Simulation Problem in CST Microwave Studio

[winson]

Hi everybody,

I'm new to CST microwave studio. Just finish constructed a structure of an L-probe patch antenna (from IEEE paper) and just run the simulation by transient time solver, the curve of the return loss(S11) against frequency that i get is different from what showing on the IEEE paper, so is it the boundary condition setting will affect the simulation results? And actually what is the function of setting the boundary condition?

During the simulation, a warning message "some PEC material is touching the boundary" was show. After change the boundary condition setting to "open(add space)" then the warning will eliminate when run again the simulation. But the s11 curve still different from the "actual" results.

Anybody can help?
Any comments will be appreciate.

Thanks.

[Admin]

so is it the boundary condition setting will affect the simulation results? And actually what is the function of setting the boundary condition?

For something like this you should use an 'infinite' boundary. I'm not sure what the paper you read did, but if you get any constants wrong (air vs vacuum, metal conductivity, temperature of environment and materials, antenna dimensions, etc), then your results will be different.

Also, what are you solving for? Do you need to mesh anything first before solving?

[winson]

Hi,

Thanks for reply!

I'm just aware got a message here. Actually i have already able to obtain the results(s11 plot or return loss) as describe in the paper that i read recently, before that i was using CST 2009 and i was not able to get the correct results until i switch to CST 2006 and without many changing of the dimension of the antenna then i can get the correct results, not very sure what is reason also.

Regarding the boundary conditon, i have set it "open(add space)" and set "et = 0" below the ground plane, and with these setting the results was nice. So i think "open(add space)" will be same meaning as what you say "infinite" boundary.

Also, what are you solving for? Do you need to mesh anything first before solving?

Erm... i'm not very sure what is your meaning. I guess you are asking me what outcome i want to get right? If so, the results that desired is a s11 plot which is the plot that reflect the return loss of the patch antenna and also how large of the bandwidth that the antenna able to achieve with certain resonance frequency. CST have a automatic mesh generator it will help to generate the mesh line that required for the solving. On the whole i'm not so sure how to tune the mesh so that we will have the less number of mesh cell vs the most accurate results and less mesh cell will means that we will have shorter calculation time.

This antenna was resonant at 2.4GHz, now i was plan to shift the resonant frequency to 0.8GHz by scale the antenna to bigger size so that shift to lower frequency, and the results was not very good too, just able to get back half of the bandwidth and 10dB of return loss at low frequency. Any idea or suggestions on how to make the bandwidth become wider and also make the return loss become larger also.

Thanks

[Admin]

On the whole i'm not so sure how to tune the mesh so that we will have the less number of mesh cell vs the most accurate results and less mesh cell will means that we will have shorter calculation time.

Lets say you have a very high mesh and you get 100% accuracy. Then lets say you do a medium mesh, it calculates twice as fast, but you get 98% accuracy. That second mesh is more optimal.

Try the same calculation at multiple meshes, and you'll find a sweet spot where its fast, but still gives good accuracy.

When I do CFD I would double the mesh and see how much the result changes. If its only 1 or 2%, you don't need that accurate of a mesh.

This antenna was resonant at 2.4GHz, now i was plan to shift the resonant frequency to 0.8GHz by scale the antenna to bigger size so that shift to lower frequency, and the results was not very good too, just able to get back half of the bandwidth and 10dB of return loss at low frequency. Any idea or suggestions on how to make the bandwidth become wider and also make the return loss become larger also.

Actually I've been doing a lot of antenna testing lately, myself. I've been doing a lot of calculations on attenuation by water on a signal. But I'm no expert!

The optimal antenna length is equal to the wavelength:

wavelength = speed of light / frequency

= 299792458 m/s / 2.4 GHz = 0.118792761 meters

But know that the material of the antenna, and diameter of the antenna, can also affect the result. Considered making a half length or quarter length antenna design?

This might help:
http://www.c-maxgroup.com/tech/antenna.php
http://www.air-stream.org.au/wavelength

[winson]

Actually I've been doing a lot of antenna testing lately, myself. I've been doing a lot of calculations on attenuation by water on a signal. But I'm no expert!

But i'm sure you are an very experience person!

Thanks for giving all that information and also the links. Actually what i'm design now is a half length antenna i think, the width of antenna's patch will be equal to lambda/2 and by tune the width it really help to shift the resonant frequency.

I have already success improve the return loss of the antenna to around 30dB, and the key tuning that perform is like increase the thickness & height of the substrate, increase the diameter of the coaxial feeding probe. But the antenna still narrow in bandwidth, still have to think some methods to improve the bandwidth.

[samkarwai]

Hi all, greetings to all,

I'm currently designing a 180 degree compact rat-race hybrid coupler for butler matrix network. I divided my rat-race into 8 sub-structures : 4 t-junctions, 2 bends and 2 straight lines. I've already obtained the results for each part of my sub-structure. But now i'm not sure how to combine the structures so that i can obtain the output for the whole rat-race hybrid coupler that is 180 degree phase difference. I'm aware of exporting the sub-structures to DXF file format but i am not sure which environment or how is the step to combine it and get the combined rat-race hybrid coupler output.

Appreciate very much if you can help. Thanks

Best & Warmest Regards,
SAM

[Admin]

I'm not familiar with microwave studio, but what it sounds like is you want a complete rat-race model so you can further analyze in microwave studio, right?

What you can do is open up another CAD modeling program, design it there, then export in a format that microwave studio accepts (dxf or whatever). Solid Works, AutoDesk 3D Inventor, etc.

Let me know if you need further help on that . . .