XANSYS Message: 38540 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Rate item:

Subject: Re: Modal cyclic symmetry ...something for the Wish List ?
Author: Juha Tyllinen
Date: 2002-06-06 06:26:00

Isn't there already a command for getting the exact mesh? MSHCopy or
something similar. Granted, it requires two steps.

JKT

> -----Original Message-----
> From: May, Gordon [mailto:Gordon.May@r...]
> Sent: 06. kes„kuuta 2002 12:40
> To: 'xansys@yahoogroups.com'
> Subject: RE: [xansys] Modal cyclic symmetry ...something for the Wish
> List ?

> Seems like VSWEEP should be modified (to include an option) so as to
> replicate the behaviour of AGEN when matching mesh patterns
> are required for
> cyclic symmetry...?

> Gordon May
> Principal Technologist - Stress Methods
> Compressor Key System
> Tel.: +44 1332 2 44041
> Fax: +44 1332 2 44874
> mailto:gordon.may@r...

> "If you speak to him of a machine for peeling a potato, he
> will pronounce it
> impossible: if you peel a potato with it before his eyes, he
> will declare it
> useless, because it will not slice a pineapple."
> Charles Babbage

> -----Original Message-----
> From: Prentice, Ian (GEAE) [mailto:Ian.Prentice@a...]
> Sent: 05 June 2002 16:04
> To: 'xansys@yahoogroups.com'
> Subject: RE: [xansys] Modal cyclic symmetry

> Like I said, the model must be in csys,1 (at least for v5.6).
> If a command
> is
> undocumented, you should ask yourself why.
> Even in csys,1 your model has some nodes which do not match -
> if you use
> vsweep
> you do not always get a matching mesh on the cut sides. To
> ensure a matching
> mesh use agen to overlay a copy of the original mesh on the
> opposite side,
> then
> merge the two areas together before vsweeping the volume.
> I noticed you are also doing 0 thru 4 nodal diameters- 0 can
> be done with
> symmetric bc's and 4 (in your case) with anti-symmetric,
> saving considerable
> computer time and filespace.
> Ian
> GEAE

> -----Original Message-----
> From: esa.prasad@p... [mailto:esa.prasad@p...]
> Sent: Wednesday, June 05, 2002 9:39 AM
> To: xansys@yahoogroups.com
> Subject: RE: [xansys] Modal cyclic symmetry

> HI,

> Here i am giving the macro which is down loaded from the
> web ofcourse
> from Ansys!! I thought atleast it should work COZ it is from
> ansys . But it
> is also not working and giving the same error.

> like

> No node matches Low node ##

> Can any one through some idea????????

