AnsweredAssumed Answered

Convergence of results (static stress analysis)

Question asked by Darshan Angadi on Mar 15, 2014
Latest reply on Jul 17, 2017 by Theo Linders

Like • Show 0 Likes0
Comment • 6

Hi,

I am a new learner of SolidWorks and like the immense capabilities it offers. I am performing static stress analysis of an assembly. It consists of 3 simple solid cylinders and two hollow cylinders. The area of interest for me is the stress at the junction of the top cylinder with the inner cylinder in the assembly.

I have performed the basic steps of simulation as per the online tutorial. However despite numerous efforts (mesh control, applying fillets, trial of automated h and p-adaptive methods, review of contacts/restraints) I am unable to get some sensible convergence of the max von Mises stress at the area of interest. I wondered if I any experienced member in the SolidWorks community can guide me on how I can improve my simulation of this assembly to obtain convergence.

I remember having read that with simulation it is as much art as it is skill - I am trying to pick up both at the moment!

With many thanks

DSA

Attachments

solidworksforumsmodels.zip6.2 MB

No one else has this question

Outcomes

Jared Conway Mar 16, 2014 1:05 PM

Are the displacements converged?

If not, deal with that first.

From there, ask yourself if convergence is really possible. At sharp corners there will be singularities that will not converge.

Start there and then post some screenshots of your model and where you are seeing problems.

Actions

Darshan Angadi @ Jared Conway on Mar 17, 2014 9:36 AM

Hi Jared,

Thank you for the prompt response. Following your suggestion I reviewed the results for both the displacement and von Mises stress. There appears to be some convergence in the displacement values but the same cannot be convincingly said for the von Mises stress values in the area of interest.

I have attached the screenshots from the simulation results. Please have a look when you get a chance. I am grateful for your guidance and suggestions.

Kind regards

DSA

Attachments

Screenshots.pdf761.2 KB

Actions

Jared Conway @ Darshan Angadi on Mar 18, 2014 12:58 AM

looks like a singularity to me. unless you add a nice fillet in there, you're going to have to take the stress right in that area with a grain of salt. lots of great discussions about this on the forum and standard to any fea solution.

Actions

Theo Linders @ Jared Conway on Jul 17, 2017 8:57 AM

hi Jared,

Maybe a stupid question but when we have a singularity like that and i want to know the maximum stress in that assembly, how can i find it?

Actions

Mikael Martinsson Mar 17, 2014 3:05 PM

For some reason the assembly expects parts different from those that you included in the zip file so I can't open your files.

But as Jared pointed out, if displacement converges while the stress continues to increase with finer mesh, it's likely that you have a stress singularity. They often occurs close to fixtures (infinite stiffness) or at sharp inner corners where you'd expect a radius on the real part. With singularities, stress will continue to rise towards infinity when you refine the mesh.

What you can do?

Either change the geometry or the fixture to better reflect the "real world part" or see if the stress converges in the vicinity of the singularity and if so, use this stress value to draw your conclusion.

Good luck.

Actions

Darshan Angadi @ Mikael Martinsson on Mar 17, 2014 3:44 PM

Hi Mikael,

Thanks for the suggestions. I have uploaded the assembly files again. Hopefully it should open this time. Please have a look and let me know your thoughts.

Regards

DSA

Attachments

Solidworks forum models.zip6.2 MB

Actions

6 Replies

Jared Conway Mar 16, 2014 1:05 PM
Mark Correct
Correct Answer
Are the displacements converged?
If not, deal with that first.

From there, ask yourself if convergence is really possible. At sharp corners there will be singularities that will not converge.
Start there and then post some screenshots of your model and where you are seeing problems.
Like • Show 0 Likes0
Actions
- Darshan Angadi @ Jared Conway on Mar 17, 2014 9:36 AM
  Mark Correct
  Correct Answer
  Hi Jared,
  
  Thank you for the prompt response. Following your suggestion I reviewed the results for both the displacement and von Mises stress. There appears to be some convergence in the displacement values but the same cannot be convincingly said for the von Mises stress values in the area of interest.
  I have attached the screenshots from the simulation results. Please have a look when you get a chance. I am grateful for your guidance and suggestions.
  
  Kind regards
  
  DSA
  Attachments
  - Screenshots.pdf761.2 KB
  Like • Show 0 Likes0
  Actions
  - Jared Conway @ Darshan Angadi on Mar 18, 2014 12:58 AM
    Mark Correct
    Correct Answer
    looks like a singularity to me. unless you add a nice fillet in there, you're going to have to take the stress right in that area with a grain of salt. lots of great discussions about this on the forum and standard to any fea solution.
    Like • Show 0 Likes0
    Actions
    - Theo Linders @ Jared Conway on Jul 17, 2017 8:57 AM
      Mark Correct
      Correct Answer
      hi Jared,
      Maybe a stupid question but when we have a singularity like that and i want to know the maximum stress in that assembly, how can i find it?
      Like • Show 0 Likes0
      Actions
Mikael Martinsson Mar 17, 2014 3:05 PM
Mark Correct
Correct Answer
For some reason the assembly expects parts different from those that you included in the zip file so I can't open your files.

But as Jared pointed out, if displacement converges while the stress continues to increase with finer mesh, it's likely that you have a stress singularity. They often occurs close to fixtures (infinite stiffness) or at sharp inner corners where you'd expect a radius on the real part. With singularities, stress will continue to rise towards infinity when you refine the mesh.

What you can do?
Either change the geometry or the fixture to better reflect the "real world part" or see if the stress converges in the vicinity of the singularity and if so, use this stress value to draw your conclusion.

Good luck.
Like • Show 1 Like1
Actions
- Darshan Angadi @ Mikael Martinsson on Mar 17, 2014 3:44 PM
  Mark Correct
  Correct Answer
  Hi Mikael,
  
  Thanks for the suggestions. I have uploaded the assembly files again. Hopefully it should open this time. Please have a look and let me know your thoughts.
  
  Regards
  
  DSA
  Attachments
  - Solidworks forum models.zip6.2 MB
  Like • Show 0 Likes0
  Actions

Convergence of results (static stress analysis)

Attachments

Outcomes

Attachments

Attachments

Related Content

Recommended Content