Transmission normalized to single source(Multiple sources simulation)


#1

My analysis contains two non-time overlapping TF/SF sources. I want to calculate the reflection and transmission from only one of the sources but transmission command uses normalization relative to the two sources. I tried to re-normalize and then normalize to the one source I am interested in but didn’t give me right results.
How can I calculate the transmission relative to a single source?
Thanks a lot.


#2

Hello, the transmission command is normalized the sourcepower command. And the sourcepower command returns the sum of all sourcepowers if there are more than one source. With some manual processes for renormalization, it should be possible to normalized the recorded power to the power of the source you want. However, it is not very clear to me why you would want to do that. If you can tell us more about the application you are trying to simulate and the name of the reference paper you are trying to reproduce the results. I could try to provide more comments.


#3

Thanks for your help.


#4

Dear, I’m not sure that I have fully understand your answer.
For example, one source is TE0 mode source and its amplitude is 0.7. And the other source is TE2 mode and its source is 0.7 too.
It I set a monitor to record the transmission, the maximum value should still be 1 or 1.4? Or the maximum one is sqrt(0.7^2+0.7^2)
Could you help me?


#5

Let’s use plane wave as an example, because it is far more easy to understand when using plane waves.

The result of the transmission command is calculated as a fraction stated on the KB. The numerator is the power recorded by the monitor in an FDTD simulation, and denominator is the sourcepower result. FDTD returns power summed coherently since FDTD is a coherently method. The sourcepower result returns power summed incoherently.

You can run the script below and check the transmission. If the two sources have the same polarization and amp_s1 = amp_s1, the transmission should be 2. If the two sources have different polarization direction, you need to edit fdtd_sum.

amp_s1 = getnamed("source_1", "amplitude");
amp_s2 = getnamed("source_2", "amplitude");

fdtd_sum = (amp_s1 + amp_s2)^2; # for sources have the same polarization
source_sum = (amp_s1)^2 + (amp_s2)^2;

trans =  fdtd_sum / source_sum;

source power test.fsp (273.8 KB)

Summing the power of the two different modes are a bit tricky because there are more factors to consider such as shape, polarization, etc. We typically do not have multiple mode sources in a FDTD simulation. Depending on the situation, you may want to consider using only one mode source at time time.


#6

Dear kchow,

I run your simulation file. And the transmission value is around 2.

In my point of view, the value should around 1, meaning 100%. Why???

And I don’t quite catch your reply. it’s easy to understand when the polarization angles are both zero. And how about the different polarization angles???

My original example is correct or not?


#7

Sorry for the late reply, I was on a business trip so wasnt able to reply in time.

yes, in my example, there are two sources, the transmission is 2, means 200%. The reason is because of the way the transmission takes the power from FDTD results (coherently), and power sum of the two sources (incoherently), as shown and explained in my last post. In this case, FDTD_power = 2^2, source_power_sum = (1^2 + 1^2), the ratio of the two becomes 2 = transmission.

if you only use one source, and set the amplitude to 2, you will see transmission = 1. In this case, FDTD_power = 2^2, source_power_sum = (2^2), the ratio of the two becomes 1 = transmission.