modeling a composite using 'sections' in shell 181
modeling a composite using 'sections' in shell 181
(OP)
Hi,
I have a thin structure which I model using the shell181 element in Ansys. As long as I provide the thickness of the structure using the 'real' definition (just one layer), the results are fine. When I try to model the same structure as a composite using 2 layered sections, the results are very different (the displacements/strains are very small). There is hardly any displacement as compared to the ones obtained using real definition.
I also try to model just one layer using sections instead of specifying thickness using 'real' command but still the results are not in accordance with what I get using the 'real' option to specify thickness. I am using keyopt3=2 option with sections, but of no help.
Would anyone have insights into this?
I have a thin structure which I model using the shell181 element in Ansys. As long as I provide the thickness of the structure using the 'real' definition (just one layer), the results are fine. When I try to model the same structure as a composite using 2 layered sections, the results are very different (the displacements/strains are very small). There is hardly any displacement as compared to the ones obtained using real definition.
I also try to model just one layer using sections instead of specifying thickness using 'real' command but still the results are not in accordance with what I get using the 'real' option to specify thickness. I am using keyopt3=2 option with sections, but of no help.
Would anyone have insights into this?





RE: modeling a composite using 'sections' in shell 181
Could you post the code you're using to define your keyopts, real constants, and sections here? Specifying a real constant at node i for thickness and a shell thickness in a section should yield identical results. Without a little snippet of your code it's hard for onlookers to help.
-Brian
RE: modeling a composite using 'sections' in shell 181
For the sections option:
et,1,shell181
KEYOPT,1,1,0
KEYOPT,1,3,2
KEYOPT,1,8,0
KEYOPT,1,9,0
KEYOPT,1,10,0
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,EX,1,,7e3
MPDATA,PRXY,1,,0.4
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,DENS,1,,1100e-6
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,EX,3,,.5e6
MPDATA,PRXY,3,,.4
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,DENS,3,,1100e-6
!*
!from bottom to top in +z direction of element coordinate system
!z axis points from bottom to top in this case (for the elements, from face 1 to 2),
sect,1,shell,,
secdata, .5,1,0.0,3
secoffset,MID
seccontrol,3.8e5,3.8e5,3.8e5, , , ,
sect,2,shell,,
secdata, 1.5,3,0.0,3
secoffset,MID
seccontrol,1e5,1e5,1e5, , , ,
!The geometry is specified, which I am omitting
asel,s,,,1,3,2
AATT, , , , ,2
alls
asel,s,,,4
AATT, , , , ,1
alls
For the real option:
et,1,shell181
KEYOPT,1,1,0
KEYOPT,1,3,2
KEYOPT,1,8,0
KEYOPT,1,9,0
KEYOPT,1,10,0
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,EX,1,,7.0e3 !g/mm-s2
MPDATA,PRXY,1,,0.4
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,DENS,1,,1100e-6 !g/mm3
R,1,0.5, , , , , , !thickness in mm
RMORE,3.8e5,3.8e5,3.8e5, , , , !shear modulus
!*
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,EX,2,,0.5e6
MPDATA,PRXY,2,,.4
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,DENS,2,,1100e-6
R,2,1.5, , , , , , !thickness in mm
RMORE,1e5,1e5,1e5, , , ,
!geometry is being omitted
asel,s,,,1,3,2
AATT, 2 ,2 ,1 ,0 ,
alls
asel,s,,,4
AATT, 1, 1 , 1 , 0 ,
alls
Any help would be greatly appreciated.
Thanks
RE: modeling a composite using 'sections' in shell 181
-Brian
RE: modeling a composite using 'sections' in shell 181
Thanks for the help. I appreciate it.
Thanks again
RE: modeling a composite using 'sections' in shell 181
Thanks!
RE: modeling a composite using 'sections' in shell 181
I had posted parts of the code above. The sections and real constants were assigned to the required areas. I had made an error while defining the density in one of the parts (i guess while defining sections). The density units were inconsistent (I have corrected it in the code pasted above), I modified it and it worked.