> /clear
> !
> ! enter the preprocessor
> !
> /prep7
> !
> ! generate some keypoints
> k,1,10,25
> k,2,11,25
> k,3,13.45,27.45
> k,4,13.45,28.45
> k,5,13.45,36.45
> k,6,18.45,28.45
> k,7,18.45,27.45
> k,8,13.45,35.45
> k,9,18.45,36.45
> k,10,20.45,35.45
> k,11,20.45,38.45
> k,12,19.95,38.45
> k,13,18.95,36.45
> k,14,19.95,37.45
> k,15,10,12.5
> k,16,11,12.5
> k,17,0,35.45
> k,18,0,36.45
> k,19,18.45,35.45
> k,20,12.8,28.45,-9.3
> k,21,14.0,28.45,-8.9
> k,22,15.3,28.45,-6.5
> k,23,16.0,28.45,-3.8
> k,24,15.2,28.45,-2.8
> k,25,14.2,28.45,-3.7
> k,26,13.5,28.45,-5.9
> k,27,12.5,28.45,-7.85
> !
> ! make some straight lines
> l,19,10
> l,10,11
> l,11,12
> l,12,14
> l,13,9
> !
> ! make a few arcs
> larc,2,3,7,2.45
> larc,1,4,6,3.45
> larc,13,14,12,1
> !
> ! generate some areas
> a,15,16,2,1
> a,2,3,4,1
> a,7,6,4,3
> a,17,8,5,18
> a,5,8,19,9
> a,19,10,11,12,14,13,9
> !
> ! make a segmented spline through some keypoints
> flst,2,9,3
> fitem,2,20
> fitem,2,27
> fitem,2,26
> fitem,2,25
> fitem,2,24
> fitem,2,23
> fitem,2,22
> fitem,2,21
> fitem,2,20
> spline,p51x
> !
> ! make an area from the segmented spline
> flst,2,8,4
> fitem,2,24
> fitem,2,25
> fitem,2,26
> fitem,2,27
> fitem,2,28
> fitem,2,29
> fitem,2,30
> fitem,2,31
> al,p51x
> !
> ! extrude this area vertically
> vext,7,,,,7
> !
> ! make a local cylindrical coordinate system
> local,11,1,0,0,0,0,-90
> !
> ! copy a keypoint 45 degrees from keypoint 10
> kgen,2,10,,,,45
> !
> ! generate an arc
> l,10,36
> !
> ! drag some areas to generate some volumes
> vdrag,1,2,3,5,6,,48
> !
> ! generate a volume from keypoints
> v,17,8,46,18,5,45
> !
> ! glue all of the volumes together
> vglue,all
> !
> ! move the working plane to be aligned with 3 keypoints
> kwplan,-1,27,25,29
> kwpave,27,31
> !
> ! offset the working plane a bit in the Z direction
> wpof,,,1
> !
> ! generate a cylinder
> cyl4,,,1.3,,,,-4
> !
> ! subtract the cylinder from volume 4 to make a hole through it
> vsbv,1,4
> !
> ! set up the meshing controls on some of the lines
> lesize,3,,,3
> lesize,6,,,6
> lesize,7,,,6
> lesize,8,,,6
> lesize,11,,,3
> lesize,13,,,3
> lesize,14,,,3
> lesize,15,,,8
> lesize,16,,,8
> lesize,18,,,2
> lesize,21,,,10
> lesize,23,,,10
> lesize,22,,,3
> lesize,25,,,4
> lesize,26,,,4
> lesize,53,,,7
> lesize,54,,,3
> lesize,55,,,7
> lesize,57,,,6
> lesize,58,,,13
> lesize,59,,,3
> lesize,60,,,13
> lesize,61,,,6
> lesize,68,,,15
> lesize,69,,,15
> lesize,71,,,15
> !
> ! set up some element types
> !
> ! first, the linear bricks
> et,1,45
> ! then the quadratic bricks
> et,2,95
> ! followed by quadratic tets
> et,3,92
> !
> ! and now the dummy mesh elements
> ! the first one will be 6 node triangles
> et,4,200,5
> ! followed by 4 node quad
> et,5,200,6
> ! followed by 8 node quad
> et,6,200,7
> !
> ! start by meshing some of the areas
> !
> ! change to element type 4, 6 node triangle MESH200
> type,4
> ! set mesh key to map mesh
> mshkey,1
> ! mesh area 3
> amesh,3
> ! change to element type 5, 4 node quad MESH200
> type,5
> ! set mesh key to free mesh
> mshkey,0
> ! change the transitioning characteristics of the free
> ! mesh to be more gradual
> mopt,trans,1.5
> ! mesh area 6
> amesh,6
> ! change to element type 6, 8 node quad MESH200
> type,6
> ! mesh area 5
> amesh,5
> ! select the areas along the cut boundary that are meshed
> asel,s,area,,3
> asel,a,area,,5,6
> ! change to cylindrical coordinate system
> csys,11
> ! copy these areas about Z axis 45 degrees
> agen,2,all,,,,45
> ! switch back to global cartesian coordinate system
> csys
> ! select everything
> allse
> ! merge coincident keypoints
> numm,kpoi,.001,.001
> ! change to element type 5, 4 node quad MESH200
> type,5
> ! mesh area 44
> amesh,44
> !
> ! mesh all of the volumes
> !
> ! change to element type 1, SOLID45
> type,1
> ! sweep mesh volume 6
> vsweep,6
> ! sweep mesh volume 7
> vsweep,7
> ! change to mapped meshing
> mshkey,1
> ! map mesh volume 2
> vmesh,2
> !
> ! change to element type 2, SOLID95
> type,2
> ! map mesh volume 3
> vmesh,3
> !
> ! change to element type 3, SOLID92
> type,3
> ! change to free meshing
> mshkey,0
> ! free mesh volume 5
> vmesh,5
> !
> ! now map the transition elements between the solid
> ! bricks and the solid tetrahedral elements
> !
> ! change to element type 2, SOLID95
> type,2
> ! change to free meshing
> mshkey,0
> ! tell ANSYS to mesh SOLID95 in tetrahedral form
> mshape,1,3d
> ! mesh volumes 8 and 9
> vmesh,8,9
> !
> ! change the SOLID95 tetrahedral elements to SOLID92
> tchg,95,92,3
> !
> ! all done generating and meshing the volume
> !
> ! turn type numbers on
> /pnum,type,1
> ! turn numbering off so that only colors are shown
> /num,1
> ! turn working plane symbol off
> wpstyle,,,,,,,,0
> ! set up the view so we can see what we have made
> /view,1,-.226,.36,.9
> ! plot the elements
> eplo
> !
> !

