XANSYS Forum Index
    Register    

FAQ    Search    Memberlist    Usergroups    SmartFeedSmartFeed    Profile    Log in
[STRUCT] 2D Axisymmetric Global Model to 3D Submodel
 
This forum is locked: you cannot post, reply to, or edit topics.   This topic is locked: you cannot edit posts or make replies.    XANSYS Forum Index -> XANSYS
Author Message
andrew.kelly
User


Joined: 06 Mar 2012
Posts: 243
Location: Columbus, OH

PostPosted: Thu Jun 19, 2014 2:00 pm  Reply with quote

Kind of curious if I can leverage the solution speed of a 2D axisymmetric assembly, then apply those displacements to a 3D solid submodel. I know this works with 3D shells & beams to a 3D solid submodel, but that doesn't apply to my assembly.

I have an assembly that can be simplified to an axisymmetric model. It solves very fast, but I need to post process things that do not lie in the radial, axial, or circumferential directions, so the axisymmetric model isn't all that helpful.

I am currently using a 3D solid model. I have DM imprints to create the faces I use for post-processing. The solve time is very slow, but the results on the imprinted faces match experimental results.

It would be fantastic if I could isolate a cylinder around these imprinted faces in the 3D model to make a submodel, then apply the displacements from the 2D axisymmetric model to the OD of the cylindrical face in the 3D submodel. That way I could have fast solution times of a 2D assembly and small 3D part, and be able to post process with the 3D solid model.

Is this possible?
_________________
Andrew C. Kelly, P.E.
Honeywell Sensing & Productivity Solutions
andrew.kelly@honeywell.com
Back to top
View user's profile Send private message Visit poster's website
jose.galan
User


Joined: 21 Oct 2008
Posts: 140

PostPosted: Thu Jun 19, 2014 5:57 pm  Reply with quote

Dear Mr. Kelly,

You may consider creating a cycling symmetry model of a small angular
sector of you structure (Advanced Guide | Chapter 7. Cyclic Symmetry
Analysis). This sector is meshed with 3D elements (solid, shells, etc).
The cyclic symmetry model is more computationally demanding than a 2D
axisymmetric model (but still smaller than the full 3d model). On the
other hand, Ansys provides some commands to expand the solution results
of your cyclic symmetry analysis to the full 360º model (or to a certain
integer number of basic sectors). In your case you would only need to
expand it to an angle large enough to contain the local region that you
want to submodel. The appropiate command in your case would be EXPAND.
With this command, the results are expanded before storing them in the
database, and new nodes and elements are created to store the additional
results. The database size increases, but you can postprocess the
results exactly as you would do for a non-cyclic 3d model. Since you
have 3d results in the database, I think that you could apply the
solid-solid submodelling technique of ansys. I have not done it myself,
though.

From the ansys manual:

Advanced Guide | Chapter 7. Cyclic Symmetry Analysis | 7.4.
Postprocessing a Cyclic Symmetry Analysis

"7.4.2.2. Using the EXPAND Command
The EXPAND command offers an alternate method for displaying the results
of a modal cyclic symmetry analysis. It is a specification command that
causes a SET operation to transform and expand the data it is reading
before storing it in the database. If you request two or more sector
repetitions, the command creates additional nodes and elements to
provide space for the extra results.

After the real-space results are stored in the database, you can plot
(PLESOL or PLNSOL), print (PRNSOL) or process them as you would those
for a non-cyclic model, a convenience if you want to process the results
in a manner unsupported by the /CYCEXPAND command. The database can
become very large, however, negating the inherent model-size advantage
of a cyclic symmetry analysis. "

Commands Reference | E Commands | EXPAND
"When you issue the EXPAND,Nrepeat command, subsequent SET commands read
data from the results file and expand them to Nrepeat sectors."

"EXPAND changes the database by modifying the geometry, the nodal
displacements, and element stresses as they are read from the results
file"

"The EXPAND command creates new nodes and elements."

I send you a link to a small example of an axisymmetric problem (thick
tube under internal pressure), solved by three methods: axisymmetric
(2d); solid (90º slice); cyclic symmetry (10º slice).
http://mae.uta.edu/~lawrence/ansys/ansys_examples.htm

Best regards,

Jose M. Galan

Constr. Engin. Dpt.

Univ. Sevilla

Spain

use the command EXPAND to expand the results into a full 3d model, with
the command solving using cyclic symmetry I think that you could do it
using tabular boundary conditions. Your idea You can extract the
displacements of your axisymmetric model at a certain radial position
(assembly you could define extract the values but you you would the

El 19/06/2014 23:00, andrew.kelly escribió:

Quote:
Kind of curious if I can leverage the solution speed of a 2D axisymmetric assembly, then apply those displacements to a 3D solid submodel. I know this works with 3D shells & beams to a 3D solid submodel, but that doesn't apply to my assembly.

