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.
From 3D input data to a 2D axial-symmetric simulation
Posted Nov 24, 2010, 12:49 p.m. EST 1 Reply
Please login with a confirmed email address before reporting spam
Hello there,
I have a question. I have a volumetric flow rate waveform that is expressed in m^3/s. So this would be ok for a 3D inlet let's say.
Suppose know that I want to do a 2D axial-symmetric simulation. The model in this case represents 1 radian? Meaning that I can divide by 2*pi my input data? Or COMSOL has another approach? For example ANSYS use a 30 degree angle.
Thanks in advance
J
I have a question. I have a volumetric flow rate waveform that is expressed in m^3/s. So this would be ok for a 3D inlet let's say.
Suppose know that I want to do a 2D axial-symmetric simulation. The model in this case represents 1 radian? Meaning that I can divide by 2*pi my input data? Or COMSOL has another approach? For example ANSYS use a 30 degree angle.
Thanks in advance
J
1 Reply Last Post Nov 24, 2010, 3:54 p.m. EST
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
Comsol always calculates in 3D, when we select a 2D or a 2D axi or even a 1D one or two spatial parametres are normalised to 1[m] resp. [m^2] (slightly physics dependent check the min physics depth or area parameter) so the equations simplify. Nevertheless, comsol always proposes GUI fields entries at "finite element" level, so body forces are N/m^3, and surface forces in Pa=N/m^2, This means that we need to implicitely integrate over a tripple integral *dxdydz. But in 2D as Z=1m we do only a double integral multiplied by the default depth i.e. "solid.d" for ST ..., respectively "beam.area* for structural beam model. Therefore I like to see the units, it reminds me bout how to normalise, thi is valid for all physics
For 2D -axi the normalisation is 2*pi*r, the loop length, but between 3.5a and v4 COSMOL has decided to change their normalsiation, so I beleive, pls check carefully, that now in v4+ the 2*pi is already incuded (but not the "r")
--
Good luck
Ivar
Comsol always calculates in 3D, when we select a 2D or a 2D axi or even a 1D one or two spatial parametres are normalised to 1[m] resp. [m^2] (slightly physics dependent check the min physics depth or area parameter) so the equations simplify. Nevertheless, comsol always proposes GUI fields entries at "finite element" level, so body forces are N/m^3, and surface forces in Pa=N/m^2, This means that we need to implicitely integrate over a tripple integral *dxdydz. But in 2D as Z=1m we do only a double integral multiplied by the default depth i.e. "solid.d" for ST ..., respectively "beam.area* for structural beam model. Therefore I like to see the units, it reminds me bout how to normalise, thi is valid for all physics
For 2D -axi the normalisation is 2*pi*r, the loop length, but between 3.5a and v4 COSMOL has decided to change their normalsiation, so I beleive, pls check carefully, that now in v4+ the 2*pi is already incuded (but not the "r")
--
Good luck
Ivar