Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Shell structure

Status
Not open for further replies.

SvenHildering

Civil/Environmental
Joined
Jul 15, 2014
Messages
2
Location
NL
Hee guys,

I just started using Ansys to create a shell structure which I will investigate on numerous subjects for my bachelor thesis. I design the structure with .mac files which I create in notepad.

The following code I made for a simple shell which can be bended in two directions and twisted:

! define variables
kxx = 3/2000 ! 1/mm kromming in de x-richting
kyy = -3/2000 ! 1/mm kromming in de y-richting
kxy = 0 ! 1/mm twist
t = 10 ! mm dikte
lx = 1000 ! mm lengte in de x-richting
ly = 1000 ! mm lengte in de y-richting
E = 2.1e11 ! N/mm2 elasticiteitsmodulus
nu = 0.00 ! - dwarscontractiecoefficient
nx = 10 ! - aantal elementen in de x-richting
ny = 10 ! - aantal elementen in de y-richting
rho = 7000 ! kg/m3 dichtheid

/PREP7
MPTEMP,,,,,,,, ! materiaal: isotroop
MPTEMP,1,0
MPDATA,EX,1,,E
MPDATA,PRXY,1,,nu
MPDATA,DENS,1,,rho ! Definieer dichtheid
ET,1,SHELL281 ! element type: 8 node quadrilateral
R,1,t,t,t,t, , , ! element dikte



*DO,i,0,nx !put nodes,
*DO,j,0,2*ny
x=-lx/2+i*lx/nx
y=-ly/2+j*ly/(2*ny)
z1=0.5*kxx*x*x
z2=kxy*x*y
z3=0.5*kyy*y*y
z=z1+z2+z3
N,,x,y,z,,,
*ENDDO
*ENDDO
*DO,i,0,nx-1 !put nodes, mid nodes op x-as,
*DO,j,0,ny
x=-lx/2+lx/(2*nx)+i*lx/nx
y=-ly/2+j*ly/ny
z1=0.5*kxx*x*x
z2=kxy*x*y
z3=0.5*kyy*y*y
z=z1+z2+z3
N,,x,y,z,,,
*ENDDO
*ENDDO

*DO,j,1,ny ! elements
*DO,i,1,nx
k1=1+(i-1)*2+(j-1)*(3*nx+2)
k2=1+i+(2+(j-1)*3)*nx+(j-1)*2
k3=1+(i-1)*2+j*(3*nx+2)
E,k3,k3+2,k1+2,k1,k3+1,k2+1,k1+1,k2
*ENDDO
*ENDDO

*DO,i,1,2*nx+1 ! hinges
D,i ,,0,,,,UX,UY,UZ,,,
D,(2*nx+1)*ny+(nx+1)*ny+i ,,0,,,,UX,UY,UZ,,,
*ENDDO
*DO,j,1,ny
D,(2+(j-1)*3)*nx+2*(j-1)+2 ,,0,,,,UX,UY,UZ,,,
D,(3+(j-1)*3)*nx+2*(j-1)+2 ,,0,,,,UX,UY,UZ,,,
*ENDDO
*DO,j,2,ny
D,1+(j-1)*(3*nx+2) ,,0,,,,UX,UY,UZ,,,
D,1+(j-1)*(3*nx+2)+2*nx ,,0,,,,UX,UY,UZ,,,
*ENDDO

ACEL,0,0,+1, ! add Self-weight
FINISH

/SOLU ! bereken
SOLVE
FINISH

When I run it, I get a visually perfect node mesh, but I also get an error which states the following:

Quadrilateral element 1 has a zero or negative determinant of the Jacobian matrix at one of its sampling locations. Midside ndoes, if any, may be poorly positioned.

Due to this error, element one can not be constructed, and the script stop running.

Can someone of you please help me fix this problem!

 
Nailed it. Forgot to edit the part where the elements are being created. Thanks!
 
You really did this the hard way. Creating an area with the RECT command and meshing it with the AMESH command is much simplier and quicker.

! define variables
kxx = 3/2000 ! 1/mm kromming in de x-richting
kyy = -3/2000 ! 1/mm kromming in de y-richting
kxy = 0 ! 1/mm twist
t = 10 ! mm dikte
lx = 1000 ! mm lengte in de x-richting
ly = 1000 ! mm lengte in de y-richting
E = 2.1e11 ! N/mm2 elasticiteitsmodulus
nu = 0.00 ! - dwarscontractiecoefficient
nx = 10 ! - aantal elementen in de x-richting
ny = 10 ! - aantal elementen in de y-richting
rho = 7000 ! kg/m3 dichtheid

/PREP7
MPTEMP,,,,,,,, ! materiaal: isotroop
MPTEMP,1,0
MPDATA,EX,1,,E
MPDATA,PRXY,1,,nu
MPDATA,DENS,1,,rho ! Definieer dichtheid
ET,1,SHELL281 ! element type: 8 node quadrilateral
R,1,t,t,t,t, , , ! element dikte



rect,-lx/2,lx/2,-ly/2,ly/2
lsel,s,loc,x,0
lesize,all,,,nx

lsel,s,loc,y,0
lesize,all,,,ny

mshkey,1
amesh,all


nsel,s,loc,x,-lx/2
nsel,a,loc,x,lx/2
d,all,ux,0,,,,uy,uz

allsel

ACEL,0,0,+1, ! add Self-weight
FINISH

/SOLU ! bereken
SOLVE
FINISH


It wasn't clear to me what you were trying to do with those boundary conditions, so I guessed. Also, take a look at your units. You have dimensions in mm, density in kg/m3 and acceleration in g's. A consistent set of units is MPA which is mm, tonne/mm3 and mm/s2.



Rick Fischer
Principal Engineer
Argonne National Laboratory
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top