I have an assembly that can be simplified to an axisymmetric model. It solves very fast, but I need to post process things that do not lie in the radial, axial, or circumferential directions, so the axisymmetric model isn't all that helpful.

I am currently using a 3D solid model. I have DM imprints to create the faces I use for post-processing. The solve time is very slow, but the results on the imprinted faces match experimental results.

It would be fantastic if I could isolate a cylinder around these imprinted faces in the 3D model to make a submodel, then apply the displacements from the 2D axisymmetric model to the OD of the cylindrical face in the 3D submodel. That way I could have fast solution times of a 2D assembly and small 3D part, and be able to post process with the 3D solid model.

Is this possible?

------------------------
Andrew C. Kelly, P.E.
Honeywell Sensing & Control
andrew.kelly@honeywell.com

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
View user's profile Send private message
dave.conover
User


Joined: 11 Oct 2011
Posts: 16

PostPosted: Fri Jun 20, 2014 6:19 am  Reply with quote

Take the cut boundary of your 3D model, collapse the nodes to the Z=0 plane
(NMODIF in the appropriate cylindrical CSYS), do the 2D cut boundary
interpolation. The resulting displacements can then be applied to the 3D
model.

Dave Conover
ANSYS, Inc.


On Thu, Jun 19, 2014 at 5:00 PM, andrew.kelly <andrew.kelly@honeywell.com>
wrote:

Quote:
Kind of curious if I can leverage the solution speed of a 2D axisymmetric
assembly, then apply those displacements to a 3D solid submodel. I know
this works with 3D shells & beams to a 3D solid submodel, but that doesn't
apply to my assembly.

I have an assembly that can be simplified to an axisymmetric model. It
solves very fast, but I need to post process things that do not lie in the
radial, axial, or circumferential directions, so the axisymmetric model
isn't all that helpful.

I am currently using a 3D solid model. I have DM imprints to create the
faces I use for post-processing. The solve time is very slow, but the
results on the imprinted faces match experimental results.

It would be fantastic if I could isolate a cylinder around these imprinted
faces in the 3D model to make a submodel, then apply the displacements from
the 2D axisymmetric model to the OD of the cylindrical face in the 3D
submodel. That way I could have fast solution times of a 2D assembly and
small 3D part, and be able to post process with the 3D solid model.

Is this possible?

------------------------
Andrew C. Kelly, P.E.
Honeywell Sensing &amp; Control
andrew.kelly@honeywell.com






+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
View user's profile Send private message
andrew.kelly
User


Joined: 06 Mar 2012
Posts: 243
Location: Columbus, OH

PostPosted: Fri Jun 20, 2014 12:43 pm  Reply with quote

jose.galan wrote:
You may consider creating a cycling symmetry model

That might be a good compromise.

I'm sure there are nuances between using Cyclic Symmetry pairs vs. Symmetry Regions in revolved parts (vs. just putting Frictionless Supports on the cut planes). Can anyone help clarify the differences?
_________________
Andrew C. Kelly, P.E.
Honeywell Sensing & Productivity Solutions
andrew.kelly@honeywell.com
Back to top
View user's profile Send private message Visit poster's website
jose.galan
User


Joined: 21 Oct 2008
Posts: 140

PostPosted: Fri Jun 20, 2014 3:31 pm  Reply with quote

They are quite different. The cyclic symmetry b.c. actually writes a set of coupling equations between the d.o.f. of corresponding nodes (i.e.located at the same radial position) at both edges (theta max and theta min) of the basic sector. Using complex notation and a cylindrical coordinate system , this relationship can be written for all the points at both (upper and lower) edges as
u(r,thetamax)=u(r,thetamin)*exp(i*k*incth)
Where incth=thetamax-thetamin
i=sqrt(-1)
k is the index number, that can take integer values from -kmax to kmax, with kmax =360/incth.
The displacements u(r,th) are complex values; they have an amplitude and a phase.
In the simplest case, k=0, the coupling eqn reduces to
u(r,thetamax)=(r,thetamin)
Ansys doesn't use complex notation. Instead, they express disp as sum of a sine and a cos term, but it is the same concept.

On the other hand, When you apply sliding (simple symmetry) boundary conditions at the edges, the displacements at the upper and lower edges are independent from each other, they are not coupled. You may obtain different displacements, in general.
I recommend you to read the cyclic symmetry section on the ansys manual. It will give you some examples where cyclic symm is applucable . For instance, a turbine.
Best regards
Jose m Gala
Univ. Sevilla
Spain
_________________
Jose M. Galan
Assistant Professor
Construction and Project Engineering Dpt.
Universidad de Sevilla
Spain
Back to top
View user's profile Send private message
Display posts from previous:   
This forum is locked: you cannot post, reply to, or edit topics.   This topic is locked: you cannot edit posts or make replies.    XANSYS Forum Index -> XANSYS
All times are GMT - 7 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You cannot download files in this forum


Powered by phpBB © 2001, 2005 phpBB Group

sleek template created by Andrew Charron