Non-linear simulation with contact

[Anonymous]

Dear Nastran In-CAD users,

I am looking to perform a simulation of non-linear contact between a ring and a cone.

Below is a diagram of the initial position as well as the final position.

I have tested many solutions but I can't find a setting that allows me to get a correct contact between these two parts. The ring is made of ABS and the cone is made of aluminum.

I’m using the last version of Nastran In-CAD as well as the last version of Inventor.

My settings are the following:

• I used half model to simplify (I tried with a quarter of a model but I have problems with the mesh...)

• No initial contact between the two geometries
• Contact with the following settings: Separation / Unsymmetric

Slave Entities:  

Master entities:

• Loads: enforced motion of 8.5mm in Y direction

• Constraints:
  • Symmetry plane ZY

• Constraint in the Y direction for the enforced motion

• Blocking of all directions and rotation on the rear face of the cone except Y translation

• Locking of the translation of the ring in Y direction on just one surface

• Non-linear settings:

• Settings of the analysis:

Questions:

1. Does anyone see anything wrong with the settings?
2. I don’t really know what to do with the ring. What boundary conditions would you apply to the ring?
3. Is this simulation possible with Nastran In-CAD?

The assembly file is available in appendix

Thank you in advance for your help,

Best regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Anonymous]

Dear Nastran In-CAD users,

I am looking to perform a simulation of non-linear contact between a ring and a cone.

Below is a diagram of the initial position as well as the final position.

I have tested many solutions but I can't find a setting that allows me to get a correct contact between these two parts. The ring is made of ABS and the cone is made of aluminum.

I’m using the last version of Nastran In-CAD as well as the last version of Inventor.

My settings are the following:

• I used half model to simplify (I tried with a quarter of a model but I have problems with the mesh...)

• No initial contact between the two geometries
• Contact with the following settings: Separation / Unsymmetric

Slave Entities:  

Master entities:

• Loads: enforced motion of 8.5mm in Y direction

• Constraints:
  • Symmetry plane ZY

• Constraint in the Y direction for the enforced motion

• Blocking of all directions and rotation on the rear face of the cone except Y translation

• Locking of the translation of the ring in Y direction on just one surface

• Non-linear settings:

• Settings of the analysis:

Questions:

1. Does anyone see anything wrong with the settings?
2. I don’t really know what to do with the ring. What boundary conditions would you apply to the ring?
3. Is this simulation possible with Nastran In-CAD?

The assembly file is available in appendix

Thank you in advance for your help,

Best regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Roelof.Feijen]

Hi @Anonymous 

Could you also add the two IPT files?

Assembly only is not enough.

I have my doubts about the Constraint in Red (Face 12).

This simulation is definitely possible in Nastran in-CAD.

[Roelof.Feijen]

Hi @Anonymous 

Could you also add the two IPT files?

Assembly only is not enough.

I have my doubts about the Constraint in Red (Face 12).

This simulation is definitely possible in Nastran in-CAD.

[Anonymous]

hi @Roelof.Feijen,

You right, sorry I forgot the .ipt files. I added them to this message.

Regarding the constraint in red (Face 12), I put this constraint after reading this article:

https://knowledge.autodesk.com/support/nastran-in-cad/troubleshooting/caas/sfdcarticles/sfdcarticles...

Maybe I misunderstood the solution. 

I thank you in advance for trying to help me, I appreciate it!

Regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Anonymous]

hi @Roelof.Feijen,

You right, sorry I forgot the .ipt files. I added them to this message.

Regarding the constraint in red (Face 12), I put this constraint after reading this article:

https://knowledge.autodesk.com/support/nastran-in-cad/troubleshooting/caas/sfdcarticles/sfdcarticles...

Maybe I misunderstood the solution. 

I thank you in advance for trying to help me, I appreciate it!

Regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Roelof.Feijen]

Thank you for the other IPT's. I will take a look at it now.

[Roelof.Feijen]

Thank you for the other IPT's. I will take a look at it now.

