Humidity boundary condition applied to an edge?

[paul.bunuan]

Is it possible to apply a steam boundary condition (e.g., select humidity and enter a value of say, 50%) to an edge of a 2D axisymmetric model? Or do you have to apply this to a surface? Thanks,

paul

[Jon.Wilde]

Sure, that is correct.

Volumes in 3D = Surfaces in 2D

Surfaces in 3D = Edges in 2D 

Hope that helps

Jon

[paul.bunuan]

So I applied a Quality BC to an edge (in my case, 0.8 to represent 80% steam quality) as well as a film coefficient (see attached image) to represent a hot steam gas being applied on a steel surface. For reference, the geometry is of the structure only, so no fluid is modeled in this case. When I attempt to solve the scenario, I get a message stating that I need to select a fluid material with Saturated in the name. My sense is that this material allocation is for the fluid that is in contact with the structure, which for a 2D case, would be an adjacent surface in contact with that edge.

Do I need to model the contacting fluid for steam quality BC to work?

[Jon.Wilde]

Hi Paul,

Yes, you would only apply a steam BC to a fluid.

Shall we take a step back and look at your model so we can get an understanding of what you are trying to achieve?

Could you share some images and goals?

Thanks,

Jon

[paul.bunuan]

The problem that I'm trying to solve is actually an inverse heat transfer problem. I have a 1" thick wall stainless steel canister  that is 30" in outer diameter. Inside it is water at room temperature that is filled to some unkown height. Above the water is a hot gas that ranges anywhere from 300-400degF (I'm assuming a low h value).

I have thermal data on the outside of the canister wall that tells me temperature vs time along the height of the canister.

My goal is to figure out what is the height of the water to within +/- 2 inches.

My first approach to solving this is to model a 2D axisymmetric profile of the canister and then apply a range of hot gas temps and water temps on the inside wall of the canister. Then run the simulation over a period of xx seconds to see if it matches my data. I set up the 2D model to just represent the canister (and not the fluid) because I thought I could represent the fluid/steel interface as a BC and therefore minimize the size of the model. Maybe I need to include the fluid?

Ideas? Comments? Am I approaching this completely the wrong way?

[Jon.Wilde]

Hi Paul,

Thank you for sharing, here are a few thoughts:

1. I think for accuracy, you shouldjust model the fluid and the internal gas (with a sliver of solid splitting the two as we cannot have two different fluids touching). This would mean CFD calculates the correct heat transfer from internal fluid to the solid canister
   The solid between the two could have the same properties as the gas or water, doesn't matter so long as you know the water height. Being 2D the overhead would be so minimal
2. Perhaps think about running a laminar, conduction only analysis (flow would be un-ticked) - or even using solid materials to represent the gasses, I doubt there will be too much internal convection
3. Ensure that you set Initial temperature Conditions to every surface (volume in 3D)
4. Utilise Monitor Points (right cick within setup to assign them) on the outside of the canister at points of interset. This way you can monitor temperatures over time in the convergence graph, again with no overhead, I find these very useful
5. The only boundary condition would be the external film coefficient on the external edge(s), which appears to be a known variable

This should work OK I think, please let me know if you have any further questions.

Thanks and have a nice weekend 🙂

Jon

[paul.bunuan]

These all sound like good suggestions, which I will start implementing. And, since I didn't see anything implying that what I'm doing is completely, I feel more confident that I'm going in the right direction. Thanks for the input! Depending on the results, I may ping on you again for more advise.

[Jon.Wilde · Accepted solution]

No problem, please do ask if you need more help. Or mark the post as a solution if it provided you with one 🙂