Air-spring modelling

[O_Giannis]

Hi all,

I have been redirected here as it is a more Abaqus forum.
I have designed a model - draft mode in the best of my abilities - I am in a beginner level.

Is there any chance someone can help me to fix some issues? I have the model so I can provide if someone is willing to support me.
I don't care that much at that level for the result values, but rather for the model to work kinematically.

Thanks in advance team!

 

[FEA way]

Did you manage to fix that issue with surface selection discussed on FEA subforum ? If yes, what are the current problems with your model ?

Sharing the cae file here is definitely a good idea if you need some specific help.

 

[NRP99]

O_Giannis

We are here to help. But you have not posted your "problem description".

 

[O_Giannis]

Hi guys, thanks for your answers. The problem lays on how I model it.
I could not fix that as there were other parameters that i should add/change to do so.

Is it ok to have a first "consultation" on the model?

https://files.engineering.com/getfi...e7578cf&file=Rear_Assembly_3D_Deformables.zip

 

[FEA way]

Most importantly, there’s something strange about the way you modeled these parts. They are all hollow and treated as shells but it’s not their midsurfaces that were extracted - just the outer surfaces. It seems that you used the Shape —> Shell —> From Solid option. It can be useful sometimes (for example when creating discrete rigid parts) but not in cases like this one where you apparently just want to utilize the fact that these parts are thin-walled and mesh them with shell elements. In such a case you should use the Assign Midsurface Region tool and then follow it with Offset Faces to automatically create midsurface. Then you may have to manually fix it because the geometry has some details that may interrupt the automatic extraction process.

 

[NRP99]

Can you explain your problem? May be some pictures can help. Sorry, not able to open the CAE file with version 2022 since I am working in yesteryear version.

 

[O_Giannis]

@NRP99, nothing to worry about, after all the model might can t work at all.

@FEA Way, maybe it could help to first say what I want to model and then get some suggestions on how to do it from you guys?

I need to model an airspring which will need:

3 Parts (Baseplate/ Sleeve/ Piston) - ideally I would like to model the in between layers of fabric (like the biased tread we have in the tires) - but It would be a pain in the a*** as I would have to create a very thin layer and introduce biased (45 deg) steel cylindric elements as bars (rebars/reinforcement).

Then there should be a tie between the piston and the sleeve, and between the sleeve and the baseplate.
(

https://www.youtube.com/watch?v=CIFdvEkBrLk

something like that).

Then there should be the cavity fluid effect with the rubber sleeve and the piston cavity as fluid cavity. The results however should depict the effect on the circumference of the sleeve.

 

[O_Giannis]

I am thinking if maybe an axisymmetric model would be easier and better

 

[FEA way]

There are some simplified ways to model such a thin fabric. You could use the skin feature for example. Another way would be to utilize composite shell elements if you want to account for the layered structure. Discrete modeling of reinforcement shouldn’t be necessary in this case.

Start by fixing these shell models, as I described above. Then you can add complexity to the model. If you need a more accurate representation of contact between the parts, continuum shell elements can be useful (the geometry needs to be solid in this case).

Build your model gradually (for example, leave the fluid cavity definition for later stages of preprocessing) and check if it still works to avoid many of the potential issues.

An axisymmetric model is indeed a good idea unless your structure has any geometric features, loads or expected responses that aren’t axisymmetric and that can’t be omitted. But in this case, you should be able to use this assumption.

Have you seen the Abaqus documentation example "Hydrostatic fluid elements: modeling an airspring" and Simuleon’s blog post "Simulating a rubber fibre reinforced air spring using Abaqus" ? Some research papers may also be useful, for example, the one titled "Design of Stiffness for Air Spring Based on Abaqus" by H. Li et al.

 

[O_Giannis]

Hi FEA way,

In case i want to see the stress propagating on the cords of the fabric layer though, do you think that I should use the skin option? or the composite shell? I am so confused on what to do in the best way possible! Obviously it would help to have some experience more in order to understand easier the challenges of each modelling technique.

I ll have a look on the researches you proposed! Definitely helpful!

Thanks a lot!

 

[NRP99]

Use of shell modelling with creating offset faces from the solid part would be way to go. Axisymmetric option is simple one for the initial understanding of the problem. But I would probably use the shell model for final analysis as done in the youtube video you shared.

 

[O_Giannis]

Thanks NRP99! I ll have to write that down on my list as another potential technique to learn!