> !second macro for cyclic symmetry
> !

> ! enter the preprocessor
> /prep7
> !
> allse
> ! generate some material properties
> mp,ex,1,30e6
> mp,nuxy,1,.3
> mp,dens,1,.00077
> !
> ! make components of the low and high edge pairs
> ! low set first (low in csys5 y direction)
> allse
> asel,s,area,,1,6
> nsla,s,1
> csys,5
> nrot,all
> cm,nodeset1,node
> !
> ! high set second
> asel,s,area,,10,11
> asel,a,area,,14,21,7
> asel,a,area,,25,43,18
> nsla,s,1
> nrot,all
> cm,nodeset2,node
> allse
> ! generate the cyclic sector with loads
> cycgen
> !
> ! constrain the model on the upper rim
> csys,5
> nsel,s,ext
> nsel,r,loc,x,20.45
> d,all,all
> allse
> !
> !
> ! enter the solver to do the analysis
> /solu
> ! define this as a modal analysis problem
> antype,2
> eqslv,spar
> mxpand,20
> modopt,lanb,20,,,,off,,2
> !
> allse
> ! Use this macro as follows:
> !
> !
> CYCSOL,ndmin,ndmax,nSector,LOW,[HIGH],tol,kmove,kpairs,sysnum
> !
> ! where: ndmin - lowest nodal diameter in range
> ! ndmax - highest nodal diameter in range
> ! nSector - number of sectors in full circle
> ! LOW - the LOW node component name (or
> MULTICYC)
> ! HIGH - the HIGH node component
> name (if any)
> ! tol - Tolerance for CECYC command
> ! kmove - If 1, move nodes to be coincident
> ! kpairs - If 1, print pairs of matching nodes
> ! sysnum - Cylinderical coordinate system for
> ! cyclic symmetry (1 by default)
> ! declare this to be a cyclic modal symmetry problem
> cycsol,0,4,8,'nodeset1','nodeset2',.001,,,11
> !
> ! enter the postprocessor
> !/post1
> ! tell ANSYS to expand the graphical image
> !expand,8,modal,1,5,45
> ! or depending on the nodal diameter you can use /expand
> ! /expand,nrepeat,MODAL,ndiam,icsys,sectorangle,,phase
> !/expand,8,modal,1,5,45
> !set,first
> !plns,u,sum
> !

> regards
> Narasimha Prasad Esa
> Alstom Power Technology Center
> United Kingdom

> "Prentice, Ian (GEAE)" on 06/05/2002
> 11:55:03 AM

> To: "'xansys@yahoogroups.com'"
> cc:

> Subject: RE: [xansys] Modal cyclic symmetry

> The model must be in CSYS,1
> Ian
> GEAE

> -----Original Message-----
> From: esa.prasad@p... [mailto:esa.prasad@p...]
> Sent: Wednesday, June 05, 2002 6:41 AM
> To: xansys@yahoogroups.com
> Subject: [xansys] Modal cyclic symmetry

> Hi,

> I am doing cyclic symmetry - modal analysis on nozzle. I have
> created High
> and Low end node components ( these nodes are in say 11
> coordinate system
> which is cylindrical) and also i checked the matching nodes
> coordinates (
> that is cyclic sector angle between corresponding nodes on
> two sides) , the
> cordinates matches.

> now i am trying to use cycsol command ( ansys5.6) it is giving the
> following error.

> No node matches Low node ##

> it is listing error for all nodes in Low and High componets,
> that means it
> is not applying cyclic symmetry ( or CE s).

> Can any one have any idea why it is happening so, what must
> be going wrong
> in this case?

> is there any undocumented field in CYCSOL command????????

> i am using cycsol,0,0,48,'low','high',0.1,0,0,11

> i am having doubt about last field which specify coordinate
> system! which
> is undocumented??

> thanks to all

> Narasimha Prasad Esa
> Alstom Power Technology Center
> United Kingdom.

Posts possibly associated with message #38540AuthorDateScore
38539Re: Modal cyclic symmetry ...something for the Wish List ?Gordon May2002/06/06 
38540Re: Modal cyclic symmetry ...something for the Wish List ?Juha Tyllinen2002/06/06 
38541Re: Modal cyclic symmetry ...something for the Wish List ?Gordon May2002/06/06 
38542Re: Modal cyclic symmetry ...something for the Wish List ?Ian Prentice2002/06/06 
38546Re: Modal cyclic symmetry ...something for the Wish List ?Juha Tyllinen2002/06/06 
38552Re: Modal cyclic symmetry ...something for the Wish List ?Dan Bohlen2002/06/06 
38558Re: Modal cyclic symmetry ...something for the Wish List ?Gordon May2002/06/06