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.
plate capacitor in frequency domain emf
Posted 26 mars 2013, 05:55 UTC−4 4 Replies
Please login with a confirmed email address before reporting spam
Hi,
I built a model of a simple two plate capacitor and I want to compute the capacitance in frequency domain with the emf interface.
I built two plates, set the material to copper, created a surrounding air and set one plate to "terminal" and one plate to "ground"
But something seems to be wrong with this model. There is no electrical potential field in the Air domain after computing the study. And the admittance is only a negative real value (see the values stored in the table ).
Maybe someone could give me a hint whats wrong with this model ?
I built a model of a simple two plate capacitor and I want to compute the capacitance in frequency domain with the emf interface.
I built two plates, set the material to copper, created a surrounding air and set one plate to "terminal" and one plate to "ground"
But something seems to be wrong with this model. There is no electrical potential field in the Air domain after computing the study. And the admittance is only a negative real value (see the values stored in the table ).
Maybe someone could give me a hint whats wrong with this model ?
Attachments:
4 Replies Last Post 26 mars 2013, 11:01 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Marius,
You are using Form Assembly mode but not imposing continuity condition. Not sure why you need assembly mode? Attached is potential distribution obtained with your model using Form Union option. Admittance is positive now.
Your PML domain has only one layer of mesh elements. Generally, one needs at least five elements through the thickness of PML domain. Also, use mapped mesh in 2D (and sweep mesh in 3D) for better performance of PML domain.
Regards,
Sergei
You are using Form Assembly mode but not imposing continuity condition. Not sure why you need assembly mode? Attached is potential distribution obtained with your model using Form Union option. Admittance is positive now.
Your PML domain has only one layer of mesh elements. Generally, one needs at least five elements through the thickness of PML domain. Also, use mapped mesh in 2D (and sweep mesh in 3D) for better performance of PML domain.
Regards,
Sergei
Attachments:
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Sergei,
thank you for your answer.
I tried to use "form union" but with this option the Terminal and Ground was "not applicable". And so i tried "form assembly".
Could you attach the model that you used for making the graphic ?
thank you for your answer.
I tried to use "form union" but with this option the Terminal and Ground was "not applicable". And so i tried "form assembly".
Could you attach the model that you used for making the graphic ?
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Marius,
Attached is model with Ground and Terminal BCs using Form Union option.
In “mef” interface, the current path must be solenoidal in order to satisfy Maxwell’s equations. It means that all current flow is excited via the exterior boundaries (in that case, closing loop is formed by return current along exterior boundaries) or closed current loop should be completely inside computational domain (or form closing path through symmetry boundary).
In your model, you tried to excite current flow via interior terminal boundaries and software didn’t allow you to do that saying that terminal BC’s are “not applicable”.
Regards,
Sergei
Attached is model with Ground and Terminal BCs using Form Union option.
In “mef” interface, the current path must be solenoidal in order to satisfy Maxwell’s equations. It means that all current flow is excited via the exterior boundaries (in that case, closing loop is formed by return current along exterior boundaries) or closed current loop should be completely inside computational domain (or form closing path through symmetry boundary).
In your model, you tried to excite current flow via interior terminal boundaries and software didn’t allow you to do that saying that terminal BC’s are “not applicable”.
Regards,
Sergei
Attachments:
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Sergei,
thank you for attaching the model and your answer.
Unfortunateley I can't open the model because I'm using Version 4.3.
But I tried to exclude the terminal and ground from the node "magnetic and electric fields". So the Boundaries of terminal and ground become exterior.
Now it seems to work. I got a potential distribution and a positive admittance
thank you for attaching the model and your answer.
Unfortunateley I can't open the model because I'm using Version 4.3.
But I tried to exclude the terminal and ground from the node "magnetic and electric fields". So the Boundaries of terminal and ground become exterior.
Now it seems to work. I got a potential distribution and a positive admittance