Radiation Heating of Tube

jorgeanrosas

Hello. I am trying to heat the inner surface of a tube from a rod radiating heat (it can be thought as a black body at 800 C). I understand that CCX can run this heat transfer in the steady state mode, but I when I add the radiation constraint I get a message saying the CCX mode does not be used in this solver.

Can somebody provide any advice?

Victor

I'm not sure why that's warned against. It might be that it sometimes generates solutions with time steps but I can't see that happening. It does seem to work OK if you ignore the warning though so as long as you're confident that CCX is alright with it, that should be OK.

Or you can use transient and very small values for specific heat capacity and density so it only ends up doing one or two time steps.

jorgeanrosas

Thank you. I tried to run it in the Transient mode, and the probelm is the same. When I select CCX, then radiation becomes an unavaliable option for me, and the internal solver does not model radiation between two elements. This is a slide of what Mecway is showing me and a simplified model of what I want to do.

It is in principle I just need to see what is the temperature map of the enclosure surrounding a constant temperature hot "rod".

Any advice? Thank you.

Victor

Does the right hand image show thermal transient? It should be available. Here's a simple example that solves.

However, this will only work in the same way that it does with the internal solver. To get radiation between surfaces, you'd have to modify the CCX input manually. You might be able to do that through CCX -> custom model definition and custom step contents but I'm not sure about how to set it up.

DaveStupple

Victor, your example above works for me, but a large model that I'm working on has the radiation unavailable as for Jorgean. It solves with Mecway's internal solver but later I'll want some of the CCX functions. I'll have to scale the model as small elements are causing it to fail with CCX (as we saw some time back) - I'll report back with how CCX reacts to the radiation.

Victor

Oh. Can you confirm that it's the thermal transient analysis type, not steady state? If so, it would be great to see an example of it being missing.

Radiation doesn't seem to work reliably with steady state with CCX. It might be something to do with the default time step size that Mecway sets or the load ramping the CCX does. Or it might just be that the examples I tried were too extreme.

DaveStupple

Victor, sorry I've not been thinking or communicating properly. I'm using thermal steady state.

When I take your simple example and switch it to CCX steady state, it does not like the radiation.

When I take my current model and switch to CCX transient, the radiation remains unchallenged. The model won't solve because I need to scale to increase the size of tiny elements. It goes through the motion of solving using internal transient, but comes up with nonsense, probably because I failed to enter density and specific heat properties. I'll have a tinker and see if I can get it to run sensibly.

Using transient could be a clunky and slow workaround, if need be.

With my present model, radiation is insignificant and could be turned off, but I want to leave it for now as the model is running too cool so I need to drive up the temperatures a bit. Hopefully radiation will remain insignificant and I'll chuck it.

JohnM

I ran this using Steady and Transient with t=100000, both ran fine and gave same answer. Running CCX 2.16, tried different solvers, all worked. Even though Mecway gives warning, RADIATE command still gets written to CCX deck.

DaveStupple

Good spot John, and thanks for the info. That's encouraging. I did wish you Happy New Year on the Salome thread, but I don't think it has published the comment (though the Forum credits me with being the latest contributor to the thread). So, even though the state of the world currently makes it almost impossible not to raise eyebrows ironically to camera, happy New Year John!