transform constraint example
Moderator: bernd
Forum rules
and Helpful information for the FEM forum
and Helpful information for the FEM forum
transform constraint example
I downloaded the transform constraint example from here: viewtopic.php?f=18&t=17127#p135715 applied a new material Calculix-Steel and a new Solver calculix. But the transform constraint seam not to work?
bernd
bernd
- Attachments
-
- transform01.fcstd
- (126.24 KiB) Downloaded 69 times
Re: transform constraint example
@Bernd,
you have applied the constraint incorrectly.
please find attached the example file. the link to the calculix manual: http://feacluster.com/CalculiX/ccx_2.10 ... de289.html
you have applied the constraint incorrectly.
please find attached the example file. the link to the calculix manual: http://feacluster.com/CalculiX/ccx_2.10 ... de289.html
Re: transform constraint example
works great. It was wrong in the downloaded file. Propably because the transform object was changed during develpment circle. I missed to set this to cylitrical.
thanks very much !
thanks very much !
Re: transform constraint example
Ok it works with the posted example, but not if I would like to do it completely new. Attached another example.
Re: transform constraint example
@Bernd,
it gives unsatisfactory results when I use the "Fusion001" geometry, when I switch to the "Fusion" geometry is works fine.
it gives unsatisfactory results when I use the "Fusion001" geometry, when I switch to the "Fusion" geometry is works fine.
Re: transform constraint example
I may found the culprit. If the GMSH mesh object is used and the mesh is done inside the active analysis, than analysis group meshing is active which gives blind fast inpurt file writing, but there seams to be some bug with this group meshing and the transform constraint. If the mesh is meshed outside the analysis I get good results in any case for GMSH meshes too. I need to have a look at this. May be I simply make a property to bre able to manually deactivate the analysis group meshing.KgoaOT wrote:@Bernd,
it gives unsatisfactory results when I use the "Fusion001" geometry, when I switch to the "Fusion" geometry is works fine.
rotational_beamBernd.fcstd
Thanks for the help ! cheers bernd
Re: transform constraint example
the pic above shows a max deflection of 1.00352 mm. Let's do some calculations ...
rectangle cross section
width = 10 mm
height = 20 mm
overall length = 200 mm
span length = 180 mm (between center of cylinder holes)
q = 100 N/mm (100 N/mm / 10 mm width = 10 N/mm2 = 10 MPa Pressure on the face)
E = 210000 N/mm2
Deformation:
max f = (5 / 384) x (q x l^4) / (E x I)
I = w x h^3 / 12
max f = (15 / 96) x (q x l^4) / (E x w x h^3)
max f = (15 / 96) x (100 x 180^4) / (210000 x 10 x 20^3)
max f = 0.973 mm deformation in z-direction. The max overall deformation will be slightliy more
rectangle cross section
width = 10 mm
height = 20 mm
overall length = 200 mm
span length = 180 mm (between center of cylinder holes)
q = 100 N/mm (100 N/mm / 10 mm width = 10 N/mm2 = 10 MPa Pressure on the face)
E = 210000 N/mm2
Deformation:
max f = (5 / 384) x (q x l^4) / (E x I)
I = w x h^3 / 12
max f = (15 / 96) x (q x l^4) / (E x w x h^3)
max f = (15 / 96) x (100 x 180^4) / (210000 x 10 x 20^3)
max f = 0.973 mm deformation in z-direction. The max overall deformation will be slightliy more
Re: transform constraint example
Attache the model updated. Furthermore the fusion was refined.
Re: transform constraint example
Stress:
max sigma_xx = M / W
W = w x h^2 / 6
M = q x l^2 / 8
max sigma_xx = (3 x q x l^2) / (4 x w x h^2)
max sigma_xx = (3 x 100 x 180^2) / (4 x 10 x 20^2)
max sigma_xx = 607.5 N/mm2 (Analysis gives 600 N/mm2 on bottom and 603 on top which is around 99 % )
max sigma_xx = M / W
W = w x h^2 / 6
M = q x l^2 / 8
max sigma_xx = (3 x q x l^2) / (4 x w x h^2)
max sigma_xx = (3 x 100 x 180^2) / (4 x 10 x 20^2)
max sigma_xx = 607.5 N/mm2 (Analysis gives 600 N/mm2 on bottom and 603 on top which is around 99 % )
Re: transform constraint example
Great!
Did you use Elasticity equation to solve this with Elmer?
Cause I got some very wrong results when I tried to do it.