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.

Comsol 3-D Transient Heat Transfer Conduction Module

Please login with a confirmed email address before reporting spam

I'm working on a project to model a connecting rod of an internal combustion engine and running into ridiculous data output. I am taking the rod to be at melting temperature and trying to cool it to room temperature.

I have a very simple CAD model of the rod and imported as an STL file.

The sub domain boundary settings were based off of the material library for AISI 4340 Steel.
Q = 0 heat source
d_ts = 1

For cp, I am basing it off the following enthalpy equation
H= C*(T-T_0)+L*H(T-T_m)

where C = 475 J/(kg*K)
T_0=1800 K initial temp
T_m = 1500 melting temp
H(T-T_m)= heaviside step function
L=226000 J/kg latent heat

In order to approximate and smooth the heaviside function, it is replaced with tanh function
so

H= C*(T-T_0)+L*[1/2*tanh( (T-T_m)/delta(T_m) )]
delta(T_m) is a smoothing parameter ~ 10 degrees
T - this is the variable, so I set the intial T(t_0) = 1800K from the initial tab

taking the derivative c_p=dH/dT = C+ L/2 * (1-tanh( (T-T_m)/ delta(T_m)) / delta(T_m)

For the boundary condition, all around the rod there is convective . I set the constant to 0 for no radiation.
T_inf = 300k
I have tried h values from 10-2000 J/kg*K

So after running the simulation, it takes a long time for the model to cool to room temperature. Example would be setting the time to 1*10^7 seconds. The exterior of the rod would be at room temperature, but a slice plot shows certain internal areas to be still at around 1800-2000K. This occurs around the areas where the two large holes are for a connecting rod. Even at lower time intervals, such as 1 hour, the max temperature actually rose higher.

As a test, I ran a simulation for a simple 3-D cube with comsol library material properties for AISI 4340 with constant properties. The plot still shows that the internal temperature is much higher the exterior after running with a large time interval. The cooling time just does not make sense.

Some advice that was given to me was to impose q for the boundary conditions.

As an example of what I am working with, I have attached a temperature distribution of the rod with a time scaling coefficient of 1/60. Note the the time at the top of the graph and temperature scale on the right. I don't have access to comsol at the moment, so I can't get a sliced view.


1 Reply Last Post 16 mai 2010, 17:53 UTC−4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 16 mai 2010, 17:53 UTC−4
Not in front of comsol at the moment, but I'm pretty sure I figured out the problem. UNITS

I modeled the rod in solidworks CAD program and exported the file to an stl. I exported the file in milimeters and comsol is defaulted to meters. Therefore my model had a scale that was up to 216 m long. That explains why it was taking so long to cool.
Not in front of comsol at the moment, but I'm pretty sure I figured out the problem. UNITS I modeled the rod in solidworks CAD program and exported the file to an stl. I exported the file in milimeters and comsol is defaulted to meters. Therefore my model had a scale that was up to 216 m long. That explains why it was taking so long to cool.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.