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.
RF-module: Possible bug in interactions between the new "Surface Magnetic Current Density" and "Surface Current Density"
Posted 2017年8月15日 GMT-4 23:06 RF & Microwave Engineering, Modeling Tools & Definitions, Parameters, Variables, & Functions, Studies & Solvers Version 5.3 1 Reply
Please login with a confirmed email address before reporting spam
In the RF-module, I am very excited to see the new "Surface Magnetic Current Density" boundary condition. It is something that I have been waiting for, for some time. That said, I think I found a particular set of interactions that produce an error.
I am particularly interested in simulating a boundary condition which has both an Electric Current and Magnetic Current. These two boundaries combined can simulate a thin meta-surface, with a bianisotropic response. In order to implement this, I set both conditions on the same boundary. I notice that in each Boundary Conditions' settings, under "Override and Contribution", they list the other as "contribution". So it appears there is no conflict.
The issue arises when I begin to assign the surface current density. This error can arise from simply constants, so I will demonstrate my issue using the example of constants. For this particular example, if I set
Js (electric): (r, phi, z) = (1, 0, 0)
Jsm (magn): (r, phi, z) = (0, 1, 0)
I get the error: Undefined variable - comp1.tEr
This combination seems to be very explicit. For example, this other combination will simulate without error:
Js (electric): (r, phi, z) = (0, 1, 0)
Jsm (magn): (r, phi, z) = (1, 0, 0)
There are random combinations that will also not work, but the most important is the one I showed above. It so happens that I need this exact relation of currents.
The obvious answer seems to define a variable, tEr, that will track the r-direction component of E on the surface, but this wont work because the variable tEr already exists. The issue seems to be the simulation improperly attempting to reference this variable before it assigns it (but... only with very specific combinations of elec/magn currents).
Note: for this error to present itself, the boundary must not be an outside-edge boundary. It must present both sides to the interior of the solution space.
Note2: Currently, 3 different web browsers are finding the "Attach File" button completely unresponsive. I will have to try and upload an example later.
I am particularly interested in simulating a boundary condition which has both an Electric Current and Magnetic Current. These two boundaries combined can simulate a thin meta-surface, with a bianisotropic response. In order to implement this, I set both conditions on the same boundary. I notice that in each Boundary Conditions' settings, under "Override and Contribution", they list the other as "contribution". So it appears there is no conflict.
The issue arises when I begin to assign the surface current density. This error can arise from simply constants, so I will demonstrate my issue using the example of constants. For this particular example, if I set
Js (electric): (r, phi, z) = (1, 0, 0)
Jsm (magn): (r, phi, z) = (0, 1, 0)
I get the error: Undefined variable - comp1.tEr
This combination seems to be very explicit. For example, this other combination will simulate without error:
Js (electric): (r, phi, z) = (0, 1, 0)
Jsm (magn): (r, phi, z) = (1, 0, 0)
There are random combinations that will also not work, but the most important is the one I showed above. It so happens that I need this exact relation of currents.
The obvious answer seems to define a variable, tEr, that will track the r-direction component of E on the surface, but this wont work because the variable tEr already exists. The issue seems to be the simulation improperly attempting to reference this variable before it assigns it (but... only with very specific combinations of elec/magn currents).
Note: for this error to present itself, the boundary must not be an outside-edge boundary. It must present both sides to the interior of the solution space.
Note2: Currently, 3 different web browsers are finding the "Attach File" button completely unresponsive. I will have to try and upload an example later.
1 Reply Last Post 2017年8月16日 GMT-4 17:34
Please login with a confirmed email address before reporting spam
Posted:
7 years ago
For anyone who is finding a similar issue, I also posted this to COMSOL support as a potential bug. They replied with the following response:
---------------------------------------------------
It is a known issue and fixed for the upcoming release. For the current
version you are using, I have modified the weak expression in the Surface
Current Density feature as
"2*mu0_const*emw.iomega*(-emw.scu1.Jsr*test(emw.tEupr)-emw.scu1.Jsphi*test(emw.tEupphi)-emw.scu1.Jsz*test(emw.tEupz))*pi*r"
So the slit condition created by the Surface Magnetic Current Density was
the cause of conflict with the general expression used in the Surface
Current Density.
---------------------------------------------------
It appears a solution will be forthcoming.
---------------------------------------------------
It is a known issue and fixed for the upcoming release. For the current
version you are using, I have modified the weak expression in the Surface
Current Density feature as
"2*mu0_const*emw.iomega*(-emw.scu1.Jsr*test(emw.tEupr)-emw.scu1.Jsphi*test(emw.tEupphi)-emw.scu1.Jsz*test(emw.tEupz))*pi*r"
So the slit condition created by the Surface Magnetic Current Density was
the cause of conflict with the general expression used in the Surface
Current Density.
---------------------------------------------------
It appears a solution will be forthcoming.