Author 
Topic 

Ashutosh98
India
7 Posts 
Posted  Oct 18 2022 : 18:39:07

I am doing a 2D analysis of a system of conductors arranged/embedded in a dielectric medium. The number of conductors in my system is N (variable controlled by me). I was initially modelling this system as 3D, but 2D seems to be getting a convereged solution much faster. The problem I am facing with the 2D analysis is that I should be getting a Capacitance Matrix of the dimension (N1)x(N1) since the last defined conductor in the .lst file will be considered ground, but instead I am getting a (N2)x(N2) capacitance matrix. I tried to investigate this problem by loading the refined 2D geometry in Fastmodel, and found that apparently Fastercap is completely ignoring the conductor which has been defined last in the .lst file. I figured that I might be doing something wrong in preparing the .lst file, so I went through the 2D examples given in the Fastercap documentation (like coax_cable_coated_2d_fine), and found they had done nothing special, and were defining the last conductor in a similar manner as me. So I tried to solve this problem crudely by introducing a dummy conductor which will get ignored by Fastercap. This seems to solve this problem, and now I am getting (N1)x(N1) matrices, but it does make me wonder what is happening which is causing Fastercap to ignore the last conductor. Also is it wise to assume that now the second last defined conductor acts as the ground?
Just to keep note: I am defining the last conductor in the same way as the conductors above it, i.e. I am using the same geometry .txt file, just its position (x and y offset) is different. Another very confusing thing which happens is that when I comment out the last defined conductor, the system remains the same according to Fastercap, and exactly the same solution is obtained. It's almost as if Fastercap needs a concluding statement or something.
It will be really helpful if someone can clear up this confusion.
Ashutosh Mukherjee 

Enrico
503 Posts 
Posted  Oct 19 2022 : 14:34:56

Hi Ashutosh,
is the last line really the last, i.e. no line feed / carriage return at the end of the line? (in a text editor, you should be able to scroll to the last empty line and not stop at the last one with some text in it)
If you have an empty line at the end, then I think I need to see the input file, if you can provide an example of the offending input.
Best, Enrico



Ashutosh98
India
7 Posts 
Posted  Oct 21 2022 : 08:38:31

Hi Enrico,
Thanks for the tip. I didn't realise an extra empty line was required at the end of the input file. Now I am able to obtain capacitance matrices with the correct dimensions.
Ashutosh Mukherjee 



Topic 


