Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

touch connection

Hello,

I would like to create a connection, which interfere one directional movement, and the reverse directional displacement is allowed. This is a connection between a fixed part, and the other designed part.
Can you please help me which connection type should be used, and how?

Thanks a lot,
Sanyo

Find more posts tagged with

English

HyperMesh

Accepted answers

All comments

Rahul_P1

sanyo55,

Looks like you need a constraint, you can constrain the nodes that are shared by both the parts (nodes in the region where the two parts are in contact) using SPC in RADIOSS bulk or similar card in other solvers,

You need to go to analysis > constraints panel or BCs > Create > Constraints to create these.

Dofs with a check will be constrained while dofs without a check will be free. So if you want the nodes in the connected region to be free to move in x direction and not in y direction or other directions you will need to check DOF2 and the other DOFs leaving DOF 1 free.

If movement is in a non axial direction, create a local coordinate system with the axes in the direction of movement and constraint, Assign the nodes which are being constrained to this coordinate system using the systems panel and then constrain the nodes as above.

Rahul_P1

Thank you for the quick reply.

I wanted to create a constraint like a half DOF. So enable to move the node for example '+x', but disable to move '-x'.
In order to do this I created a rigid part, which is fully fixed and created an interface connection between the analysed and the fixed part. With slide connection the 'y' and 'z' movement is allowed, but the '+x' is fixed and the '-x' as well.
I hope this explanation is now more clear.

Thanks a lot!

Rahul_P1

Hello,

I think, that the versions are up to date now, I attached the copied system information.
But unfortunatly the touch connection not yet working.

Thanks in advance

Rahul_P1

Hi,

I would rather suggest to do a explicit contact option on this point. I do not know the complexity of your model but via Analysis -> Contact Surfs you can create two Contact surfaces on your two parts.

Via Analysis -> Interfaces you can combine them together and define master and slave contact surfs.

In this case contact is properly modeled and you can change the value in the interface card to freezed if you do not want to deal with friction or anything like that.

Hope this helps.

Rahul_P1

Thanks Merula! Now I did it exactly the same way, but it seems, that when the two parts would move away from each other, it is not allowed. Which setting should work like a 'touch connection'?

Rahul_P1

sanyo55

Contacts may be the way to go for you as merula suggests, I however need a clearer idea of what you are trying to simulate here, If you can explain with an image as to the kind of interface you want also maybe give an idea of what reality condition you want to simulate I can perhaps suggest the kind of contact best suited for you.

Rahul_P1

Ah I see,

Freeze is not an option then because it sticks the two together no matter what.

Try Slide or Stick. You have to create an interface group with the type of contact. The Contact card is explained here:

http://www.altairhyperworks.com/hwhelp/Altair/hw11.0/help/hwsolvers/hwsolvers.htm?contact_bulk.htm

Rahul_P1

sanyo55,

Here is a simple procedure for that just in case
Interfaces panel --> select type as CONTACT --> add the master and slave csurfs --> edit the contact and set type in card image to SLIDE.

Rahul_P1

Thanks

Altair Forum User

Hi,

I have similar problem. In this case the model is working fine (let's forget about the connection between the bearing and the gearbox), i.e. the bearing pushes away the flange. However, if I change the direction of the force, in which case the bearing should slip out and has no effect on anything, the bearing still pulls the flange. What do I have to modify to make the bearing available to push the flange but not pull?

Thank you

aecyd

Unable to find an attachment - read this blog