Axial follower force on a PIPE element

[dersman]

Hello all,

I'm doing a large deflection analysis with a string of PIPE20 elements. I want a force to work in the pipe's axial direction at any time during the analysis, hence I need a follower force. I know this requires applying an axial pressure since forces in Ansys always point in the same direction.

My question is if there is any way to apply an axial pressure to a pipe element. Maybe there is some possibility by creating a surface at the pipe's end and applying the pressure to that?

Thanks!

 

[meshparts]

Hello dersman,

as far as I know, forces in Ansys are defined in the nodal coordinate system. I usualy don't work with large deflections, but I suppose in such a case, the nodal coordinate system will also rotate/move. So the forces should be all the time tangential to the pipeline.

Regards,
Alex

 

[dersman]

First of all, thanks for your reaction!

During a large deflection analysis the forces work in the global coordinate system, which does not rotate with the model. I am pretty sure that the only way to get an axial follower force is to apply an axial pressure. I just dont know if that is possible and if so, how to do it...

I hope there is someone out there who has the solution for this problem.

Dersman

 

[Stringmaker]

Dersman,
You question had me curious to whether this could be done or not. After doing a little research on it I found that it is possible to apply pressure loadings not only in the lateral direction (normal) but also in the axial direction of a beam as well which is what you're after. Check out the SFBEAM command. There's also an short description and illustration in Sec. 2.5.7.1 of the Basic Guide in the documentation.

Good luck,
-Brian

 

[dersman]

Thanks for your respons!

I tried the SFBEAM command before but I only get it to work with real beam elements. Not with derived beam elements like the PIPE series. I think it says in the manual that the SFBEAM command should work with PIPE elements as well, but it just doesn't work for me.

Once again, I'm using PIPE20 elements (Ansys 8.1) and I want an axial pressure. According to the manual an axial pressure for this element means LKEY=2.

Any idea why this doesn't work?

Thanks very much!

 

[Stringmaker]

Dersman,
I'm on my way out of the office for a long weekend here or else I would make up a test case to try using it as well. Under the PIPE20 element description it gives the following:

Surface Loads
Pressures --
1-PINT, 2-PX, 3-PY, 4-PZ, 5-POUT

The above seems to indicate that by specifying LKEY=2 you should be obtaining the proper loading. Have you tried both positive and negative pressures to see if that has an effect? I'm stumped as well...I'll delve a little deeper into this after the weekend.

Good luck,
-Brian

 

[Drej]

It seems logical that if you need to apply an axial force, you can do so by defining an internal pressure (available as part of the standard pressure PPRES command). To do this, calculate your axial force as a function of internal pressure as part of the theory of thin cylinders, and multiply through by the cross sectional area (CSA), that is:

Axial (longitudinal) force = (Pd/4t)*CSA

where P = internal pressure, d = pipe OD, t = wall thickness of pipe.

Hope this helps.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[dersman]

Thanks for now, I won´t have much time to work on the problem this weekend, but I will continue this thread on monday for sure. I hope you will keep helping me ´cause I know together we will solve this thing!

Dersman

 

[dersman]

Drej,

I can't see how I can apply an axial force by applying an internal pressure, in case of an open pipe. It would work in case of a closed pipe, but then I would also influence the cirumferencial and radial stress which is unwanted.

Maybe I don't understand exactly what you mean, but there must be a way to apply a pressure on the cross sectional area of the pipe's free end in my opinion...

 

[cbrn]

Hi,
giving a look at the Help for PIPE20 allowed to read: "the transverse pressures PX, PY and PZ [...] are defined in the Global Cartesian coordinate system". Though this seems a bit incoherent (why would one have to "orient" each force direction, when a triad of non-colinear vectors do already form a "base" of the 3D space ???), it might explain why you don't get any correct pressure application: probably the directions are not compatible, i.e. they give a null resultant.

Regards