Select nodes on a plane which is part of another plane

[vidiii]

I'm trying to select certain nodes on a plane, which is a part of another plane.

BATCH:

NSEL,S,LOC,Z,h_b/2+h_p ! Selects the desired plane [WORKS OK]
D,all,VOLT,0

NSEL,S,LOC,Z,-h_b/2-h_p ! Selects the desired plane [WORKS OK]
D,all,VOLT,0

NSEL,S,LOC,Z,h_b/2 ! Selects the desired plane, but only a part of the nodes on the plane must have the constraint [WORKS OK]
NSEL,U,LOC,X,0,L1 ! Unselect the nodes that are selected from 0 to L1 on the X-axis [NOT OK]
NSEL,U,LOC,X,L2,L3 ! Unselect the nodes that are selected from L1 to L3 on the X-axis [NOT OK]
D,all,VOLT,0

NSEL,S,LOC,Z,-h_b/2 ! Selects the desired plane, but only a part of the nodes on the plane must have the constraint [WORKS OK]
NSEL,U,LOC,X,0,L1 ! Unselect the nodes that are selected from 0 to L1 on the X-axis [NOT OK]
NSEL,U,LOC,X,L2,L3 ! Unselect the nodes that are selected from L1 to L3 on the X-axis [NOT OK]
D,all,VOLT,0

allsel,all

The first part works OK, but the second part does not. The idea is to select the nodes that are on the h_b/2 plane (XZ plane) and then unselect all the nodes from 0 to L1 and L2 to L3 for the X-axis (0<=L1<L2<=L3).
This doesnt seem to work properly, the wrong nodes get selected and get constrained. Does anybody have an idea what I'm doing wrong?

Hope to hear from you. Thanks in advance.

With the kindest regards,
Vidi

 

[meshparts]

Hello,

All seems to be OK. Check if the coordinate system is set as you wish.

 

[vidiii]

I'll illustrate the problem with an example batch script. The unselecting part doesnt seem to unselect the needed nodes. The wanted nodes (from L1 to L2) are not selected.

Thanks for your help.

Batch

/COM

/PREP7
SMRT,OFF

ET,2,SOLID95

MP,EX,2,22E10
MP,DENS,2,7500

L1 = 0e-3
L2 = 15e-3
LS1 = 22e-3
LS2 = 11E-3
LS3 = 16E-3
h_b = 0.09e-3
h_p = 0.21e-3
b = 10e-3
b2 = 30e-3

K,1,0,-b/2,-h_b/2
K,2,LS1,-b/2,-h_b/2
K,3,LS1+LS2,-(b2-b)/2,-h_b/2
K,4,LS1+LS2+LS3,-(b2-b)/2,-h_b/2
K,5,LS1+LS2+LS3,(b2-b)/2,-h_b/2
K,6,LS1+LS2,(b2-b)/2,-h_b/2
K,7,LS1,b/2,-h_b/2
K,8,0,b/2,-h_b/2

a,1,2,3,4,5,6,7,8

VOFFST,1,h_b, ,

K,17,L1,-b/2,h_b/2
K,18,L1,-b/2,h_b/2+h_p
K,19,L1,b/2,h_b/2+h_p
K,20,L1,b/2,h_b/2

K,21,L2,-b/2,h_b/2
K,22,L2,-b/2,h_b/2+h_p
K,23,L2,b/2,h_b/2+h_p
K,24,L2,b/2,h_b/2

K,25,L1,-b/2,-(h_b/2)
K,26,L1,-b/2,-(h_b/2+h_p)
K,27,L1,b/2,-(h_b/2+h_p)
K,28,L1,b/2,-(h_b/2)

K,29,L2,-b/2,-(h_b/2)
K,30,L2,-b/2,-(h_b/2+h_p)
K,31,L2,b/2,-(h_b/2+h_p)
K,32,L2,b/2,-(h_b/2)

a,17,18,19,20
a,21,22,23,24

a,25,26,27,28
a,29,30,31,32

v,17,18,19,20,21,22,23,24
v,25,26,27,28,29,30,31,32
vglue,all

ESIZE,2e-3,0,

TYPE, 2
MAT, 2
REAL,
ESYS, 0
SECNUM,
!*
MSHAPE,1,3D
MSHKEY,0
!*
FLST,5,3,6,ORDE,2
FITEM,5,4
FITEM,5,-6
CM,_Y,VOLU
VSEL, , , ,P51X
CM,_Y1,VOLU
CHKMSH,'VOLU'
CMSEL,S,_Y
!*
VMESH,_Y1
!*
CMDELE,_Y
CMDELE,_Y1
CMDELE,_Y2
!*


csys,0

NSEL,S,LOC,X,0
D,all,UX,
NSEL,S,LOC,X,0
D,all,UY,
NSEL,S,LOC,X,0
D,all,UZ,


NSEL,S,LOC,Z,h_b/2+h_p ! ok
D,all,UX,0
NSEL,S,LOC,Z,-h_b/2-h_p ! ok
D,all,UX,0

!***PROBLEM***
NSEL,S,LOC,Z,h_b/2 ! not ok
NSEL,U,LOC,X,0,L1
NSEL,U,LOC,X,L2,L3
D,all,UX,0
NSEL,S,LOC,Z,-h_b/2 ! not ok
NSEL,U,LOC,X,0,L1
NSEL,U,LOC,X,L2,L3
D,all,UX,0
allsel,all
!***PROBLEM***

FINISH
/SOL
!*
ANTYPE,2
!*
MSAVE,0
!*
MODOPT,LANB,10
EQSLV,SPAR
MXPAND,0, , ,0
LUMPM,0
PSTRES,0
!*
MODOPT,LANB,10,5,100000, ,OFF
/STATUS,SOLU
SOLVE
FINISH

 

[SKJoe]

Hi vidiii,

If you look on selection syntax, two values must be imput min and max. So when I want to select some nodes in some location on line I do that this way :
nsel,s,loc,x, L1-q, L1+q, where q is a very small value. Since my one very terrible experience I have not been using (...I hope this is correct tense) only
nsel,s,loc,x,L1. Similarly, I would use
NSEL,U,LOC,X,0-q,L1+q, or only NSEL,U,LOC,X,-q,L1+q instead
NSEL,U,LOC,X,0,L1.
Maybe it sounds weird, but you can try that...

Regards,
Lubo

 

[cbrn]

Hi,
or, could you possibly reverse the problem and "reselect" the nodes you want instead of "unselected" the undesired ones?

Regards

 

[meshparts]

Hello,

allso try changing the selection tolerance with SELTOL.

Regards,
Alex