MICRESS Forum

Hi Micress forum,

I'm trying to model directional solidification in 1D. I'm hoping to run the simulation with a set thermal gradient and cooling rate, then after a set time, stop the simulation and restart it with no cooling rate. How do you set up a 1D simulation in MICRESS? So far i've tried to set the thermal conditions to 1d_temp but it gets rid of the option to set a cooling rate in process conditions, which i need control of during my simulations. I'm also slightly confused as what to set as the initial position of the 1D temperature field when i use the 1d_temp option for thermal conditions. How does the temperature field in 1D work in MICRESS? Any help would be much appreciated.

Many thanks,

Matthew

Hi Matthew,

with the '1D_temp' option, you lay an extra 1D temperature field over the simulation domain in Z-direction for solving temperature. It can be larger than the simulation domain.

I am not sure that this can be combined with 1D simulation area, i.e. in MICRESS setting the geometry in x and y to 1. Like it is done in the Benchmark example B016_diffusion_control.

Best,
Ralph

Dear Matthew,

The question of using a 1d-simulation domain in MICRESS is completely independent from using a 1d-temp boundary condition or not. The 1d-temp option can be used if you want to calculate the temperature boundary condition for the simulation domain in 1d-approximation using a 1d-temperature field. If you want to set the temperature boundary condition instead (e.g. as a constant cooling rate), you should not use the 1d-temp option.

Bernd

Apologies I'm very confused. So how would you recommend setting up a simulation in 1D. And is it possible to set up a simulation in 1D whereby you can change the cooling rate?

Hi Matthew,

To set up a 1d-simulation in MICRESS (one-dimensional microstructure simulation domain) you simply set the size of the simulation domain to 1 for the x- and y-direction, and in z-direction to the size of your choice. This so far is independent on your choices for the thermal boundary conditions which you define afterwards.

Then, for the thermal boundary conditions, you have 3 fundamental choices (independent of the dimensionality you have chosen for the simulation domain above):

1.) You set the temperature (and its change with time and along the z-direction, optionally with some inclination). In the most simple case you will just define a constant temperature rate and temperature gradient. You do that by specifying "no_latent_heat" in the "Model" section of the input file.

2.) You calculate the temperature and its change with time by specifying a heat extraction rate (global heat balance). This includes the effects of latent heat. In the simplest case you will define a constant heat extraction rate or a heat transfer coefficient together with an environment temperature). You could combine this further with a prescribed temperature gradient in z-direction, although this typically does not make much sense.

3.) You calculated the temperature and its change with time and space by using a more complex macroscopic 1d-temperature field approach. This takes latent heat within the simulation domain into account. Latent heat within the 1d-temperature field (outside the microstructure simulation domain) is solved indirectly using tabulated data which may be iteratively obtained (Homoenthalpic Approach). As you already experienced, this requires a more extensive input of the parameters required to solve the 1d-temperature problem.

Which of the three approaches is best for you depends on your requirements. From what you had written in your first question above, probably approach 1.) is fine for you, and you just define a constant cooling rate and temperature gradient. Method 3) would only apply if you have transient conditions (like L-PBF or welding) or a small temperature gradient, so that latent heat effects must be taken into account.

Bernd