[Knowledge]

Hi

When I unable the constant physics I can not get the solution/solver manager to work. I get an error message. When I try to uploade the error file or open it I can't!

What can I do ?

[ghorrocks]

Leave it at defaults then. It is an expert parameter, so it is not guaranteed to work all the time, and if you choose to use it you have to be looking for unexpected behaviour.

[Knowledge]

Originally Posted by ghorrocks

Leave it at defaults then. It is an expert parameter, so it is not guaranteed to work all the time, and if you choose to use it you have to be looking for unexpected behaviour.

I never really get a result-oriented answer I can leave it but can you tell me how I can solve the problem?

I think maybe the problem comes up because I have an interface between the water and air and not a solid-domain. I have a bottle, half of it is filled with water and the rest is air and all it's surface is heated up with the same temperature which is constant. Therefor I have a multicomponent-problem.

I have looked into the Steam Jet tutorial as I think it is similar to my problem in regard to the fluids (I have water and air): https://ansyshelp.ansys.com/account/...%23cfxSteaOver

However, I can still not figure out an answer to my question. How can I solve this problem by enabling multi-components (water and air) in a transient time domain when the volume of water and air are specified?

Note: I am new into ANSYS and therefor you should provide me with details like in the ANSYS manual to understand how to follow your instructions.

[Gert-Jan]

1) you are not new to ANSYS anymore. You are here for months already, and I ranked you as an 'expert' a few minutes ago ;-)
Alternatively change your name from 'knowledge' to 'ignorance'.

2) If you at first instance don't explain what problem you are trying to solve and only tell us your frustration, how can we help?

3) you should not model this as you proposed. You have a bottle with water and air. This is just a multiphase case. You should leave Constant Physics on and define 2 fluids: air and water. Initialise the bottom volume with water and the top with air. And off you go.

[Knowledge]

Originally Posted by Gert-Jan

1) you are not new to ANSYS anymore. You are here for months already, and I ranked you as an 'expert' a few minutes ago ;-)
Alternatively change your name from 'knowledge' to 'ignorance'.

2) If you at first instance don't explain what problem you are trying to solve and only tell us your frustration, how can we help?

3) you should not model this as you proposed. You have a bottle with water and air. This is just a multiphase case. You should leave Constant Physics on and define 2 fluids: air and water. Initialise the bottom volume with water and the top with air. And off you go.

I have defined the two domains (water and air). The problem arise when I add the material for water and air. It is impossible to assign a different material to each fluid. When I assign water to water-domain it automatically also assign water to air-domain even though I change it.

How can I enforce different material for each domain?

[Gert-Jan]

As you describe it, you don't have two domains. You have one volume inside your bottle. So you have one domain.
If you would model the heat transfer through the glass, then you have two domains: inside the bottle as fluid domain and glass as a solid domain.
If you would include the air outside the bottles as well, then you can end up with three domains: inside and outside the bottle and the glass itself.
In that case you could consider to switch off constant physics since the fluid inside and outside the bottle could differ in physical treatment, or as fluid composition.

Now, your domain is filled with two fluids: air and water. This is a 2-phase calculation, not a mutlicompnent calculation.
If you would distinguish N2 and O2 in air, then it is multicomponent.

[Knowledge]

Originally Posted by Gert-Jan

As you describe it, you don't have two domains. You have one volume inside your bottle. So you have one domain.
If you would model the heat transfer through the glass, then you have two domains: inside the bottle as fluid domain and glass as a solid domain.
If you would include the air outside the bottles as well, then you can end up with three domains: inside and outside the bottle and the glass itself.
In that case you could consider to switch off constant physics since the fluid inside and outside the bottle could differ in physical treatment, or as fluid composition.

Now, your domain is filled with two fluids: air and water. This is a 2-phase calculation, not a mutlicompnent calculation.
If you would distinguish N2 and O2 in air, then it is multicomponent.

I can see that it is not a problem when I have one solid. The problem first arise when I define distinct material for each fluid (water and air). I agree with you that I have one domain inside the bottle as it is multiphase. But how can I define the top volume and bottom volume without having two distinct domains?

You can view the attached file. The top is air and the bottom is water. I have a specific volume I want to simulate for each fluid.

[Gert-Jan]

You first need to describe what exact question you are trying to answer with CFD using this setup. Because you are trying to split physics that cannot be split in real life.

So, what are you trying to mimic? Isn't this a X-Y-problem (https://en.wikipedia.org/wiki/XY_problem)?

