Learning Forum

[Berkay Özfidan]

Hello,

 

-->

-->-->-->

-->

Sure, here is the direct translation of the text to English:

Thermo-Structural Analysis for Turbocharger Turbine Housing

I am working on a thermo-structural analysis for a turbocharger turbine housing for my school project. I need to define supports from the boltholes in the flange areas of my part, which consists only of the turbine housing (as required). After completing my steady-state thermal analysis with appropriate boundary conditions, I couple the results with my static structural analysis. When I define fixed and cylindrical supports to account for thermal expansion, I observe abnormally high stresses (around 5000 MPa) around the boltholes. I do not encounter any issues when I apply the 3-2-1 method to vertices, but I am required to define supports from the inner surfaces of the boltholes for the part that will be fixed with the boltholes. Additionally, I will optimize the part and modify the shape of the boltholes to examine the temperature-stress distributions. Applying the 3-2-1 method makes the boltholes in the flange area seem ineffective.  

Note: In addition, I have tried all combinations of free/fixed options for the Tangential-radial-axial options for Cylindrical Support. Also, i did apply fine mesh around boltholes. But results were not changed.

[peteroznewman]

I suggest you add a separate flange body with a bolt hole pattern to match the bolt holes in the turbine housing. The flange can be a rectangular block that can have the 3-2-1 kinematic constraint pattern to ground.  It can be the same material as the turbine housing so the CTE will match.  It can be given a similar temperature to the base of the turbine housing.  Use 4 fixed joints to connect the bolt hole pairs to each other. Make the fixed joint have behavoir = deformable.  Taking those steps will go a long way to removing artificially induced thermal stress by fixing those holes directly to ground.

Another approach is to use a single remote displacement on the four bolt holes and set behavior = deformable and set all six DOF to 0.

[Berkay Özfidan]

Thanks for your answer, 

Can you explain why the cylindirical support doesn't work in my case? As i mentioned I tried all the combinations for cylindirical support options (tangential-radial-axial free/fixed).

I did try your suggestons but dummy component (adding/drawing another flange) is not allowed, also can you explain remote displacement part a bit more in detail. Because i did try remote displacemenet as: 1)Select 4 boltholes inner surfaces(as it's shown in my figure above), 2)insert remote displacement, 3)behavior = deformable and 4)set the all six DOF to 0. 

But the point is once i tried with only one bolthole instead 4 boltholes results were same but constraint was changed. So it's impossible.

Second thing is, do you suggest that path: 1) Select 4 boltholes inner surfaces, insert remote displacement (this time placing remote displacement x-y-z coordinates due to dummy flange component back side(opposite side due to contact surface).(dummy flange which is not necessarily dwarn in this case) )