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.
Circularity problem in Electric-Currents model due to sigma(J)
Posted Apr 30, 2017, 7:14 a.m. EDT 3 Replies
Please login with a confirmed email address before reporting spam
Hello,
I'm trying to use the electric-currents model to see the current density distribution in a material (superconductor) that has an Electrical-conductivity vs Current density dependency (the electrical conductivity changes according to the applied current density in it). I tried to enter that dependency into the material properties but the solver tells it cant be solved due to circularity.
Does anyone knows how can I overcome that issue?
I'm trying to use the electric-currents model to see the current density distribution in a material (superconductor) that has an Electrical-conductivity vs Current density dependency (the electrical conductivity changes according to the applied current density in it). I tried to enter that dependency into the material properties but the solver tells it cant be solved due to circularity.
Does anyone knows how can I overcome that issue?
3 Replies Last Post May 16, 2017, 12:19 p.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hello Ilan,
The circular definition comes from the fact that the definition of J contains sigma.
Could you express the conductivity as a function of the electric field E instead?
Best,
Jeff
The circular definition comes from the fact that the definition of J contains sigma.
Could you express the conductivity as a function of the electric field E instead?
Best,
Jeff
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hi Jeff,
I'm not sure how could that help. I have to have some dependency in the current density (as in a real superconductor). The dependency can come from either the conductivity or electrical field. They both end up with circularity.
I thought about:
1. Using the PDE model instead and defining it there but I'm not that familiar with the PDE model and don't know how to define there the "normal current density" and "ground" for where should the current flow.
2. Using the magnetic field formulation and and after solving, converting the magnetic field to current density but I don't how to define the appropriate magnetic field to get the wanted induced current in my geometry.
Any advice?
I'm not sure how could that help. I have to have some dependency in the current density (as in a real superconductor). The dependency can come from either the conductivity or electrical field. They both end up with circularity.
I thought about:
1. Using the PDE model instead and defining it there but I'm not that familiar with the PDE model and don't know how to define there the "normal current density" and "ground" for where should the current flow.
2. Using the magnetic field formulation and and after solving, converting the magnetic field to current density but I don't how to define the appropriate magnetic field to get the wanted induced current in my geometry.
Any advice?
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Updated:
8 years ago
Think of it this way:
Let's say that in a linear iteration, you have an approximation to the V field.
If sigma is expressed as a function of J, you cannot compute J based on its definition as sigma times E because sigma cannot be computed without already knowing J: circular dependency.
BUT, if sigma is available as a function of E, you can get J by multiplying sigma(E) by E (Since E can be computed from V, which you have).
If you need further assistance, please contact the Technical Support Team (support@comsol.com).
Best regards,
Jeff
Let's say that in a linear iteration, you have an approximation to the V field.
If sigma is expressed as a function of J, you cannot compute J based on its definition as sigma times E because sigma cannot be computed without already knowing J: circular dependency.
BUT, if sigma is available as a function of E, you can get J by multiplying sigma(E) by E (Since E can be computed from V, which you have).
If you need further assistance, please contact the Technical Support Team (support@comsol.com).
Best regards,
Jeff