[Anonymous]

You're welcome, that's very kind of you @Roelof.Feijen 🙂

I hope you can help me move forward a little bit with this simulation 

Regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Anonymous]

Regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Anonymous]

You're welcome, that's very kind of you @Roelof.Feijen 🙂

I hope you can help me move forward a little bit with this simulation 

Regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Anonymous]

Regards,

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Roelof.Feijen · Accepted solution]

Hello @s.gugelmann,

I have been struggling with your model for some time, although this should not be too difficult to solve.
There is a problem with Cone_Halb.ipt. First, I thought it had something to do with the derived part missing. However, it is not. I save the file as a Step and imported it again (I saved it as Ring_Halb2.ipt by mistake). Placed everything in a new assembly and recreated the analysis.

Note the changes that I made in relation to your analysis:
1. I used a quarter of the model instead of a half. This makes it easier to constrain / stabilize the model. It will also reduce the number of contact elements and nodes.

2. Contacts
Penetration type = Symmetric contact
Stiffness factor = 0.01 (because of the difference in stiffness between the ABS and Steel. With Steel – Steel I would use a Stiffness factor of 1, but not in this case.)
Master face should be the outer face, slave should be the inner face.
I selected less contact faces, which creates less contact elements.
Added a maximum activation distance of 10 mm. Again to create less contact elements.

3. Nonlinear setup > Advanced Settings
Stiffness Update Method = AUTO
Convergence on only Displacement and Work.

4. Mesh use linear instead of parabolic. I preferred linear in these kind of analyses.

The files without results are attached.
If you want to have the files with results, you can download them here https://we.tl/t-LdUAsyZTIP

(Note: Link is available for 7 days).

[Roelof.Feijen · Accepted solution]

Hello @s.gugelmann,

I have been struggling with your model for some time, although this should not be too difficult to solve.
There is a problem with Cone_Halb.ipt. First, I thought it had something to do with the derived part missing. However, it is not. I save the file as a Step and imported it again (I saved it as Ring_Halb2.ipt by mistake). Placed everything in a new assembly and recreated the analysis.

Note the changes that I made in relation to your analysis:
1. I used a quarter of the model instead of a half. This makes it easier to constrain / stabilize the model. It will also reduce the number of contact elements and nodes.

2. Contacts
Penetration type = Symmetric contact
Stiffness factor = 0.01 (because of the difference in stiffness between the ABS and Steel. With Steel – Steel I would use a Stiffness factor of 1, but not in this case.)
Master face should be the outer face, slave should be the inner face.
I selected less contact faces, which creates less contact elements.
Added a maximum activation distance of 10 mm. Again to create less contact elements.

3. Nonlinear setup > Advanced Settings
Stiffness Update Method = AUTO
Convergence on only Displacement and Work.

4. Mesh use linear instead of parabolic. I preferred linear in these kind of analyses.

The files without results are attached.
If you want to have the files with results, you can download them here https://we.tl/t-LdUAsyZTIP

(Note: Link is available for 7 days).

[Roelof.Feijen · Accepted solution]

Hi @Anonymous,

Note that I made some changed to the default Nastran settings / parameters.

Not sure what will happen if you the analysis that I created.

Otherwise Reset the parameters to default.

[Roelof.Feijen · Accepted solution]

Hi @Anonymous,

Note that I made some changed to the default Nastran settings / parameters.

Not sure what will happen if you the analysis that I created.

Otherwise Reset the parameters to default.

[Anonymous]

Hi @Roelof.Feijen,

That's fantastic! Thank you so much for the time you took to explain and solve my problem!

I'm going to need a little time to analyze your simulation and understand my mistakes. But I have everything I need!
I also download the complete file, it also works for me! Great. Thank you also for the detailed explanations of the contact settings, it helps me a lot.

As you may have noticed, I am not yet a Nastran In-CAD specialist, but I am very good with Autodesk Moldflow. So if you have any problem with Moldflow, let me know!

Best regards,

Serge

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München

[Anonymous]

Hi @Roelof.Feijen,

That's fantastic! Thank you so much for the time you took to explain and solve my problem!

I'm going to need a little time to analyze your simulation and understand my mistakes. But I have everything I need!
I also download the complete file, it also works for me! Great. Thank you also for the detailed explanations of the contact settings, it helps me a lot.

As you may have noticed, I am not yet a Nastran In-CAD specialist, but I am very good with Autodesk Moldflow. So if you have any problem with Moldflow, let me know!

Best regards,

Serge

Serge Gugelmann - Technik

s.gugelmann@ricone.de

RICONE GmbH

Trausnitzstraße 8

D-81671 München