Learning Forum

[deepesh.p.gurdasani]

How to model rivet and double rivet joint? Can we use beam connection? Or if in 3D, what contacts to be given ? 

[Deepak]

Hi,

Please refer to this post: Lap Joint Rivet Analysis

Thanks,

Deepak

[deepesh.p.gurdasani]

Thanks for link of discussion. Can we use beam connection instead of 3D rivet?

[Deepak]

For double rivets, using a beam connection is not a feasible option.

[deepesh.p.gurdasani]

For double riveted, we have to mandatorily go ahead with the 3D model and connections ? Or anyother method similar to beam connection ? 

[Deepak]

You can try with 1D line Elements.

[deepesh.p.gurdasani]

ok. But in beam connection we get to choose the surface areas of imprints , right ? so on first plate top and second plate bottom, both the areas can be selected. Any specific reason that this wll give incorrect answer? 

[Deepak]

You cannot scope the beam connection to the middle plate in a Double rivet Joint.

[deepesh.p.gurdasani]

'Scope to the middle plate'? There are only two plates that we are considering here, right? 

[Deepak]

Sorry, I thought butt joint; if you are using Lap Joint, you can go with the use of a beam connection.

[deepesh.p.gurdasani]

Thanks for the clarification. Also IF we consider Butt joint, we should use 'Extract' option in spaceclaim to convert rivet into line body, right? in this case, we will have to remove rivet head and tail parts, right because those parts wont be converted into 1D.

[Deepak]

Yes, you are right.

[deepesh.p.gurdasani]

 

Also, for Riveted butt joint, The contacts to be given to 1D elemnt would be same as the link mentioned by you in one of the above messages ? 

Here, upper vertex of 1D element to upper surface (Or projected surface of the head that was removed) on the plate – Frictional

1D element shank to holes of “only” “middle plate” (As first and third plate will have contact with the vertex of 1D element) – Frictional 

Lower vertex of 1D element to third plate’s surface (Or projected surface of the tail that was removed) – Frictional

AND Contacts between 1st to 2nd plate & 2nd to 3rd plate – Frictional 

Is it correct? 

 

[deepesh.p.gurdasani]

If anyone can guide? 

Thanks in advance.

[deepesh.p.gurdasani]

Is there a need to create separate post for this query ?

[Deepak]

Hi,

Sorry for the delay in my response,

You may define the MPC contact between vertex and head imprints. The remaining contacts are correct.

Thanks,

Deepak 

[deepesh.p.gurdasani]

Thanks for your response. MPC won't work for 'Frictional'. MPC is only working in 'Bonded'. In real scanario the head and tail of rivets are to be in contact as 'Frictional', right ?

[Deepak]

Now you replaced the head with a 1D element, bonded contact would be a better option.

[deepesh.p.gurdasani]

Any specific reason for that, Deepak? because real scenario wouldn't have bonded contact with the plate.

[Deepak]

Frictional contact is a nonlinear contact type and is slightly more computationally expensive than linear contacts like bonded or no separation. However, using a linear contact in place of a frictional one in such scenarios would be an oversimplification and could adversely impact the accuracy of the results.

[deepesh.p.gurdasani]

Ok. I tried this. Till I am using 'Force' as a load, this runs ok but as soon as I change/apply it to 'acceleration', it shows error "Internal magnitude exceeded' . 

[Deepak]

 

Make sure the model is constrained properly.

Can you share any screenshots of the model?

 

[deepesh.p.gurdasani]

This is the model (Quite simple model)

 

 

 

[Deepak]

Let me try it at my end.

[deepesh.p.gurdasani]

Hello Deepak, did you try?