Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Problems using electro-thermal interaction with radiative heat transfer
Posted 29 ott 2009, 16:41 GMT-4 3 Replies
Please login with a confirmed email address before reporting spam
Hi everyone. I am starting using Comsol modeling the temperature field of a buried cable installed into a PVC duct (2D problem). I have already solved the problem considering heat transfer by conduction. To do this, I have used the Electro-Thermal Interaction Application Mode (Perpendicular induction heating: Steady-state) included in the AC/DC Module. However, when I try to implement radiation I have this message:
Error: 6253
Complex number found in irradiation expression.
- Expression: J
- Geometry: 1
- Boundaries: 5,6,7,8,13,14,17,18
Failed to evaluate expression.
- Expression: d(((dVolbnd_htgh*epsilon_htgh*(Gm_htgh+Famb_htgh*s igma_htgh*Tamb_htgh^4-sigma_htgh*T^4)*test(T))-(0))*(dvol),{test@0})
I think the problem is that in this application mode the variable T has become a complex number (I don´t know why, but it´s true. When I click on the T surface plot I get this: Value: 341.171896-6.440595e-6i [K], Expression: T, Position: (-0.013275,-0.006)).
Could this behaviour be a program bug? or Should I have to change anything in my model to make variable T a real number again?
I hope someone can help me.
Thank you for your suggestions.
Error: 6253
Complex number found in irradiation expression.
- Expression: J
- Geometry: 1
- Boundaries: 5,6,7,8,13,14,17,18
Failed to evaluate expression.
- Expression: d(((dVolbnd_htgh*epsilon_htgh*(Gm_htgh+Famb_htgh*s igma_htgh*Tamb_htgh^4-sigma_htgh*T^4)*test(T))-(0))*(dvol),{test@0})
I think the problem is that in this application mode the variable T has become a complex number (I don´t know why, but it´s true. When I click on the T surface plot I get this: Value: 341.171896-6.440595e-6i [K], Expression: T, Position: (-0.013275,-0.006)).
Could this behaviour be a program bug? or Should I have to change anything in my model to make variable T a real number again?
I hope someone can help me.
Thank you for your suggestions.
3 Replies Last Post 1 nov 2009, 03:09 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
2 decades ago
Hi
Indeed it sounds strange, these type of questions are typically what I submit to Comsol support.
In the mean time to get around, if you are in 2D and have tubes inside tubes as geometry, probably your view factor is close to 1 so you can anyhow simplify the isue and use a heat exchange just sigma*(T1^4-T2^4) over the boundaries. Anyhow in air radiative heat exchange is rather "soft" and often small compared to conduction or even convection, except if you are cooking or really melting your cable.
Good luck
Ivar
Indeed it sounds strange, these type of questions are typically what I submit to Comsol support.
In the mean time to get around, if you are in 2D and have tubes inside tubes as geometry, probably your view factor is close to 1 so you can anyhow simplify the isue and use a heat exchange just sigma*(T1^4-T2^4) over the boundaries. Anyhow in air radiative heat exchange is rather "soft" and often small compared to conduction or even convection, except if you are cooking or really melting your cable.
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
2 decades ago
Thank you very much for your suggestions. I´ll try to do it, but I have to read more about Comsol to learn how to do it (I´m just staring using it). Anyway, thank you because now I know I´m not the problem and could be a program bug.
Meanwhile I have also tried using surface to ambient radiation instead of surface to surface radiation, as inner surface of the duct and external surface of the cable are almost concentric and the view factor is near 1. Doing this I have no problems, but I am not sure this is correct, so I don´t know if results are valid, because temperature field is exactly the same I obtain when considering only conduction. As cable temperature is around 50ºC, I think radiation has no effect as you told me (I am not melting the cable).
Anyhow, if radiation has low effect, next step is modeling natural convection inside the duct. In this moment I´m fighting to it, because it doesn´t converge.
Thank you again for your help.
Meanwhile I have also tried using surface to ambient radiation instead of surface to surface radiation, as inner surface of the duct and external surface of the cable are almost concentric and the view factor is near 1. Doing this I have no problems, but I am not sure this is correct, so I don´t know if results are valid, because temperature field is exactly the same I obtain when considering only conduction. As cable temperature is around 50ºC, I think radiation has no effect as you told me (I am not melting the cable).
Anyhow, if radiation has low effect, next step is modeling natural convection inside the duct. In this moment I´m fighting to it, because it doesn´t converge.
Thank you again for your help.
Please login with a confirmed email address before reporting spam
Posted:
2 decades ago
Hi
well if you start with COMSOL you will have some reading, I had to read the books several times to feel comfortable with them, because there is really a lot of matter, in fact it's "all Physics" rewritten in COMSOl approach, with their conventions, not always 100% in line with my (now older) training habits, but its all crrect so it's just to learn it their way. But you will get there, and then it's really fun.
In any case, I must strongy suggest that you always verify your model calculations: do a simple thermal calculations for the heat exchange by an anlytical formulation, get your area, the two temperatures and then multiply by sigma the Boltzmann constant (with the correct power of T in Kelvin, and it's the difference of the T^4 in absolute temperaature and not the ^4 of the difference !) this you find in most thermal analysis books, do not forget to check your units.
This gives you the amount of heat exchange (if you are in 2D you must consider areas as length per meter "depth").
Good luck
Ivar
well if you start with COMSOL you will have some reading, I had to read the books several times to feel comfortable with them, because there is really a lot of matter, in fact it's "all Physics" rewritten in COMSOl approach, with their conventions, not always 100% in line with my (now older) training habits, but its all crrect so it's just to learn it their way. But you will get there, and then it's really fun.
In any case, I must strongy suggest that you always verify your model calculations: do a simple thermal calculations for the heat exchange by an anlytical formulation, get your area, the two temperatures and then multiply by sigma the Boltzmann constant (with the correct power of T in Kelvin, and it's the difference of the T^4 in absolute temperaature and not the ^4 of the difference !) this you find in most thermal analysis books, do not forget to check your units.
This gives you the amount of heat exchange (if you are in 2D you must consider areas as length per meter "depth").
Good luck
Ivar