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.

please make comment and help for my problem(especially Ivar)

Please login with a confirmed email address before reporting spam

Hi

I am doing a heat conduction analysis in 3D. what I need to do is to move or slide a hot box over another cold box from head to tail. so as the hot box(upper one) is going over the bottom one there should be a time dependent temperature contour in in the bottom one. I have tried deformed geometry and moving mesh (where I see no temperature sense in the bottom box as the upper one moves forward after one step) and also although I do not want to use parametric sweep it is again does not answer to my problem. please take a look at the attached picture and see how the temperature contour is not correctly distributed as the upper box slides :((

Please help me with your comments. I am worried if is this thing doable in COMSOL?

thanks


10 Replies Last Post 05.08.2011, 11:40 GMT-4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.03.2011, 07:06 GMT-4
Hi

could it be linked to the reply from here ?

www.comsol.eu/community/forums/general/thread/15851/#p43553

--
Good luck
Ivar
Hi could it be linked to the reply from here ? http://www.comsol.eu/community/forums/general/thread/15851/#p43553 -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.03.2011, 10:52 GMT-4
Hi Ivar

I did what was said but did not answered :(

all problems are in 2D but my problem is in 3D does it makes changes? for example one which is odd is

in ALE I can define just free deformation, when I define prescribed mesh velocity or mesh displacement there is no object selectable and all surfaces became not applicable!!

could you say where is my problem please,
could you take a look at the file pleasee.

Navid
Hi Ivar I did what was said but did not answered :( all problems are in 2D but my problem is in 3D does it makes changes? for example one which is odd is in ALE I can define just free deformation, when I define prescribed mesh velocity or mesh displacement there is no object selectable and all surfaces became not applicable!! could you say where is my problem please, could you take a look at the file pleasee. Navid


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 25.03.2011, 11:00 GMT-4
Ivar

could you please take a look at the file in previous message and see what is my problem please :( I am stuck in this simple problem for a long long time :( and it makes me so confused. I just want to move a block over the other and get a rational temperature contour in the bottom block :( in 3D

I appreciate it,

Ivar could you please take a look at the file in previous message and see what is my problem please :( I am stuck in this simple problem for a long long time :( and it makes me so confused. I just want to move a block over the other and get a rational temperature contour in the bottom block :( in 3D I appreciate it,

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 26.03.2011, 07:29 GMT-4
Hi

now I do not have 4.0 so my 2D HT+ALE model attached a 4.1.0.185 is probably not useful for you.

What I did was to use ALE instead of deformed geometry (see the KB), with assembly mode and an identity continuity condition for the contact surface

Then there is an open discussion about the frame to use, I believe the spatial (default) one is the best, in my model the results are wrong if I select one of the others, but on another thread, there was something wrong in the case of a rotary heat exchange in the same physics, but it could be something else too, such models should ideally be remade from scratch time to time, to be sure that all tries and changes has not changed a setting a couple of levels down

--
Good luck
Ivar
Hi now I do not have 4.0 so my 2D HT+ALE model attached a 4.1.0.185 is probably not useful for you. What I did was to use ALE instead of deformed geometry (see the KB), with assembly mode and an identity continuity condition for the contact surface Then there is an open discussion about the frame to use, I believe the spatial (default) one is the best, in my model the results are wrong if I select one of the others, but on another thread, there was something wrong in the case of a rotary heat exchange in the same physics, but it could be something else too, such models should ideally be remade from scratch time to time, to be sure that all tries and changes has not changed a setting a couple of levels down -- Good luck Ivar


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28.03.2011, 11:14 GMT-4
Hi Ivar

I am very thankful for your model. I tried to open your file in version 4.1 but it said "expected end of central directory record signature" ! is it kind of zipped or something. I think this would be really helpful fro me because I think I have problem with how to do the 'assembly mode and an identity continuity condition for the contact surface" because when you do the 3D simulation the contact surface is just one surface and i do not know how continuity works if it should be between two surfaces how it would work when we have just one surface between top and bottom blocks!

I appreciate your helps, I feel it is gonna to work by your help.

