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.
Problems modelling a microstrip resonator
Posted 25 feb 2016, 10:58 GMT-5 Low-Frequency Electromagnetics, RF & Microwave Engineering, Geometry Version 5.2 2 Replies
Please login with a confirmed email address before reporting spam
Hi there,
I am trying to model a microstrip resonator with COMSOL and I am having some problems with it since it's the first time I use this software.
Basically I followed the steps on the example " SMA connectorized wilkinson power divider" to generate my model but I don't know why it's not working at all. The microstrip is built with an SMA connector, attached to the substrate and with two thin layers on the top and bottom of the substrate as shown in the geometry of the model attached.
Probably the error is on the physics of the model but honestly, I don't know where to look for to learn how to use them. If you simulate the model, you could see that the electric field only appears on the volume next to the lumped port but you couldn't see on the conductive parts that are in contact with the port. I tried different configurations and probably it's a really stupid problem but I am not able to solve it. Any advice or explanation for that?
In the example I commented before (" SMA connectorized wilkinson power divider"), they apply a transition boundary condition for the microstrip but in my case it says that on the main part of the microstrip I generated this condition is not applicable, why? I also tried to modify the geometry of the example to my model but I didn't get any good results.
I also tried to generate the model avoiding to work with work planes and thin layers, and working with blocks directly but when I work with blocks with a thin component (around 30um) I obtain sometimes mesh errors or the computation time increases a lot and in the same way as before, I can only see the electrical field on next to the lumped port and not in the rest of the conductive parts.
Any advice with this model?
Thanks in advance!
Jaime
I am trying to model a microstrip resonator with COMSOL and I am having some problems with it since it's the first time I use this software.
Basically I followed the steps on the example " SMA connectorized wilkinson power divider" to generate my model but I don't know why it's not working at all. The microstrip is built with an SMA connector, attached to the substrate and with two thin layers on the top and bottom of the substrate as shown in the geometry of the model attached.
Probably the error is on the physics of the model but honestly, I don't know where to look for to learn how to use them. If you simulate the model, you could see that the electric field only appears on the volume next to the lumped port but you couldn't see on the conductive parts that are in contact with the port. I tried different configurations and probably it's a really stupid problem but I am not able to solve it. Any advice or explanation for that?
In the example I commented before (" SMA connectorized wilkinson power divider"), they apply a transition boundary condition for the microstrip but in my case it says that on the main part of the microstrip I generated this condition is not applicable, why? I also tried to modify the geometry of the example to my model but I didn't get any good results.
I also tried to generate the model avoiding to work with work planes and thin layers, and working with blocks directly but when I work with blocks with a thin component (around 30um) I obtain sometimes mesh errors or the computation time increases a lot and in the same way as before, I can only see the electrical field on next to the lumped port and not in the rest of the conductive parts.
Any advice with this model?
Thanks in advance!
Jaime
Attachments:
2 Replies Last Post 29 feb 2016, 18:39 GMT-5
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hi Jaime,
I advise to you mail your question to: support@comsol.com
Best,
François Mandias
I advise to you mail your question to: support@comsol.com
Best,
François Mandias
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
This is just my experience, and I am no expert. But I can say from adapting that particular model, creating the microstrip on a plane and placing it at the same height as the substrate works. Also it simplifies geometry to use unions and differences and to uncheck 'keep interior boundaries'. You have to be clever in how you build it, otherwise two conductors may accidentally be touching and shorting, or other such misc problems
Assuming youve built a workable geometry, under Component, go Definitions->Selections->Explicit. Change to 'Boundaries'. Then select all metal boundaries (or check 'select all boundaries' and then deselect the non metal ones). Rename it, say, 'Conductor boundaries'. Likewise add another explicit for the substrate boundaries 'Substrate Boundaries'. And probably another for the substrate domain as well. Call it 'Substrate Domain'.
In Materials, you can add two blank ones, each with er=ur=1 and sigma=0. Let one of them be 'All boundaries' and the other be the domain where the air is. Create another material for FR4 for the substrate, and in the drop down bar choose Substrate Domain.
In the physics section add another PEC. In the drop down for the boundary selection, choose 'Conductive boundaries'. If you want it to be copper, in the material section, add copper, and choose the Conductor boundaries there. then in EWM, choose Impedance Boundary and then Conductor boundaries. Check out the manual for the differences between Impedance, Scattering and Transition boundaries. Add 3 ports and define them as such, using 'coaxial'.
define the frequency range in Study->frequency domain. Run the study, and in result you can view the S parameters from derived values->global evaluation.
Your comsol is an older version than im using so I cant open your model, but those are just a few quick takaways that I got from working with this model. Hopefully it is somewhat helpful, and again skim the RF Module user guide. Good luck.
Assuming youve built a workable geometry, under Component, go Definitions->Selections->Explicit. Change to 'Boundaries'. Then select all metal boundaries (or check 'select all boundaries' and then deselect the non metal ones). Rename it, say, 'Conductor boundaries'. Likewise add another explicit for the substrate boundaries 'Substrate Boundaries'. And probably another for the substrate domain as well. Call it 'Substrate Domain'.
In Materials, you can add two blank ones, each with er=ur=1 and sigma=0. Let one of them be 'All boundaries' and the other be the domain where the air is. Create another material for FR4 for the substrate, and in the drop down bar choose Substrate Domain.
In the physics section add another PEC. In the drop down for the boundary selection, choose 'Conductive boundaries'. If you want it to be copper, in the material section, add copper, and choose the Conductor boundaries there. then in EWM, choose Impedance Boundary and then Conductor boundaries. Check out the manual for the differences between Impedance, Scattering and Transition boundaries. Add 3 ports and define them as such, using 'coaxial'.
define the frequency range in Study->frequency domain. Run the study, and in result you can view the S parameters from derived values->global evaluation.
Your comsol is an older version than im using so I cant open your model, but those are just a few quick takaways that I got from working with this model. Hopefully it is somewhat helpful, and again skim the RF Module user guide. Good luck.