Pressure loading function of normal angle

[Rem38]

Hi everyone !

I'm currently working on a mechanical simulation which aims to size a tool in compression.
Compression faces aren't staight and the pressure act as a function of the normal angle. Actually the pressure is constituted by a normal and a tangential (shear) component. At zero degree, pressure is completely normal, at 90 degrees, pressure is completely tangential. Please find attached a little illustration of the pb.
Does anybody know how it is possible to apply a load (pressure) which directly depends on geometry ?

It seems that surface traction could be the solution, but I have no idea of how to configure it, and I can't find anything on Abaqus documentations.

In advance, thanks for your help !

 

[Mustaine3]

I'm not sure I understand your question. It sounds like you are asking for something, that the regular pressure load is doing automatically. This load is always applied in normal direction at each element face.

 

[Rem38]

Hi Mustaine3, thanks for answering.

Yes, sure, with the regular pressure load the load is always applied normal at each direction face. Actually I don't ask about the normal direction but about the value of the pressure with respect to the normal angle.
The pressure I want to apply is completely linked to the normal angle. If the normal angle is 0°, the pressure should be 100 MPa (for example), for 45° pressure should be 106 MPa, and for 90° it should be 50 MPa, the pressure function is written on the illustration.

Thanks for your help

 

[Mustaine3]

Ahh, ok. But I think it is not possible to do what you want. There is imho no way to link the load to the normal direction; not even in the DLOAD or ORIENT subroutines.

Why do you want to do that? What surface load in real life requires this?

 

[Rem38]

Ok, thanks for your answer.
The field i'm working on is the powder metallurgy. My company produce sintered mechanical part for the automotive sector.
The powder is compressed and then "heat" to be sinter. During the compression, the powder behavior is quiet similar to fluid : the axial pressure produce a radial one. So the tools we use are subject to both pressure components, the resultant is normal to the surface and his value depends on the normal angle.

For the moment I'm used to split the surfaces "by normal angle", I was looking for a better way to do it.