How to set-up temperature of steam inside of vessel

[danijel.radenkovic]

Hello to all,

I have some difficulties to set-up input parameters for my analysis. I want to do an analysis of heat transfer from steam (150°C on the purple surfaces) to the vessel. Outer temperature is 20°C. I would like to see what will be the maximum temperature on the outer surface of the vessel if the temperature of fluid inside is 150°C ? This is just an simplified example.

Any help is very appreciated.

Danijel

[John_Holtz]

Hi Danijel,

One approach is to apply a convection load to the inside surface to represent the steam (I would think a relatively large convection coefficient) and a different convection load to the outside surface to represent the free air (relatively small convection coefficient, depending on whether you want to represent a natural convection state or have some air blowing around the vessel).

You can use the methods in any heat transfer text book to calculate the convection coefficients. The inside surface is probably close to 150C, so you could just use a large value to force the inside temperature to be 150. A large convection coefficient is generally about 1000 times the thermal conductivity of the material.

[danijel.radenkovic]

---

@John_Holtz wrote:

Hi Danijel,

One approach is to apply a convection load to the inside surface to represent the steam (I would think a relatively large convection coefficient) and a different convection load to the outside surface to represent the free air (relatively small convection coefficient, depending on whether you want to represent a natural convection state or have some air blowing around the vessel).

You can use the methods in any heat transfer text book to calculate the convection coefficients. The inside surface is probably close to 150C, so you could just use a large value to force the inside temperature to be 150. A large convection coefficient is generally about 1000 times the thermal conductivity of the material.

---

Hello Mr. Holtz,

Thank you! I have done some simulation and I am comparing and trying to understand results. Is it possible somehow to see the stress results for some specific node on the mesh?

Danijel 

[John_Holtz]

The answer is yes.

A thermal analysis calculates the temperatures and heat fluxes.

A linear static stress analysis calculates the displacements and stresses.

So you now need to perform a linear static stress analysis and use the temperatures from the thermal analysis as a load. See any of the following links for background information:

• "Thermal Stress of a Tank" in the Help > User's Guide > Examples > Linear and Dynamic Stress gives a step-by-step tutorial.
• "Temperatures" in the Help > User's Guide > Setup and Perform Analysis > General Information > Loads and Constraints" describes the basic steps of applying temperature loads to a stress analysis. For your case, scroll down to the section Apply Temperature Profiles from Thermal Analyses.
• "How to Perform a Thermal Stress Analysis" article on the Knowledge Network gives the steps for transferring temperatures from a steady state heat transfer analysis to a linear stress analysis. (I see that this article is slightly outdated, but close enough to the current software that you should be able to follow it.)

Have fun!

[danijel.radenkovic]

---

@John_Holtz wrote:

The answer is yes.

A thermal analysis calculates the temperatures and heat fluxes.

A linear static stress analysis calculates the displacements and stresses.

So you now need to perform a linear static stress analysis and use the temperatures from the thermal analysis as a load. See any of the following links for background information:

• "Thermal Stress of a Tank" in the Help > User's Guide > Examples > Linear and Dynamic Stress gives a step-by-step tutorial.
• "Temperatures" in the Help > User's Guide > Setup and Perform Analysis > General Information > Loads and Constraints" describes the basic steps of applying temperature loads to a stress analysis. For your case, scroll down to the section Apply Temperature Profiles from Thermal Analyses.
• "How to Perform a Thermal Stress Analysis" article on the Knowledge Network gives the steps for transferring temperatures from a steady state heat transfer analysis to a linear stress analysis. (I see that this article is slightly outdated, but close enough to the current software that you should be able to follow it.)

Have fun!

---

Hello Mr. Holtz,

Thank you very much for helping me.

I have done it. I will post the video of my workflow cause I would like to hear your opinion.

