2D Heat Transfer Simulation with motion

[Anonymous]

Hello!

I'm trying to simulate contact heat transfer between materials. It is 2D transient simulation and I attached linear motion to the upper die. After 2-3 time steps simulation crashes. 

I have tried two diffent ways of modeling planes in Inventor:

1) all planes as individual sketches and whole file saved as one part

2) all planes as individual parts and loaded in CFD as assembly file

Mesh on air and solid is the same size as suggested in other topic. Hot region has finer mesh.

I noticed that when I assign air material to the plane it only attaches "visible" part and it trims around solid; leaving a hole after hiding solids.

After disabling motion simulation runs perfectly.

Cheers,

MK

[Jon.Wilde]

For motion, move the part above and outside of the air in CAD, then use the 'initial position' to move it back into the correct place in CFD. This way we build the fluid mesh before overlaying the solid one 😉

You might also need a smaller time step, no idea without seeing the model settings though 🙂

What are you actually trying to achieve? Does significant heat transfer happen before the two parts touch? 

What is the overall goal? The main question you are trying to answer?

Thanks,

Jon

[Anonymous]

Thanks for answer 🙂

I'm trying to evaluate contact heat transfer coefficient between Al7075 and Steel. Ofc should be much higher than cooling Aluminum plate in air 😉 This is simplified model for hot stamping. I will be comparing results with exact real experiment. 

I finally ignored motion and simulated with dies closed down. 

BTW I tried your tips, moved the solid plane in design out and moved it back by initial position. Time step is 0.01s, 1 iteration per time step and it still doesn't work.

MK

[Jon.Wilde]

I guess it depends on the speed they are moving, if you needed motion I suspect there is a solution there 🙂

I asked about what you needed as I suspected the same, just start with them together and set initial Temperature conditions to every surface (real world volume) and that might be sufficient.

Did it work out OK?

Thanks,

Jon

[Anonymous · Accepted solution]

With dies closed works well 🙂

But I'm still wondering why it doesn't work with moving object... I've changed time step to 0.001 and motion is 25mm/sec so it's not that fast. I've tried with uniform mesh 0.5mm and 1.0mm.  But I think problem is somewhere else. After changing results to velocity I've spotted that it's growing very fast to irrational values. It would go to infinity but program stops. It happends on the edge of a fixed solid. Happends with and without mesh wall layer...

I can remember I've had something very similar when I was simulating vortex shedding but it was at inlet of air velocity. In this one there is no velocity, just pressure at the air plane edges.

MK 

[Jon.Wilde]

Strange! I am happy to take a look if you would like to share a CFZ file 🙂

Did you start with the moving solid outside of the fluid? Might be worth testing this also, please offset it a known distance above the model if you do choose to share.

Thanks!

[Anonymous]

Nah, wasting time. I've got my solution. I wanted motion just to make it look good 😄

Anyway thank you for help 🙂

I've deleted corrupted scenarios.

Cheers,

MK