Thanks
Hi Ivar I am very thankful for your model. I tried to open your file in version 4.1 but it said "expected end of central directory record signature" ! is it kind of zipped or something. I think this would be really helpful fro me because I think I have problem with how to do the 'assembly mode and an identity continuity condition for the contact surface" because when you do the 3D simulation the contact surface is just one surface and i do not know how continuity works if it should be between two surfaces how it would work when we have just one surface between top and bottom blocks! I appreciate your helps, I feel it is gonna to work by your help. Thanks

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28.03.2011, 13:44 GMT-4
Hi Ivar

I wanted to say I can not use HT in solids. in my problem I have a vertical cylindrical hole in the bottom block say in location x and is filled with another material something like discontinuity in bottom box. so if the heat wants to transfer in bottom box from right to left then it will not be same situation for the temperature contours after the hole.

and also in the contact surface in bottom block we should have the same temperature as upper box but as you see in figure, I have defined motion in x direction 0.0025 m/s and the length is 5 cm so when time is 20 sec all the top surface should be somehow red but it is still blue :(

and the main problem is th temperature distribution after the hole as you see is not simulating real contours as hot box moves.

please take a look at pic and file. thank you so much Ivar
Hi Ivar I wanted to say I can not use HT in solids. in my problem I have a vertical cylindrical hole in the bottom block say in location x and is filled with another material something like discontinuity in bottom box. so if the heat wants to transfer in bottom box from right to left then it will not be same situation for the temperature contours after the hole. and also in the contact surface in bottom block we should have the same temperature as upper box but as you see in figure, I have defined motion in x direction 0.0025 m/s and the length is 5 cm so when time is 20 sec all the top surface should be somehow red but it is still blue :( and the main problem is th temperature distribution after the hole as you see is not simulating real contours as hot box moves. please take a look at pic and file. thank you so much Ivar


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28.03.2011, 14:23 GMT-4
Hi Ivar

I could open your file, that is definitely what I want thank you :), I am going over it to see how you did the simulation. actually my main problem was how to define the pairs. when I do the 3D simulation the contact surface counts as one common surface and it is not like on surface from the top block and one from the bottom to have a common surface.

anyhow thank you sooo much :)

I am going over it please let me know if you have any comment on how to have COMSOL to not combine surfaces and let me to select the surfaces separately.

Thanks again
Hi Ivar I could open your file, that is definitely what I want thank you :), I am going over it to see how you did the simulation. actually my main problem was how to define the pairs. when I do the 3D simulation the contact surface counts as one common surface and it is not like on surface from the top block and one from the bottom to have a common surface. anyhow thank you sooo much :) I am going over it please let me know if you have any comment on how to have COMSOL to not combine surfaces and let me to select the surfaces separately. Thanks again

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 28.03.2011, 15:26 GMT-4
I haven't read the whole thread, only your initial question. So this may have been already suggested:
If the deformation in question is strictly rigid body motion or rotation, easiest way to do what you want is to take advantage of convective terms in the heat equation. For example, you can simulate the effect of body translation by simply assigning it a velocity in the translation direction. I haven't thought about how you would model the ends of the box domains though.
good luck
Ozgur
I haven't read the whole thread, only your initial question. So this may have been already suggested: If the deformation in question is strictly rigid body motion or rotation, easiest way to do what you want is to take advantage of convective terms in the heat equation. For example, you can simulate the effect of body translation by simply assigning it a velocity in the translation direction. I haven't thought about how you would model the ends of the box domains though. good luck Ozgur

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 29.03.2011, 00:36 GMT-4
Hi

of you have only 1 boundary, it is because you are in "Finish Union mode" this implies unicity on all boundaries, you should use "Assembly Finsih mode inthe geoemtry, and for a simple case "without" imprints. Indeed my model is v4.1.0.185, so you must have that version to load it, I do no longer have the older version up running

--
Good luck
Ivar
Hi of you have only 1 boundary, it is because you are in "Finish Union mode" this implies unicity on all boundaries, you should use "Assembly Finsih mode inthe geoemtry, and for a simple case "without" imprints. Indeed my model is v4.1.0.185, so you must have that version to load it, I do no longer have the older version up running -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 05.08.2011, 11:40 GMT-4
I would like to add a supplementary question to this thread. I don't have a picture of what I'm going to describe, but hopefully, it's clear enough...Extending from Navid's model, what if I wanted to have a protrusion from the top block into the bottom block and have it traverse through the material of the bottom block (such as if the bottom domain was a liquid and the top solid). After each time increment, the mesh would need to update to accommodate the new topology...or maybe not a new topology since the connectivity isn't necessarily changing (or is it?). 1) Is this possible strictly with the GUI (i.e. no Matlab Livelink or other programming required) and, if so, 2) How would I implement it in a similar fashion to what Ivar has done in this example?

Thank you.
I would like to add a supplementary question to this thread. I don't have a picture of what I'm going to describe, but hopefully, it's clear enough...Extending from Navid's model, what if I wanted to have a protrusion from the top block into the bottom block and have it traverse through the material of the bottom block (such as if the bottom domain was a liquid and the top solid). After each time increment, the mesh would need to update to accommodate the new topology...or maybe not a new topology since the connectivity isn't necessarily changing (or is it?). 1) Is this possible strictly with the GUI (i.e. no Matlab Livelink or other programming required) and, if so, 2) How would I implement it in a similar fashion to what Ivar has done in this example? Thank you.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.