Buckling modes shapes

- May 14, 2020 at 10:53 am
  
  AJ998
  Subscriber
  
  Hello everyone,
  
  I am making a buckling analysis, and I was wondering if I can get the function of the resulting mode shapes,
  
  I know the shape of the first buckling mode of a cantilever beam, and the function looks like
  
  M(x) = A*(1-cos((1+1/2)* pi*x/L))
  
  I would like to get that formula for the buckling modes of the structure I am analyzing, can I get that from ANSYS WB, if so how ?
  
  Thanks alot !
- May 15, 2020 at 12:36 am
  
  peteroznewman
  Subscriber
  
  No, the analysis does not come with an equation for the buckled shape.
  
  You can take a line of nodes along the length, output their deformation and use a curve fitting program to obtain the best fit to a number of different equations.
- May 15, 2020 at 4:47 am
  
  AJ998
  Subscriber
  
  But I don't have a 1D shape, so it would be hard to account for the modes entirely doing this method. But how can I You can take a line of nodes along the length, output their deformation?
  Also could I visualise the UPGEOM command?
  Thanks alot !
- May 15, 2020 at 4:47 am
  
  AJ998
  Subscriber
  
  But I don't have a 1D shape, so it would be hard to account for the modes entirely doing this method. But how can I You can take a line of nodes along the length, output their deformation? Also could I visualise the UPGEOM command? Thanks alot !
- May 16, 2020 at 12:24 am
  
  peteroznewman
  Subscriber
  
  Why do you want a formula for the buckled shape? Why is the buckled shape itself insufficient?
- May 17, 2020 at 10:59 am
  
  AJ998
  Subscriber
  
  I would like to use these shapes and sum them up to make a nonlinear post buckling analysis. I tried to do with UPGEOM command, but I cannot visualize it. So I thought if I had the equations I could add the perturbation in the geometry manually.
- May 17, 2020 at 11:00 am
  
  AJ998
  Subscriber
  
  what element are you using, maybe making a 2D analysis would work better that 3d ?
- May 17, 2020 at 2:04 pm
  
  peteroznewman
  Subscriber
  
  AJ988, you can create a perturbation and update the geometry manually or use UPGEOM. Here is a video on nonlinear buckling that is not specific to ANSYS. I particularly like the 2-step perturbation force method, since you don't need to any perturbation in the geometry, and the force is not present at the end of the analysis.
  
  OguzhanA, the website may have been broken when you tried to create a New Discussion, but it is working now.
- May 18, 2020 at 9:17 am
  
  AJ998
  Subscriber
  
  yes, but after i do what you are suggesting, I have the error : the Jacobin is negative, and highly distorted elements, I would like to be able to see the new geomtery and the mesh to be able to fix it. to be more precise i used a similar method to this
  
  https://www.simutechgroup.com/tips-and-tricks/fea-articles/94-fea-tips-tricks-buckling
  
  if i cannot visualize the new geometry in order to fix my mesh. i would to generate myself by combining the buckling modes. but how can i take a line of nodes along the length, output their deformation?
- May 18, 2020 at 10:03 pm
  
  peteroznewman
  Subscriber
  
  AJ998, you will get better suggestions if you provide images so we can see what you are doing.
  
  What I suggested was to apply a very small perturbation load in step 1, which should not create highly distorted elements or negative Jacobians. If you got that, your load is way too high. Use a tiny load that only moves the structure by a small fraction of a wall thickness.
  
  In step 2, you ramp up the main buckling load, while ramping off the small perturbation load. This is when you can get highly distorted elements and negative Jacobians. What is important is to have a large number of substeps and plot the force up to the point when the distorted elements stops the solution. It might not matter that the solution stopped if you got a good plot of the force.