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.
Equivalent of a "ground" boundary at infinity ?
Posted 16 giu 2010, 10:46 GMT-4 3 Replies
Please login with a confirmed email address before reporting spam
Hello everybody,
i'm actually using COMSOL v3.5a with AC/DC module to simulate how go electric fields and associated forces in various kind of solutions at nanoscale but i get one main problem.
I'm trying to simulate the electric field generated by a charge pattern at the surface of a 100nm PMMA (polymer) layer once immersed in a hexane solution.
The system/geometry is quite easy (see joined document) but the major issue is that i need to put some boundary to the ground to get a solution returned by COMSOL. Of course, depending on the way i define the ground boundary, my electric field lines change a lot. Furthermore, normally and physically speaking, the ground is "ad infinitum".
I hope you will have a few minutes to have a look on my model and give me advices about this.
Thanks in advance for your attention,
Etienne
i'm actually using COMSOL v3.5a with AC/DC module to simulate how go electric fields and associated forces in various kind of solutions at nanoscale but i get one main problem.
I'm trying to simulate the electric field generated by a charge pattern at the surface of a 100nm PMMA (polymer) layer once immersed in a hexane solution.
The system/geometry is quite easy (see joined document) but the major issue is that i need to put some boundary to the ground to get a solution returned by COMSOL. Of course, depending on the way i define the ground boundary, my electric field lines change a lot. Furthermore, normally and physically speaking, the ground is "ad infinitum".
I hope you will have a few minutes to have a look on my model and give me advices about this.
Thanks in advance for your attention,
Etienne
Attachments:
3 Replies Last Post 18 giu 2010, 07:50 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
The way I understand it, is that you have basically two situations:
a) either symmetry zero charge or
b) ground
all others are equivalent if you look at the eqautions and fill them in with the default values proposed,
this gives you basically two options: field lines parallel to the border or perpendicular to the border.
What would happen if you have a half a circle and not a square (as external border) ?
And what do you think is more representative ?
I'm not that sure what to say
This is better viewed if you look/plot at the "log10(normE_emes)" (I find log10 easier to read than ln) you have fields changes over > 7 orders of magnitude, this is the full precision range of floating point numbers I beleive, and below about 1E-4, relative value of the field, you see some numerical non linearities appear (concentration points, I beleive these are typical numerical issues due to the representation of numbers (binary is not decimal ! spacings are not even, you tend to end up with singularitiy points)
Good luck
Ivar
The way I understand it, is that you have basically two situations:
a) either symmetry zero charge or
b) ground
all others are equivalent if you look at the eqautions and fill them in with the default values proposed,
this gives you basically two options: field lines parallel to the border or perpendicular to the border.
What would happen if you have a half a circle and not a square (as external border) ?
And what do you think is more representative ?
I'm not that sure what to say
This is better viewed if you look/plot at the "log10(normE_emes)" (I find log10 easier to read than ln) you have fields changes over > 7 orders of magnitude, this is the full precision range of floating point numbers I beleive, and below about 1E-4, relative value of the field, you see some numerical non linearities appear (concentration points, I beleive these are typical numerical issues due to the representation of numbers (binary is not decimal ! spacings are not even, you tend to end up with singularitiy points)
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Ivar,
You got a great idea by using a half circle instead of a square. Finally, i will use a "huge" circle (compared to my PMMA layer) with boundaries at ground and "filled" with hexane which surrounds my whole system. This is I think the most representative.
Furthermore, I have learned from COMSOL support that now you can implement a mass at infinity by using infinite elements (only available in v.4). This would help too.
I will certainly acknowledge you in my next paper !
Thanks again
Etienne
You got a great idea by using a half circle instead of a square. Finally, i will use a "huge" circle (compared to my PMMA layer) with boundaries at ground and "filled" with hexane which surrounds my whole system. This is I think the most representative.
Furthermore, I have learned from COMSOL support that now you can implement a mass at infinity by using infinite elements (only available in v.4). This would help too.
I will certainly acknowledge you in my next paper !
Thanks again
Etienne
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
Good that you progress, well there is another thread discussing this too, with some other nice ideas, check the forum from the last days
have fun Comsoling
Ivar
Good that you progress, well there is another thread discussing this too, with some other nice ideas, check the forum from the last days
have fun Comsoling
Ivar