[WanderleyGrespan]

Hi guys. I need a help. I'm not so good on simulations but I trying.

This is a furnace, and inside there's a cylinder (Steel) initially with 480°C and it will be cooled.

There's 12 inlets with a mass flow(air) and only one Outlet and I want to know how is the decreasing of the temperature along the time. I have considered only heat transfer by convection.

At the images you can see some of the settings that I did for solving the problem.

And I need to know: Can I trust the results? If I can, how much?

Apparently I don't have to many knowledge about time steps and convergence and for this reason I ask your opinion about what I did and what I can do different.

Please, it will be very helpful for me.

Thanks!!!

[ghorrocks]

Can I trust the results?

No. That simulation clearly has not converged.

First of all read this FAQ: https://www.cfd-online.com/Wiki/Ansy...gence_criteria

In this simulation you clearly have a large separation in time scales. The solid domain is much slower than the fluid domain. So use a solid time scale factor to accelerate convergence in the solid domain. A factor of 10, 100 or 1000 should do it.

[WanderleyGrespan]

You are right Gleen. My time step it's not good.

But Can you se the other settings?

I forgot to post before. Can you give me more ideas and tips for solving the problem.

How I said before, I'm still leaning and it will be very helpful

[ghorrocks]

Oh, I see your simulation is transient. In that case forget about the solid time scale factor suggestion. That only works steady state.

For CHT simulations you should include imbalances as part of your convergence criteria.

Unless you have done a time step sensitivity check your time step will be wrong. Instead I recommend you use adaptive time stepping, homing in on 3-5 coeff loops per iteration. Make sure the minimum and maximum are wide enough you never hit them. Also the initial time step needs to be sensible. Then the solver will automatically find its own time step size.

[WanderleyGrespan]

Following your tips about the adaptive time step, I must say, i've never used this option and for this I don't have knowledge of how I supposed to do.
What do you recommend for those options?

[WanderleyGrespan]

Do you have any comment about my fluid models tab (turbulence model)??

[evcelica]

It really seems like you should do some of the CFX tutorials. They are very beneficial and you will learn all that you are asking.

Instead of modeling those little inlets, I would have used source points injecting fluid, but that's besides the point. You need to go over the tutorials.

[WanderleyGrespan]

Originally Posted by evcelica

It really seems like you should do some of the CFX tutorials. They are very beneficial and you will learn all that you are asking.

Instead of modeling those little inlets, I would have used source points injecting fluid, but that's besides the point. You need to go over the tutorials.

Thanks Evcelica.

"Some of the CFX tutorials." Can you be more specific about what tutorials you are talking about?

[ghorrocks]

All of them

If that is too much for you then look through the list and do the ones which look relevant to you.