Heat treatment of a ring - thermal expansion and distortion 2

[wuumbs]

Hello everybody.

I'm trying to simulate the heat treatment of a ring to analyse the thermal expansion and distortion.

The inside diameter of the ring is 2700 mm and the outside diameter is 2400 mm. It's made of 42CrMo4. So as you can see, it is very heavy.

At first tried to simulate the heating-up process. As you can see at the following picture, I created a ring and 20 identic supports. These supports are made of GX40CrNiSi25-20.

I picked "coupled temperature-displacement" element type for all parts. Then I created both materials and added the following material properties: conductivity, density, young's modulus and expansion coeff. alpha.

In the initial step I set up a displacement for the support's bottom side (U1, U2, U3). Then i predefined a constant temperature of 20 °C for the ring and all supports. At least I created a surface-to-surface contact between the bottom side of the ring and the surfaces the ring is linked to.

For the surface-to-surface contact i used the following contact properties:

- Normal behavior: Hard contact, default, allow separation after contact

- tangential behavior: penalty, isotropic, friction coeff. 0.2

Then i created a step - type: coupled temp-displacement (steady-state)

In this step I just added a BC for the heating-up process: temperature of 870 °C. This temperature is added to all parts (supports & ring).

When i run the simulation, its working. But in the post-processing the ring is just sliding/moving on the supports... there is no thermal expansion / distortion.

PS: In reality the oven is preheated to 500 °C. When this process is finished, the ring (with 20°C) is put into the oven. Then the oven gets heated up to 870 °C. After the heating-up process follows a cooling process back to 20 °C. It's very important for me to get this in my simulation too, but i actually I have no ideal how to handle this.

Do you have some information for me how to get my simulation working? I'm thankful for every answer.

PSS: Sorry for my bad english, my mother language is german.

 

[corus]

For this model you should have used symmetry to reduce the size of model to one quarter of its size. Presumably the loads would be due to self weight as well as the thermal load. It could be that the displacements are just showing some kind of rigid body motion. Symmetry restraints will prevent that.
Incidentally you could have improved the mesh quality by using structured regions to obtain a much more uniform mesh. The supports also seem to be made up of tetrahedral elements, although it's hard to tell from the picture. You'll get better results using brick elements throughout.