Ansys outputs obviously wrong result on a simple static mechanical problem, why?
 

Hi,

I am a new user of ansys. I was stuck at a problem which I've no idea what I did wrong.

The problem is described detailed as below.

First, I use solidwork to build a simple 3D beam structure.

Structure thickness is 25um.
When i build the structure using solidworks, I choose unit system MMGS(milimeter, gram, second), and saved the structure as parasolid(.x_t) file.

Then I imported the structure into Ansys 15.0(also tried Ansys 16.0), did the following steps:
1. define element ( solid 20node 186)
2. Import material properties. (material is silicon)
3. mesh(using smart mesh, set the level to 1)
4. Set analysis type(static)
5. define boundary conditions(set the four supporting beam ends fixed in all DOF)
6. add a force to the central shuttle, negative Z direction, value 10uN.

7. solve LS

The result I got is like this(contour plot-nodal solution-Z component of displacement)

The result is contradictory to common sense, how could the upper supporting beams bent toward the opposite direction of the force? How could the lower part of shuttle has a negative displacement while beam a positive direction pulling force?

I wish somebody could point out my mistakes.

Thanks in advance.

Hi Li,

Your workflow seems correct, but there must definitively be a mistake somewhere. Did you check your log file?
I’ve quickly recreated your model with no particular issue. If nothing you couldn’t find something in the log file or input file, it’s better to upload them so people can have a look.

Hi Gweher,

Thanks very much for your help.

I am a new user of Ansys, right now I mainly use the GUI to perform a simulation, don't have much knowledge about ansys code.
Below is the log file.


/BATCH
/COM,ANSYS RELEASE Release 16.1 BUILD 16.1 UP20150325 09:39:03
/input,menust,tmp,''
/GRA,POWER
/GST,ON
/PLO,INFO,3
/GRO,CURL,ON
/CPLANE,1
/REPLOT,RESIZE
WPSTYLE,,,,,,,,0
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
~PARAIN,'Part2','x_t','.\Desktop\',SOLIDS,0,0
/NOPR
/GO
/FACET,NORML
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/DIST,1,0.924021086472,1
/REP,FAST
/AUTO,1
/REP,FAST
!*
ET,1,SOLID186
!*
MPREAD,'Silicon_original','mp','Desktop\'
SMRT,6
SMRT,OFF
SMRT,6
SMRT,7
SMRT,1
MSHAPE,1,3D
MSHKEY,0
!*
CM,_Y,VOLU
VSEL, , , , 1
CM,_Y1,VOLU
CHKMSH,'VOLU'
CMSEL,S,_Y
!*
VMESH,_Y1
!*
CMDELE,_Y
CMDELE,_Y1
CMDELE,_Y2
!*
FINISH
/SOL
FLST,2,4,5,ORDE,4
FITEM,2,3
FITEM,2,9
FITEM,2,13
FITEM,2,19
!*
/GO
DA,P51X,ALL,0
/VIEW,1,,1
/ANG,1
/REP,FAST
/VIEW,1,,,-1
/ANG,1
/REP,FAST
/AUTO,1
/REP,FAST
/VIEW,1,1,1,1
/ANG,1
/REP,FAST
/USER, 1
/VIEW, 1, 0.671256814260 , 0.397635938763 , -0.625539726567
/ANG, 1, 45.2186943371
/REPLO
/ZOOM,1,SCRN,0.892146,0.262017,1.027138,0.109644
FLST,2,1,1,ORDE,1
FITEM,2,1340
!*
/GO
F,P51X,FZ,-10e-6
/AUTO,1
/REP,FAST
/VIEW,1,1,1,1
/ANG,1
/REP,FAST
/STATUS,SOLU
SOLVE
FINISH
/POST1
!*
/EFACET,1
PLNSOL, U,Z, 0,1.0
/VIEW,1,,1
/ANG,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST
/DIST,1,1.08222638492,1
/REP,FAST

Hi Gweher,

Did you apply the same boundary condition as I did? If yes, would you please upload your model file, I would like to try on my computer to see if it is the problem of model?

Thanks,
Chengjun Li

Hi Li,

Well one possible cause could be the definition of your material properties (through the "MPREAD,'Silicon_original','mp','Desktop\' ") Did you try with standard mat properties?



I can share my input file if needed.

Hi Gweher,

Thanks a lot!

Just like what you said, it is the problem of material properties.

Other than density, young's modulus, possion ratio, the material property file I imported has specific heat, thermal expansion, thermal conductivity, resistivity as well.

I deleted those material properties(specific heat, thermal expansion, thermal conductivity, resistivity), and run the simulation again. This time I could get the correct result.

Do you know why including those material properties will lead to wrong result?

Is it because the system will not do a static structural analysis by including those material properties?

Thanks.

Thanks for the feedback Li. As mentioned before, for me it was one of the “only” origin of your problem.
It is hard to say without knowing how you defined those additional material properties. The solution you have now is identical to mine as well and it is a static structural analysis. If properly defined the additional material properties will not affect your test case.