[Knowledge]

Originally Posted by Gert-Jan

You first need to describe what exact question you are trying to answer with CFD using this setup. Because you are trying to split physics that cannot be split in real life.

So, what are you trying to mimic? Isn't this a X-Y-problem (https://en.wikipedia.org/wiki/XY_problem)?

I want to figure out how long it takes to heat up the entire water to 60 degrees Celsius in a bottle and therefor naturally the rest of the volume will be air. The bottle will be surrounded by a constant temperature of 60 degrees Celsius.

[Gert-Jan]

Why do you think you need to include the air? Did you perform any analysis by hand that tells you that conduction from air to water is a relevant effect?

[Knowledge]

Originally Posted by Gert-Jan

Why do you think you need to include the air? Did you perform any analysis by hand that tells you that conduction from air to water is a relevant effect?

Yes, if the air accounts for more than 5-10 % of the volume it is not negligible. And for all my cases it accounts for more than 5-10
%.

[Gert-Jan]

It is sound engineering practice to first do some simple hand calculations. Therefore, calculate m.cp.dT for water and air. Then again evaluate how much the air will contribute in terms of Watts.

[Knowledge]

Originally Posted by Gert-Jan

It is sound engineering practice to first do some simple hand calculations. Therefore, calculate m.cp.dT for water and air. Then again evaluate how much the air will contribute in terms of Watts.

No problem I will do that.
So I should look into the mass (m), constant pressure (Cp) and how the temperature vary in time (dT).

Can you please give an example from a random case on how to do that? Just to be sure?
I am here to learn things by people who knows more than me. I will not mind have an detailed explanation to understand how to do in the future.

[Gert-Jan]

I don't want to be rude but if you don't know what mcpdT is, one should not allow you to use CFD.

Just type it in Google.

[Opaque]

From the thread so far, it seems you have not attempted to run the multiphase tutorials.

Please read the documentation regarding the differences between a multi-component model and a multiphase model as provided/implemented in ANSYS CFX. It is not about how you think it should be, but how the software is providing the access to the models you are interested in.

There is absolutely no need for using non-constant physics for the problem you have described this far.

You will save considerable time, fewer frustrations, and considerable more help from those in this forum once you get the basics right.

Hope the above helps you.

[Knowledge]

Originally Posted by Gert-Jan

It is sound engineering practice to first do some simple hand calculations. Therefore, calculate m.cp.dT for water and air. Then again evaluate how much the air will contribute in terms of Watts.

Now I have calculated that for water approximately there will be used 25 Watts and for air it is 0.1 Watts which is a big difference. Obviously I will say that air will not play a role. Therefor air will not use a significant amount of the energy to be heated. Do I understand it correct now?

[Gert-Jan]

Definitely. I would say you can ignore it.

Remember, this is valid for most cases like you examine now. There are always inaccuracies. e.g. due to the higher temperature the water will expand, leading to a change in heat exchanging area. This will be a minor effect. A more important effect will be the viscosity decrease due to higher temperature. Then mixing due to buoyancy will be enhanced.

You should determine all these effects and determine what can be ignored or not, depending on the question that you want to answer using CFD. Always keep that in mind!

Some time ago, I had a question regarding the shape of meniscus as function of Temperature. Then I needed to include the air and surface tension since the question is completely different. For this I even had to switch to Fluent.

[Knowledge]

Originally Posted by Gert-Jan

Definitely. I would say you can ignore it.

Remember, this is valid for most cases like you examine now. There are always inaccuracies. e.g. due to the higher temperature the water will expand, leading to a change in heat exchanging area. This will be a minor effect. A more important effect will be the viscosity decrease due to higher temperature. Then mixing due to buoyancy will be enhanced.

You should determine all these effects and determine what can be ignored or not, depending on the question that you want to answer using CFD. Always keep that in mind!

Some time ago, I had a question regarding the shape of meniscus as function of Temperature. Then I needed to include the air and surface tension since the question is completely different. For this I even had to switch to Fluent.

But is it impossible to use CFX in my case? If air is negligible what the material of the area then be? Can I leave it empty and just simulate the water and wall?

[Gert-Jan]

No CFX is perfectly fine.
Leave out the air completely. I would consider the air-water-interfaces as an adiabatic wall, with free slip.

[Knowledge]

Originally Posted by Gert-Jan

No CFX is perfectly fine.
Leave out the air completely. I would consider the air-water-interfaces as an adiabatic wall, with free slip.

thanks for helping