1d_temp

[Bernd]

Dear sunny,

if you want to setup a simulation like e.g. the one illustrated by the "AlCu_Temp1d_dri" example, of course, you need physical parameters like the heat diffusivity and the thermal boundary conditions which correspond to your process.
Furthermore, especially if phase transformations are supposed to take place also outside the microscopic simulation domain (i.e. in the outside part of the 1d temperature field), you need data on the effective enthalpy and heat capacity as a function of temperature. One possible approach is to use MICRESS for getting these data in a prior simulation run. If you perform a simulation with 1d_temp, or without 1d_temp but with latent heat, you get the average enthalpy and cp (and eventually heat diffusivity) data as a function of temperature in the .dTLat output. This output can directly be used as input for a subsequent MICRESS simulation with 1d temperature field, as illustrated in the "AlCu_Temp1d" example. The "time dependency" you mention in the tabular data is nothing more than a remnant of the prior simulation, because the complete .dTLat file has been used here.

The important question is now, how the first simulation has to be set up to obtain a useful dataset for the subsequent second simulation. The answer is of course that there is no general rule and that it depends a lot on the conditions. Sometimes, an iterative approach seems the best solution, like we have demonstrated recently for equiaxed castings ("homoenthalpic approach"). For directional solidification or solid state processes, things may be much easier.

In the present implementation, averaging of the latent heat data (.dTLat) in the case of 1d_temp coupling is done on the whole microstructure domain if the domain is quadratic (Nx=Nz), and for a central part of the domain which corresponds to the width of one cell of the temperature field in z direction, otherwise. We may change this if more control over the averaging zone is needed in future.

I have put the "AlCu_Equiaxed_dri" example for you on the temporary download area of the micress homepage!

Bernd

[Bernd]

What is 1d_temp option and in which part of the driving file it appears?
In a calculation domain, how nonlinear temperature gradient could be defined by “1d_temp” option?
Which data each column of the file containing one-dimensional temperature field represents?

Dear mtoloui,

I just implemented a generalized input of temperature profiles for different time steps ("Connection Points") from file (option "profiles_from_file"). Thus, it will be no longer necessary to use the 1d_temp option for this purpose! The update will be available with the next release. I can send send you an updated beta-version soon.

Bernd