what's the interpretation of these results?(an antenna in two different spaces)

[project.email]

Dear friends
I've designed an antenna for (402-405MHz)frequency band to implant ir in a lossy medium the antenna configuration ,the S11 result and the port signals (i.e incident and reflected signals)are shown in Fig.1
In the other case I simulated the mentioned antenna in the free space and got the S11,port signals .
So the question here is:
Why in the free space(i,e in Fig2) the S11 result has ripples in it's nature?
I know that S11 in Fig2, tell us that the reflected power is almost equal to the incident power so, this antenna is not suitable for 402-405MHz in the free space but it is suitable for the lossy medium in the former frequency band.
What's the other information that these figures tell us?


the results of the antenna in the free space:

 

[FvM]

As |S11| > 1 doesn't represent physical reality, the correct topic is simulation artefacts, I presume.

 

[volker_muehlhaus]

Why in the free space(i,e in Fig2) the S11 result has ripples in it's nature?

It seems that your simulation time period was not long enough. As you can see, there is still a lot of energy at the end of the time plot. This leads to ripple in the S-parameters. Increase the length of the simulation period.

 

[project.email]

As |S11| > 1 doesn't represent physical reality, the correct topic is simulation artefacts, I presume.

dear FvM you are right
because the S11 is the ratio of the V(reflected) to the V(incident) and if it be larger than 1 the reflected wave should be larger than incident wave that it is not physically possible.
but assume the S11 is actually the return loss plot.
so what's your interpretation of these plots?
Regards.

 

[project.email]

It seems that your simulation time period was not long enough. As you can see, there is still a lot of energy at the end of the time plot. This leads to ripple in the S-parameters. Increase the length of the simulation period.

Dear volker_muehlhaus

I changed the excitation signal duration like the following figure and increased the time duration but the previous result obtained again without any particular change.(i.e in aspect of ripple existance)



I know that this antenna has been designed for the lossy medium not for free space because it's resonating better in the lossy medium than the free space but the fact that disturb me is that why in the free space the rerun loss plot has some ripples?
It's the main question:
What dose happen, when we face with ripple in return loss plot?

 

[volker_muehlhaus]

I changed the excitation signal duration like the following figure and increased the time duration but the previous result obtained again without any particular change.(i.e in aspect of ripple existance)

Your results show the expected, significant change: the amplitude of the ripple is much smaller now (0.00x dB).

The fact that there is some ripple is related to your time dimain simulation method, which needs FFT to transform the time domain results into frequency domain. If the time period over which you have simulated is too short, and there is still energy in the model that did not leave through the ports, there will be some ripple on the S-parameters. The higher the residual energy in the system, the bigger that error/ripple.

In your case, even the small 0.00x dB ripple is seen because your antenna has (almost) total reflection in this frequency range. Or maybe there's a mistake in the model where you created a short/open circuit.

 

[FvM]

The results suggest that you have placed your antenna in a box with pefectly reflecting boundaries instead of a reasonable free space simulation enviroment (with absorbing boundaries). If the simulation is intended to show an antenna in free space, you should change the setup.

 

[project.email]

The results suggest that you have placed your antenna in a box with pefectly reflecting boundaries instead of a reasonable free space simulation enviroment (with absorbing boundaries). If the simulation is intended to show an antenna in free space, you should change the setup.

in fact, the simulation is intended to show an antenna in the lossy medium .
so, the simulation for this medium is acceptable rather than free space,right?

 

[FvM]

A lossy medium in a reflecting box will work, but your free space simulation is apparently wrong. If the boundaries for the "lossy medium" simulation are setup correctly is a different question. Due to the medium losses, the antenna characteristic looks regular at first sight.

 

[volker_muehlhaus]

I've done EM solver support for many years, including these time domains solvers. As already mentioned above: truncated time signals will lead to ripple in the S-parameters. You don't see the ripple in the lossy medium plot because that has a much larger scale (15dB vs. 0.006dB).

I definitely agree to FvM that absorbing boundary conditions should be used, to enable radiation. However, I doubt that the ripple is related to the boundary conditions because your second run with longer simulation period shows much reduced ripple. It the ripple was from the boundary conditions in an electrically large box, it should not change with the simulation period.

From your plot, I cannot see if you have used wrong boundary conditions which prevent radiation, or if it's just the different permittivity of the medium that has tuned the antenna too a much different resonance frequency.

 

[project.email]

These figures show the boundary condition for two cases



the antenna ,lossy medium and the surrounding air boundary condition in the second case is shown as below:



I think in two cases I've applied the boundary condition, correctly.

The type of the boundary condition in each of two cases are open boundary in the CST software.