Low pass filter using CST simulation

[johnjoe]

Results with parasitics is better than a result with garbage!

1) CST or other simulation programs are nothing to play with. They're expensive, many people put a lot of effort into it to make a great product. And now everybody can download illegally the software (without even paying respect for the developers) and ask questions that demonstrate that they even didn't understand the basic concepts behind these tools.

2) I think, it's not the intention of such products to produce "some" results. They should predict the real performance as best as possible, and the user has to set up the software correctly to get reasonable results. If he don't understand something, he should ask instead just doing something. By the way, CST has a good help function which explains most topics.

Even if he place a ground plane far away or does some other stuff, I'm absolutely sure he will ask one day why there is such big difference between measurement and simulation results. I think it's not the best to teach something.

- - - Updated - - -

I did not use discrete ports

You have to place your waveguide ports at the sides of the traces.

 

[Shahendaeid]

1-I used a licensed program i did not download it from the internet
2-I used wave giude ports not discrete ports

 

[johnjoe]

Ok, everything fine. The previous comments were not about you.
Place the waveguide ports at the side of the trace.

 

[pragash]

hi,
since you have third layer which is conductor, its not a problem at all. you can use waveguide port as well. not an issue. all you need to do is make sure your waveguide port touches filter input and the thrid layer conductor. refer to video below for assistant.

https://www.youtube.com/watch?v=k54XWrcI5Bc

 

[Shahendaeid]

Thanks a lot for your help.I get better result but i still have this error "The input reflection of the S-parameters seems to be large at 8 GHz. A constant mesh adaptation sample will not be changed"

 

[pragash]

its ok! the method i have mentioned will work for waveguide port as well!

 

[Shahendaeid]

its ok! the method i have mentioned will work for waveguide port as well!

It is ok when I make the frequency range maximum 5 Ghz but when I increase the frequency range for example to 10 G HZ i got the error i mentioned before and also the following errors (At least one propagating mode is not considered at port 1.
At least one propagating mode is not considered at port 2)

 

[gyz182]

(At least one propagating mode is not considered at port 1.
At least one propagating mode is not considered at port 2)

I believe this indicates higher order modes propagating at higher frequencies which wasn't accounted for in your waveguide port excitation.

 

[Shahendaeid]

I also got the following error Mesh adaptation terminated because the maximum number of passes is reached. could anyone know how to solve it

 

[FvM]

Let me guess, you are running CST Student Edition? Then it's a helpless case. Otherwise there's a serious problem in your design definition, as in the previously shown screenshots (post #11) where the antenna structure isn't appropriately meshed.

 

[gyz182]

I also got the following error Mesh adaptation terminated because the maximum number of passes is reached. could anyone know how to solve it

I could be wrong, but you might have activated adaptive mesh refinement without realising or the intention to do so? When adaptive mesh refinement is activated, the mesh is refined to improve accuracy of results. Sometimes, as mesh gets finer, the results are not necessarily improving by much. Therefore, a max number of passes is set to terminate mesh adaptation even if S-parameters have not converged.

So, the error mentioned above could just be the system telling you that no further refinement will be conducted as the max number of passes is reached.

 

[Shahendaeid]

I get the correct result finally thanks all for your help

 

[tejaswiau09]

port should be connected to all the three filter strip, dielectric and ground metal layer