4]

In each and every analysis that I have performed in the last 20+ years (including the 32+ that have found their way into the published literature), I have added a load (typically in the form of a traction pressure) that I have termed the pressure thrust. So, yes, there is literature stating that. Otherwise, the nozzle neck longitudinal stress is zero (plus the F/A from the external loads), which is fundamentally incorrect.

 

[Mohit_singh]

@TGS4- I admit your comment that nozzle thrust need to be included, but when my piping loads are calculated by stress engineer, then he would already include the pressure thrust in piping loads.
@jtseng123 Thanks for your advise, Will do some more research over that

 

[TGS4]

It is unlikely that the pressure thrust would be included in the piping stress loads. Highly unusual.

 

[GandalfTheGrey]

I second TGS4, the pressure thrust is required to balance the forces, and is a common practice where I work.
One check you can do is don't apply any external forces and check for the membrane stress.

 

[KevinNZ]

Another vote here to add a pressure tension load to the nozzle for FEA.

Unless it is requested, the piping designer does not usually report pressure loads. They assume pressure is included in the vessel side loads.

 

[jtseng123]

Digging out from piles of files, finally found the literature talking about nozzle pressure thrust. Worth reading to understand.

 

[TGS4]

When you read this paper, you will see that the "gold standard" is FEA (NozzlePro), which imposes the full pressure thrust. The remainder of the paper goes on to argue whether or not pressure thrust loads should be additional to the WRC 107/297/368 methods.

Nowhere in this paper does it say to not apply pressure thrust to FEA. jtseng123, you had originally said:

No, you don't add pressure thrust in both WRC107 or FEA.

The paper itself says that the "gold standard" applies pressure thrust to the nozzle. Will you finally admit that you are incorrect in your assertion that application of pressure thrust to FEA is not necessary?

 

[Spoonful]

It depends,
If your nozzle, for example is attached with blind flange, the there will be a full pressure thrust acting on the nozzle away from the shell.
If your nozzle, for example is connected to solid piping with solid anchor. Then there is no pressure thrust on to your nozzle.

 

[DriveMeNuts]

Spoonful,
If early on in a project, the pipe loading design had not been even started and the project needed the equipment to be purchased and manufactured immediately, as the leader vessel engineer responsible for the technical requisition of the equipment, you would be the person who makes assumptions about nozzle loading so that the equipment can be purchased immediately.

For a vessel operating at a single static internal pressure, stresses are typically greatest when the full pressure thrust is applied to the nozzle, therefore it is best to assume that the nozzle will experience full pressure thrust.

As a result a much greater number pipe loading scenarios are covered for when the piping data does arrive later in the project.

You would only consider designing without pressure thrust if you know that a nozzle is going to be "connected to solid piping with solid anchor", if you are aware of this arrangement at the beginning of the project. However, I would still include full pressure thrust, because if the anchor is removed later on from the design or during plant modification then the stress go up and you are stuffed.

Including full pressure thrust provides flexibility/insurance for future changes to the design or physical equipment, usually for little or no financial cost.

 

[TGS4]

All of you (or any of you) who say that the pressure thrust in the nozzle doesn't exist if the attached piping is connected to a "solid anchor", need to show me a drawing of what you mean. Are you actually saying that, in the presence of pressure only, the longitudinal stress in a pipe in that situation is zero? Show my a drawing, with a free body diagram, and a description of the situation.

My understanding of the issue is with respect to the non-FEA methods of WRC 107/537 or WRC 297, or WRC 386 - is it appropriate to supplement the external axial load with an additional axial load equal to the pressure thrust. That argument is a completely different matter, and should not be conflated with the discussion about whether or not to include pressure thrust in an FEA.

 

[PV1108A]

Hi TGS4,
If there is a line stop between the nozzle and first elbow (for example), such as shown in Figure 5 of this webpage (

https://communities.bentley.com/pro...essure-thrust-on-nozzles-in-autopipe-nozzle),

wouldn't this be a case where the pressure thrust would not be included in the FEA for the nozzle?
Thank you,
PV

 

[TGS4]

If, by chance, you had what is shown in Figure 5 of that article, the "thermal" loads would be extremely large, and would quite likely result in a failure. That is, possibly, the most stupid support arrangement. If that's the only situation where there may be a fully-counteracted pressure thrust, then I rest my case.

Nevertheless, the longitudinal stresses in the nozzle/pipe would still equal P*D/(2*t), but it would be counteracted by an opposite force from the line stop. Plus some Poisson effect.

 

[BJI]

Only where there is a line stop with no gaps before the first elbow (like in the example), and where the vessel and vessel anchor (with no gaps) is infinitely stiff, would you see no pressure thrust in the nozzle neck. This one theoretical example doesn't reflect any real applications that I have come across, therefore it wouldn't be acceptable to exclude pressure thrust in FEA for any normal application.