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.
Why does my model (transmission through a interface) work well at most incident angles but failed at some angles?
Posted 16.09.2010, 03:53 GMT-4 1 Reply
Please login with a confirmed email address before reporting spam
I want to study the transmission and reflection of a periodical structure. So a simple model was build for testing.
The model simulates light incident on a interface of air and a material (refractive index: 1.5).
I compared the transmission curve with the theoretical one. At most incident angles, especially at small incident angles, the model works well and gives reasonable transmission. However, at some angles, for example 76 degree, the transmission deviates the theorectical results much.
You can see the details in the attached files.
I will be very grateful if anyone can help me.
The model simulates light incident on a interface of air and a material (refractive index: 1.5).
I compared the transmission curve with the theoretical one. At most incident angles, especially at small incident angles, the model works well and gives reasonable transmission. However, at some angles, for example 76 degree, the transmission deviates the theorectical results much.
You can see the details in the attached files.
I will be very grateful if anyone can help me.
1 Reply Last Post 21.09.2010, 01:38 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
I havent manage to test it out but I have a few thoughts: oscillation in the results, can come from several items such as not enough mesh elements (particularly when you com at grazing insicence and your mesh density is not "square", then from the element type (higher order can give some oscillations around steep gradients (try to refine the mesh and see if there are changes). Thereafter you have the order of the integration variables, the default is 4th order which means that these can give oscillations, particularly around steep gradients and with insufficient mesh points (try reducing the integration order in the advance tab). And finally there is some internal numerical "damping" to improve model convergence, this can result in oscillatin results particularly at strong transitions, here is a full chapter devoted to this is the doc.
These are the elements that I can propose to check out first, I agree for such a "simple" case one could expect to get a better fit, but this is typically the model validation & verification issue that one must always do to be sure that the model represents truely the problem one want to study
--
Good luck
Ivar
I havent manage to test it out but I have a few thoughts: oscillation in the results, can come from several items such as not enough mesh elements (particularly when you com at grazing insicence and your mesh density is not "square", then from the element type (higher order can give some oscillations around steep gradients (try to refine the mesh and see if there are changes). Thereafter you have the order of the integration variables, the default is 4th order which means that these can give oscillations, particularly around steep gradients and with insufficient mesh points (try reducing the integration order in the advance tab). And finally there is some internal numerical "damping" to improve model convergence, this can result in oscillatin results particularly at strong transitions, here is a full chapter devoted to this is the doc.
These are the elements that I can propose to check out first, I agree for such a "simple" case one could expect to get a better fit, but this is typically the model validation & verification issue that one must always do to be sure that the model represents truely the problem one want to study
--
Good luck
Ivar