Robert Koslover
Certified Consultant
Please login with a confirmed email address before reporting spam
Posted:
4 years ago
4 janv. 2021, 16:42 UTC−5
Well, if I was modeling the transfer of heat from a heated wire to some environment around it, I probably wouldn't try to constrain the temperature on the wire surface, since I doubt that I would know its exact surface temperature distribution in advance. Rather, I would model the sources of heat (ohmic losses due to an electric current, perhaps?) in the wire and account for the thermal properties of the wire (thermal conductivity and specific heat) and surrounding environment. I'd probably constrain the outermost computational boundary to some fixed temperature (like a room temperature), although other boundary conditions might possibly apply there. In a time-domain model, I would define the temperature everywhere, starting at time t=0, and define and apply a time-domain excitation (source of heat), and let the problem advance in time for small enough time steps to model the details adequately, and for as long a time as I needed for my application. In a steady-state (equilibrium) model, I wouldn't need initial volume temperature conditions, just all the right boundary conditions, and a source of heat. Ok, so if those comments don't help you figure out what to do, I suggest you post your model to the forum. You might also want to study some of the thermal models already provided by Comsol, in the Applications Library. Good luck.
-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Well, if I was modeling the transfer of heat from a heated wire to some environment around it, I probably wouldn't try to constrain the temperature on the wire surface, since I doubt that I would know its exact surface temperature distribution in advance. Rather, I would model the sources of heat (ohmic losses due to an electric current, perhaps?) in the wire and account for the thermal properties of the wire (thermal conductivity and specific heat) and surrounding environment. I'd probably constrain the outermost computational boundary to some fixed temperature (like a room temperature), although other boundary conditions might possibly apply there. In a time-domain model, I would define the temperature everywhere, starting at time t=0, and define and apply a time-domain excitation (source of heat), and let the problem advance in time for small enough time steps to model the details adequately, and for as long a time as I needed for my application. In a steady-state (equilibrium) model, I wouldn't need initial volume temperature conditions, just all the right boundary conditions, and a source of heat. Ok, so if those comments don't help you figure out what to do, I suggest you post your model to the forum. You might also want to study some of the thermal models already provided by Comsol, in the Applications Library. Good luck.