Learning Forum

[st2021]

(Branched off /forum/discussion/29031/meaning-of-constants-in-rmodif-in-apdl-commands/p1

the purpose eventually is to model friction of a stationary block that touches a rotating block, and the resulting temperature increases, both bodies made of structural steel

after replicating the tutorial like https://www.youtube.com/watch?v=_hrZVJg8b3o now I am adding rotation via a body-ground rotating joint, and joint load. The problem is the temperature of the rotating block starts to decrease.

simplified the model, forget the friction, just have one rectangular thin block rotating around axis perpendicular to its large face (with body-ground rotating joint and joint load). same issue

how can it be avoided? I would like friction to be transferred to heat, and no heat losses to be included in the model

Thanks

[1shan]

Are you using a revolute joint or a cylindrical joint? A revolute joint has only 1 DOF (along axis of rotation), this means there will be no axial movement/ normal reaction, so no friction. Also, first start with a simple transient structural, check if the contact results are realistic and then do the coupled field analysis.
Regards Ishan.

[st2021]

Hi, thank you for your reply. It is a revolute joint. The body is flexible, not rigid. So I was hoping with rotation in one axis of rotation friction would still work: it would deform locally in whatever area the second block is touching.
I don't mind if there is no friction, but it is not supposed to cool down, is it? It is a simulation without coupling, only structural thermal.

[st2021]

To put it simpler: a flat thin rectangular body is rotating around one axis perpendicular to its large face. There are no other degrees of freedom. in a transient structural simulation TEMP is undefined. When I enable structural thermal via "et,matid,226,11" APDL command, and start to rotate the body, the result is that its temperature drops. Increasing number of time steps helps to keep the temperature closer to its initial value but then the number of time steps required for this is massive. It appears to make no sense to me from physics or numerical point of view. Advice would be appreciated. Many thanks.