Problem with layers and regions in 3D Log Out | Topics | Search
Moderators | Register | Edit Profile

FlexPDE User's Forum » User Postings » Problem with layers and regions in 3D « Previous Next »

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

Andrea Siefers (asiefers)
New member
Username: asiefers

Post Number: 1
Registered: 07-2006
Posted on Wednesday, July 26, 2006 - 10:45 am:   

Hi, I'm also relatively new to this program and have been using it to help create a system that measures electrical properties of a sample 15x5 microns in dimension. The system consists of 4 Cu contacts that are attached to 4 Pt contacts which are attached at different places to the sample (in our simulation the sample is Ag). Voltage is set at 2 of the Cu contacts and measured across the remaining 2.

The problem we are having is concerning the boundaries of the layers and regions. As you can see in the attached file, the Cu contacts are on the bottom layer, the Pt on the next layer and the sample sits on the top. When I first ran this script I defined everything as just "regions" and didn't have an 'outer boundary' region that encompasses everything else. I ran into the problem that the regions that overlapped eachother were actually being extruded into all layers, which skews the electrical measurements. After consulting the user guide, I added an 'outer boundary' region and changed the rest of the regions to "limted regions". Now, depending on what version I'm running, I either get a message that the surfaces are crossed (version 5) or that the mesh is tangled (version 4).

I found the following post in this forum:



Robert G. Nelson (rgnelson)
Moderator
Username: rgnelson

Post Number: 505
Registered: 06-2003
Posted on Tuesday, December 13, 2005 - 06:22 pm:

--------------------------------------------------------------------------------
FlexPDE allows you to say a lot of different things in your script, and unless I know what you said, it is difficult for me to guess what has happened.

There is one known error in the processing of Limited regions:

If you have two Limited Regions existing exclusively in adjacent layers, and if these regions overlap, then in the surface separating the layers, the regions will be treated according to the 2D rule that regions stated later in the script will overlay and hide parts of regions stated earlier in the script.

This may cause the outline of the occluded region to be absent in the surface mesh, resulting in tangling of the 3D mesh.

If one of these regions is totally contained in the other, then you can cure the problem by stating the larger region first in the script.

If there is no total containment, you might be able to cure these troubles by making both regions present in both layers, assigning the appropriate material properties in the regions. Depending on your layout, this might merely cascade the trouble to the next layer.

If neither of these things describes your case, you will have to send me the script so I can see what you are attempting to do.



Is this the problem I'm experiencing? And if so, is there any way I can remedy the script to get accurate results?

Thank you for your time!
application/octet-stream3D electrical contacts
3D contacts 03.pde (5.8 k)
Top of pagePrevious messageNext messageBottom of page Link to this message

Robert G. Nelson (rgnelson)
Moderator
Username: rgnelson

Post Number: 664
Registered: 06-2003
Posted on Friday, July 28, 2006 - 12:51 am:   

The crash seems to be due to the fact that some components are absent in various surfaces where they are in fact needed. I have not dug through your rather complex layout to discover the details of this.

I suggest that to start off, you do not use LIMITED REGIONS, as the logistics of which layers and surfaces need to be specified makes it more difficult to understand. Later, when you have a better grasp of the way layers and regions are constructed, you can make more compact specifications using LIMITED REGIONS.

If you use standard REGIONS, you merely need to specify the material properties for each layer that differs from the default.

Notice that when you specify regions that partially overlay one another, the one named last will over-ride and hide the shared portion of the region named earlier.

If in fact you want different materials in different layers of these overlapped regions, you should declare the overlapped part as a separate region, so you can properly assign material properties.

You might do well to start by building your problem domain one layer at a time. Satisfy yourself that the first layer is correct, then add the second, etc. Use the "Domain Review" button or menu item to walk through the construction process.

Some other comments about your existing script:
1) the region 1 boundary is not closed. The last two points are the same point, and the curve does not close. Use "TO CLOSE" as the last leg of each bounding path to automate the closing process.
2) The specification cond=0 in region 1 makes the equation collapse to 0=0. This is not a legal equation. Use a nonzero cond or delete region 1.

Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action:

Topics | Last Day | Last Week | Tree View | Search | Help/Instructions | Program Credits Administration