[danijel.radenkovic]

---

@John_Holtz wrote:

Hi Danijel,

One approach is to apply a convection load to the inside surface to represent the steam (I would think a relatively large convection coefficient) and a different convection load to the outside surface to represent the free air (relatively small convection coefficient, depending on whether you want to represent a natural convection state or have some air blowing around the vessel).

You can use the methods in any heat transfer text book to calculate the convection coefficients. The inside surface is probably close to 150C, so you could just use a large value to force the inside temperature to be 150. A large convection coefficient is generally about 1000 times the thermal conductivity of the material.

---

Hello Mr. Holtz,

Regarding to your advice, I have one more question: I have tried to find convection coefficient on the internet and I have found it only for the room temperature. What would be convection coefficient for the convection load on inside surface if the temperature of the inner surface is about 150° or 210°? What convection coefficients do you suggest please?

[John_Holtz · Accepted solution]

Hello,

I think your situation is too specialized to find convection coefficients published anywhere. And it is not something that I can easily calculate because of the number of parameters involved.

Have you tried three analyses where you used the convection value you did find, something 100 times larger, and something 1E6 times larger? This is just a guess, but the steam probably has a much larger capacity to provide heat than the air has to remove it. So I would expect there to be just a small difference in the temperature results in the three analysis. If that is the case, it may not be necessary to have a "precise" value for the convection coefficient.

If you want a reasonable estimate, my suggest is to get a textbook on Heat Transfer. They generally have several chapters related to the calculation of the convection coefficient depending on arrangement (horizontal, vertical, flat, round, and so on), motion (still gas/fluid, moving gas/fluid), and the book may even have the material properties that you need for the fluid (thermal conductivity, density, Prandtl number, and so on).

[danijel.radenkovic]

---

@John_Holtz wrote:

Hello,

I think your situation is too specialized to find convection coefficients published anywhere. And it is not something that I can easily calculate because of the number of parameters involved.

Have you tried three analyses where you used the convection value you did find, something 100 times larger, and something 1E6 times larger? This is just a guess, but the steam probably has a much larger capacity to provide heat than the air has to remove it. So I would expect there to be just a small difference in the temperature results in the three analysis. If that is the case, it may not be necessary to have a "precise" value for the convection coefficient.

If you want a reasonable estimate, my suggest is to get a textbook on Heat Transfer. They generally have several chapters related to the calculation of the convection coefficient depending on arrangement (horizontal, vertical, flat, round, and so on), motion (still gas/fluid, moving gas/fluid), and the book may even have the material properties that you need for the fluid (thermal conductivity, density, Prandtl number, and so on).

---

Hello Mr. Holtz,

That's is the reason why I have asked for the coefficient, I am getting in both of the cases pretty similar temperature results.

Here are the results:

Thank you very much for your help! For now, I have got the results which are acceptable for static analysis, but I decided to learn much more about thermal analysis. Thank you for your time!

Danijel

[John_Holtz]

50 W/mm^2/C is a huge convection coefficient. Do you have several jet engines blowing air around the outside of the vessel?

Your large convection coefficients are forcing the outside temperature of the vessel to be the same as the air temperature (or whatever fluid is outside the vessel). A typical convection coefficient for air is on the order of 4 to 15 W/m^2/C (4E-6 to 15E-6 W/mm^2/C) depending on the orientation and speed of the air.

[danijel.radenkovic]

---

@John_Holtz wrote:

50 W/mm^2/C is a huge convection coefficient. Do you have several jet engines blowing air around the outside of the vessel?

Your large convection coefficients are forcing the outside temperature of the vessel to be the same as the air temperature (or whatever fluid is outside the vessel). A typical convection coefficient for air is on the order of 4 to 15 W/m^2/C (4E-6 to 15E-6 W/mm^2/C) depending on the orientation and speed of the air.

---

Hello Mr. Holtz,

I made mistake, but even when I set the outside convection coefficient to 5 W/mm^2/°C there is no big difference between results. As I said in the last reply, I have to learn much more about heat transferring.

Thank you!

Danijel