| Author | Message | ||
     Rizwaan (ritzyritz) Junior Member Username: ritzyritz Post Number: 3 Registered: 02-2007 |
Hello Everybody, I have been trying to solve the problem I am going to mention here on my own for last few days, but in vain. Only as a last resort I am posting it here in this forum. Here it goes: The problem is slightly complicated. It comprises of two domains - a 3D and 2D, simultaneously. There are two separate sets of PDEs for each domain. Both the systems of PDEs can be easily solved for their respective domains in FLEXPDE separately. Complication arises when the 2D domain is attached to the 3D domain on one of its surfaces. The 3D domain is a simple 3D plate fixed on one end and loaded on another. The rest of the four lateral faces are stress free. On one of these four lateral stress free surfaces, a 2D thin film is bonded to the 3D plate. Remember, the systems of equations for the 3D-plate and the 2D-thin-film are different. But, the displacement field output from the 3D-plate has to go as an input to the 2D-film. Only then the system of PDEs of the 2D-thin-film can be solved. With the above mentioned background, could anybody, please, suggest me whether this problem could be handled with FLEXPDE? If yes, then how to go ahead with it? Thanking in anticipation Rizwaan | ||
| Rizwaan (ritzyritz) Member Username: ritzyritz Post Number: 4 Registered: 02-2007 |
Hello Mr. Nelson, I have recently posted a problem. But could not avail any reply from anybody. Could you, please, put some thought to my problem, and do the needful. Thanks, Rizwaan | ||
| Robert G. Nelson (rgnelson) Moderator Username: rgnelson Post Number: 767 Registered: 06-2003 |
FlexPDE does not have provision for running coupled 3D and 2D systems. But you should be able to represent the 2D surface as a thin 3D layer. Put in a fictitious large diffusion term in the normal direction to force uniform values throughout the film. All Variables must exist and be defined throughout the domain, but you can use CONTACT boundaries to disconnect values in separate regions. (For this purpose you should put the layer on the side of your figure, as at present you can not make an extrusion surface a CONTACT boundary.) Provide simple equations for the variables that are meaningless in any region. You can define switching parameters in the regions and multiply terms by these parameters to turn off and on the terms of the equations. If the variables in the thin film are different than the variables in the 3D region, you could save memory space by using the same variable name for different quantities in the two regions. Simply use a CONTACT boundary with very high resistance to decouple the two regions. |