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.
Reproducing a 'Floating Potential' via weak constraints
Posted 2016年5月4日 GMT+2 14:34 Version 5.2 1 Reply
Please login with a confirmed email address before reporting spam
Hello,
In order to understand weak constraints (to be used in a more complicated problem), I am trying to reproduce a 'Floating Potential' node (in electrostatics) by weak constraints.
To go systematically, I did the following in the attached model:
CASE 1: Electric Potential
I have set a fixed potential on the test electrode (left circle), -1V. I made a global evaluation to calculate the charge: intop1(nx*up(es.Dx)+ny*up(es.Dy)) - the integral on the electrode surface gives the charge, which resulted in -2.4152e-11 C/m
CASE 2: Floating potential
(Electric Potential disabled). I have specified a floating potential on the same electrode, and plugged in the value -2.4152e-11 C/m from the previous calculation. I get the same pattern
CASE 3: (Floating potential disabled)
- Global constraint: intop1(nx*up(es.Dx)+ny*up(es.Dy))+2.4152e-11[C/m], which is supposed to set the total charge of the electrode
- Weak constraint: nx*up(es.Ey)-ny*up(es.Ex), which is the cross product between the normal vector of the surface and E. This constraint prescribes that E is normal to the surface (i.e. no tangential component, the electrode surfaces is on isopotential).
The field pattern I get is completely different.
I am trying to read more about weak constraints and understand it better. It will go faster if somebody gives a smart hint.
From the 'Floating potential' node I deduced that it works by setting a constraint on V at the boundary (to some floating value), rather than requiring E to have a vanishing tangential component. How would this be done in a weak constraint node?
Thank you
In order to understand weak constraints (to be used in a more complicated problem), I am trying to reproduce a 'Floating Potential' node (in electrostatics) by weak constraints.
To go systematically, I did the following in the attached model:
CASE 1: Electric Potential
I have set a fixed potential on the test electrode (left circle), -1V. I made a global evaluation to calculate the charge: intop1(nx*up(es.Dx)+ny*up(es.Dy)) - the integral on the electrode surface gives the charge, which resulted in -2.4152e-11 C/m
CASE 2: Floating potential
(Electric Potential disabled). I have specified a floating potential on the same electrode, and plugged in the value -2.4152e-11 C/m from the previous calculation. I get the same pattern
CASE 3: (Floating potential disabled)
- Global constraint: intop1(nx*up(es.Dx)+ny*up(es.Dy))+2.4152e-11[C/m], which is supposed to set the total charge of the electrode
- Weak constraint: nx*up(es.Ey)-ny*up(es.Ex), which is the cross product between the normal vector of the surface and E. This constraint prescribes that E is normal to the surface (i.e. no tangential component, the electrode surfaces is on isopotential).
The field pattern I get is completely different.
I am trying to read more about weak constraints and understand it better. It will go faster if somebody gives a smart hint.
From the 'Floating potential' node I deduced that it works by setting a constraint on V at the boundary (to some floating value), rather than requiring E to have a vanishing tangential component. How would this be done in a weak constraint node?
Thank you
Attachments:
1 Reply Last Post 2016年5月4日 GMT+2 15:49
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
I discovered this example:
www.comsol.com/model/download/262881/models.mph.loaded_spring.pdf
which is a perfect analogue of my problem.
That is:
1 - use a 'Global equations' node, where one requires that the integrated n*D on the electrode surface equals the prescribed charge (Q0 - calculated from CASE1), introducing a state variable Vprobe
2 - set a Potential node on the probe electrode with potential Vprobe
It works almost OK, except that the solution at the end gives intop1(n*D)=Q0/2. The charge on the electrode is only half of the prescribed value... Any tips?
Thank you
www.comsol.com/model/download/262881/models.mph.loaded_spring.pdf
which is a perfect analogue of my problem.
That is:
1 - use a 'Global equations' node, where one requires that the integrated n*D on the electrode surface equals the prescribed charge (Q0 - calculated from CASE1), introducing a state variable Vprobe
2 - set a Potential node on the probe electrode with potential Vprobe
It works almost OK, except that the solution at the end gives intop1(n*D)=Q0/2. The charge on the electrode is only half of the prescribed value... Any tips?
Thank you
Attachments: