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.
Boundary conditions problems in transient heat transfer
Posted 2 feb 2018, 10:07 GMT-5 Heat Transfer & Phase Change, Modeling Tools & Definitions, Parameters, Variables, & Functions 2 Replies
Please login with a confirmed email address before reporting spam
Hello,
I tried to make a time dependent 3D heat transfer to the ground simulation for a period of one year. Boundary conditions: h = 24 W/m2K, Te = linear interpolation function, defined at Global Definitions section based on hourly registered real temperatures (introduced as heat flux boundary condition - Robin?..). T ground = 9 deg C (introduced as temperature boundary condition - Dirichlet).
I made 3 different simulations starting with the same steady-state initialization, at the time-step with different simulation periods: 24 h (calculation step 0.2), 760 h, 8760 h (one year). I analised the boundary temperatures (Tsi), heat fluxes during the first 24 h for each simulation and the values were different for the three simulations. Why do I not get the same results?
Is there a limit for the number of input values used for the interpolation function, which may have an effect on the imposed boundary conditions on different time steps?
For the heat flux there are different values for inward heat flux and the normal total heat flux (extracted from Derived Values -> Surface Integration), for the same simulation. Can anyone explain the meaning of the two terminologies?
Kind regards, Ancuta Magurean
Hello Ancuta Magurean
Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.
If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.
Please login with a confirmed email address before reporting spam
Hi.
There is a thread somewhere in the forum explaining this. The fact that you are getting different result has to do with each solver timestep. I think that the timestep is chosen to be something like 0.001(final_time). (not sure) As so, your results may differ if the problem is not well defined, the mesh fine enough, and some other factors. Check the log file for each simulation and see the timesteps taken by the solver.
Regarding you 2nd question: normal total heat flux means the flux that is perpendicular to the surface. I suppose that "inward heat flux" is the flux that goes towards the domain and passes through that surface but the incident angles is not necessarily 90º.
Hope this helps.
Best, Nichal
Please login with a confirmed email address before reporting spam
Hello Nichal,
Thank you for your answer!
I asked for technical support and I found what was the issue. Also, I found a really useful discussion on the forum. Anyway, for those who will have the same problem, I will let the "solution" here.
So it was a problem off time step call in the solver. The time step from the Time dependent Study Settings (Times), actually is used by the program only for the output data. The program make it's own steps of simulation, but this time steps settings are hidden in the Time Dependent Solver menu -> Time stepping -> Steps taken by solver, and by default they are set as "FREE", which make sense why the results were different for different length of the period of the simulation for the same model in my test simulation. In order to force the time step that you want for the solver to solve related to the entry known variables, "Strict" option can be used.
Best regards, Ancuța Măgurean