Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
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.
Cooling chamber
Posted Mar 10, 2011, 8:24 a.m. EST Version 3.5a, Version 4.2a 2 Replies
Please login with a confirmed email address before reporting spam
Hi,
I am doing a work placement at Roctool, a french company working on induction heating of industrial molds for thermoplastic injection.
I am currently using the COMSOL 3.5a version to simulate the joule effect induced by a current wandering through a copper wire.
I am running a 2D simulation, using the "conduction-transient analysis" (heat transfer) and "perpendicular induction currents-time harmonic analysis"(quasi-statics,magnetic).
By applying an external current density and setting the resistive heating (Qav_emqa*(t<X)) as a heat source in the conduction subdomain settings, the simulation is working just fine.
The results would be quite realistic if it wasn't for the cooling system. For the moment, I've only been able to apply a constant heat transfer coefficient (around 10000 W/(m^2.K), which is an approximation of the heat transfer coefficient of water for turbulent flow) on the boundary of the cooling channel.
I do obtain results, but in order to improve them I wish to switch to a cooling chamber model, located behind the heating system. The idea is to absorb the heat provided by induction using a change of phase of the working fluid (using the latent heat to have a much more efficient cooling).
I'm still quite new to Comsol, therefore I would like to now if you happen to have preexisting research on this case or if similar models that are available (I've been searching on the website, without success). The problem is the company doesn't have all the application modes, only Comsol multiphysics, ac/dc module, heat transfer module and structural mechanics module. On the few similar cases i've found, it seems the chemical engineering module is quite useful to caracterize the change of phase, and input a latent heat value. Is it possible to do without?
So far the simulation hasn't been too complex, but I would like to add a velocity field for the fluid, linked with its change of density and dynamic viscosity. I intend to keep the ambient pressure of the cooling chamber constant, since the liquid phase and vapor phase remain at an equilibrium by using a heat exchanger. My internship supervisor has told me that the point isn't to start from scratch, but rather to find preexisting similar subjects and simulations to help.
If anyone of experience would be ever so kind as to provide information on how to handle the problem, I would be very grateful.
It seems the moderator mr.Kjelberg, has answers to everything ;)
I am doing a work placement at Roctool, a french company working on induction heating of industrial molds for thermoplastic injection.
I am currently using the COMSOL 3.5a version to simulate the joule effect induced by a current wandering through a copper wire.
I am running a 2D simulation, using the "conduction-transient analysis" (heat transfer) and "perpendicular induction currents-time harmonic analysis"(quasi-statics,magnetic).
By applying an external current density and setting the resistive heating (Qav_emqa*(t<X)) as a heat source in the conduction subdomain settings, the simulation is working just fine.
The results would be quite realistic if it wasn't for the cooling system. For the moment, I've only been able to apply a constant heat transfer coefficient (around 10000 W/(m^2.K), which is an approximation of the heat transfer coefficient of water for turbulent flow) on the boundary of the cooling channel.
I do obtain results, but in order to improve them I wish to switch to a cooling chamber model, located behind the heating system. The idea is to absorb the heat provided by induction using a change of phase of the working fluid (using the latent heat to have a much more efficient cooling).
I'm still quite new to Comsol, therefore I would like to now if you happen to have preexisting research on this case or if similar models that are available (I've been searching on the website, without success). The problem is the company doesn't have all the application modes, only Comsol multiphysics, ac/dc module, heat transfer module and structural mechanics module. On the few similar cases i've found, it seems the chemical engineering module is quite useful to caracterize the change of phase, and input a latent heat value. Is it possible to do without?
So far the simulation hasn't been too complex, but I would like to add a velocity field for the fluid, linked with its change of density and dynamic viscosity. I intend to keep the ambient pressure of the cooling chamber constant, since the liquid phase and vapor phase remain at an equilibrium by using a heat exchanger. My internship supervisor has told me that the point isn't to start from scratch, but rather to find preexisting similar subjects and simulations to help.
If anyone of experience would be ever so kind as to provide information on how to handle the problem, I would be very grateful.
It seems the moderator mr.Kjelberg, has answers to everything ;)
2 Replies Last Post Jan 9, 2012, 7:58 a.m. EST
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
Well in fact I do not have answers to everything, and the questions you are asking there are somewhat outside my usual field. Phase change is an interesting domain but there I consider myself only as an amateur.
One thing before setting up a full CFD and phase change model you could trey to find some intermediate formula driven physics to represent your non-linear exchange with better representativity than a plain heat sink
Normally, you can reconstruct everything by yourself with the basic COMSOL module, but it can be rather tedious. For the modules, you should take note that with the v4 (r)evolution the modules have changed from applications to physics. So you must discuss in detail with your local rep which modules you really need, as these have changed now in the new version.
It will definitively be a plus to a) use v4 far more efficient to build models, and b) to gain time by having most physics pre-coocked by using the existing modules replying to your need and domain of application(s).
In v4 you also have a new feature (not yet official, it was presented at the COMSOL conference in Paris last year) the Model Builder that should allow you to build your own "physics" settings, so that you do not need to fill in all the formulas each time
--
Good luck
Ivar
Well in fact I do not have answers to everything, and the questions you are asking there are somewhat outside my usual field. Phase change is an interesting domain but there I consider myself only as an amateur.
One thing before setting up a full CFD and phase change model you could trey to find some intermediate formula driven physics to represent your non-linear exchange with better representativity than a plain heat sink
Normally, you can reconstruct everything by yourself with the basic COMSOL module, but it can be rather tedious. For the modules, you should take note that with the v4 (r)evolution the modules have changed from applications to physics. So you must discuss in detail with your local rep which modules you really need, as these have changed now in the new version.
It will definitively be a plus to a) use v4 far more efficient to build models, and b) to gain time by having most physics pre-coocked by using the existing modules replying to your need and domain of application(s).
In v4 you also have a new feature (not yet official, it was presented at the COMSOL conference in Paris last year) the Model Builder that should allow you to build your own "physics" settings, so that you do not need to fill in all the formulas each time
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
I do not have 3.5a up running any more, but here is an example in 4.2a for tweaking Cp to get a latent heat equivalent (if I haunt got it wrong though pls check it out carefully).
The real trick is that one must help the solver to see the short Cp spikes, if you try to solve the model with Study 2 (same as Study 1 but in default time series free mode) the result is quite different (and quite wrong) !
--
Good luck
Ivar
I do not have 3.5a up running any more, but here is an example in 4.2a for tweaking Cp to get a latent heat equivalent (if I haunt got it wrong though pls check it out carefully).
The real trick is that one must help the solver to see the short Cp spikes, if you try to solve the model with Study 2 (same as Study 1 but in default time series free mode) the result is quite different (and quite wrong) !
--
Good luck
Ivar
Attachments: